UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The prospective memory deficit theory of compulsive checking Cuttler, Carrie 2008

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2008_fall_cuttler_carrie.pdf [ 2.68MB ]
Metadata
JSON: 24-1.0066985.json
JSON-LD: 24-1.0066985-ld.json
RDF/XML (Pretty): 24-1.0066985-rdf.xml
RDF/JSON: 24-1.0066985-rdf.json
Turtle: 24-1.0066985-turtle.txt
N-Triples: 24-1.0066985-rdf-ntriples.txt
Original Record: 24-1.0066985-source.json
Full Text
24-1.0066985-fulltext.txt
Citation
24-1.0066985.ris

Full Text

THE PROSPECTIVE MEMORY DEFICIT THEORY OF COMPULSIVE CHECKING by CARRIE CUTTLER B.A., The University of British Columbia, 2001 M.A., The University of British Columbia, 2003 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Psychology) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) September 2008 © Carrie Cuttler, 2008 ABSTRACT Checking compulsions are the most common manifestation of obsessive-compulsive disorder (OCD), yet the mechanisms which contribute to them are not well understood. According to one prominent theory — the memory deficit theory — individuals’ compulsions to check are fueled by a deficit in memory which makes it difficult for them to remember performing a previous action (e.g., locking a door). The main goal of this dissertation is to examine the link between memory deficits and checking compulsions. This examination in carried out in the context of two domains of memory: retrospective memory and prospective memory. A review of the literature on memory in OCD shows that previous research on the memory deficit theory has focused almost exclusively on the domain of retrospective memory, the ability to remember previously learned information and events. More importantly, the review demonstrates that deficits in this domain of memory are not unique to checkers and therefore do not hold the power to explain the compulsion to check. The review further examines the memory deficit theory in the domain of prospective memory, the ability to remember to carry out actions (e.g., lock a door). It reviews two of the studies presented in the dissertation which demonstrate deficits in sub-clinical checkers’ prospective memory and it provides some supplementary analyses which show that deficits in prospective memory are unique to checkers and therefore may hold the power to explain the compulsion to check. Three empirical studies demonstrating that sub-clinical checking compulsions are associated with subjective and objective deficits in prospective memory comprise the body of the dissertation. Two of the studies show that the link between checking compulsions and objective deficits in prospective memory is direct and independent from elevations in depression, anxiety and distractibility associated with checking compulsions. The results are used as initial support for the theory that checking compulsions may develop in part as a compensatory reaction to deficits in prospective memory. If individuals frequently forget to perform tasks they may develop intrusive doubts about whether they performed important tasks and when the perceived consequences of a failure are serious these doubts may lead to checking. TABLE OF CONTENTS ABSTRACT ii TABLE OF CONTENTS iii LIST OF TABLES vi LIST OF FIGURES vii PREFACE viii ACKNOWLEDGMENTS ix STATEMENT OF CO-AUTHORSHIP xi CHAPTER 1: GENERAL INTRODUCTION 1 Dissertation Overview and Obectives 2 LITERATURE REVIEW: CHECKING-IN ON THE MEMORY DEFICIT THEORY OF CHECKING COMPULSIONS 5 General Method 8 Scope of Reviews 8 Expectations 9 Organization of Result Tables 10 Memory Deficits and Meta-Memory Deficits in Retrospective Memory 13 Results 13 Memory for Actions 13 Memory for Source Information 16 Memory for Visual and/or Spatial Materials 19 Memory for Verbal Materials 23 Summary and Discussion 27 Memory Deficits and Meta-Memory Deficits in Prospective Memory 31 Results 31 Supplementary Analyses 33 Summary and Discussion 36 General Discussion 37 References 41 CHAPTER 2: SUB-CLINICAL CHECKERS’ PROSPECTIVE MEMORY IS IMPAIRED 52 Study Overview 54 Method 55 Participants and Design 55 Instruments 56 Padua Inventory 56 Prospective and Retrospective Memory Questionnaire (PRMQ) 57 Prospective Memory Questionnaire (PMQ) 57 III Personal Belonging Task .58 Phone Call Reminder Task 58 Neurocognitive Filler Tests 58 Procedure 59 Results 60 Data Preparation 60 Objective Measures of Prospective Memory Performance 61 Subjective Measures of Prospective Memory Performance 62 Discussion 64 Conclusion 67 References 68 CHAPTER 3: SUB-CLINICAL COMPULSIVE CHECKERS SHOW IMPAIRED PERFORMANCE ON HABITUAL, EVENT- AND TIME-CUED EPISODIC PROSPECTIVE MEMORY TASKS 72 Study Overview 74 Method 75 Participants and Design 75 Instruments 76 Padua Inventory 76 Prospective and Retrospective Memory Questionnaire (PRMQ) 76 Prospective Memory Questionnaire (PMQ) 77 Cognitive Failures Inventory ((CFQ) 77 Beck Depresssion Inventory (BDI) 78 State Trait Anxiety Inventory (STAI) 78 Personal Belonging Task 78 Lexical Decision Task 79 Procedure 79 Results 79 Data Preparation 79 Subjective Prospective Memory Performance 80 The Influence of Depression, Anxiety and Distractibility 82 Objective Prospective Memory Task Performance 85 The Influence of Depression, Anxiety and Distractibility 85 Discussion 86 Conclusion 89 References 90 CHAPTER 4: SUB-CLINICAL COMPULSIVE CHECKERS SHOW IMPAIRED PERFORMANCE ON HABITUAL, EVENT- AND TIME-CUED EPISODIC PROSPECTIVE MEMORY TASKS 94 Study Overview 98 Method 100 Participants and Design 100 Instruments 101 Habitual Prospective Memory Task 102 Episodic Prospective Memory Tasks 102 Time-Cued Prospective Memory Questionnaire (TCPMQ) 103 Prospective Memory Questionnaire (PMQ) 104 Prospective and Retrospective Memory Questionnaire (PRMQ) 105 iv Cognitive Failures Inventory ((CFQ) 105 Padua Inventory 106 Beck Depresssion Inventory (BDI) 106 State Trait Anxiety Inventory (STAI) 107 Personality Assessment Instruments 107 Task Relevance Ratings 107 Procedure 108 Results 108 Data Preparation 108 Habitual Prospective Memory Task Performance 109 Episodic Prospective Memory Task Performance 109 Self-Reported Prospective Memory Performance 113 The Influence of Depression, Anxiety and Distractibility 115 Habitual Prospective Memory Task Performance 115 Episodic Prospective Memory Task Performance 115 Self-Reported Prospective Memory Performance 117 Discussion 119 Conclusion 122 References 124 CHAPTER 5: GENERAL DISCUSSION 128 Limitations 134 Future Directions 136 Contributions and Implications 137 References 139 Appendix (BREB Certificate of Approval) 143 V LIST OF TABLES Table 1.1: A Summary of the Results of Research on Action Memory 14 Table 1.2: A Summary of the Results of Research on Source Memory 17 Table 1.3: A Summary of the Results of Research on Visual Memory 20 Table 1.4: A Summary of the Results of Research on Verbal Memory 24 Table 1.5: A Summary of the Results of Research on Prospective Memory 32 Table 1.6: Classification Criteria and Mean Padua Inventory Subscale Scores for Checkers (n = 26), Washers (n = 36) and Controls (n = 97) 33 Table 2.1: A List of all Instruments and Tasks Assigned to Participants, Arranged According to When They Were Administered, Showing the Time Required to Complete Each 59 Table 3.1: Results of a Series of Linear and Hierarchical Linear Regression Analyses Using Checking, Depression, State and Trait Anxiety and Distractibilty to Predict Ratings on the PMQ and PRMQ Subscales 83 Table 3.2: Zero-Order Correlations Among Checking, Depression, State Anxiety, Trait Anxiety, Distractibility and Ratings on the PMQ and PRMQ Subscales 84 Table 4.1: Description of the Episodic Prospective Memory Tasks Assigned in Each Session for Each Condition and the Total Number of Low and High Checkers who Completed Each Task..101 Table 4.2: Zero-Order Correlations Among Checking, Depression, State Anxiety, Trait Anxiety, Distractibility and the Various Measures of Prospective Memory 116 Table 4.3: Partial Correlations Between Severity of Checking Compulsions and Ratings on the TCPMQ, PMQ and PRMQ Subscales After Controlling for Depression, State and Trait Anxiety and Distractibility 118 Table 5.1: Zero-Order Correlations Between the Residualized Checking Variable and Each of Depression, State Anxiety, Trait Anxiety, Distractibility and the PMQ and PRMQ Subscales in the Sample of Participants in the Second Study (Chapter 3) 133 Table 5.2: Partial Correlations Between the Residualized Checking Variable and Ratings on the PMQ and PRMQ Subscales After Controlling for Depression, State and Trait Anxiety and Distractibility in the Sample of Participants in the Second Study (Chapter 3) 133 vi LIST OF FIGURES Figure 1.1: Proportion of Comparisons Revealing a Memory Deficit (left side) and Meta-Memory Deficit (right side) in Each Domain of Memory and for Each Type of Subject Group Comparison 28 Figure 1 .2: Failure Rates (+SE) on the Prospective Memory Test for Controls (n = 97), Washers (n = 36) and Checkers (n = 26) 34 Figure 1 .3: Mean Ratings (+SE) on the Four Subscales of the Prospective Memory Questionnaire (PMQ) for Controls (n = 97), Washers (n = 36) and Checkers (n = 26) 35 Figure 1.4: Mean Ratings (+SE) on the Two Subscales of the Prospective and Retrospective Memory Questionnaire (PRMQ) for Controls (n=97), Washers (n=36) and Checkers (n=26) . . ..36 Figure 2.1: The Frequency of Clock-Checking Responses on the Phone Call Reminder Task Across the Three Checking Groups 62 Figure 2.2: Self-Rated Frequency of Prospective Memory Failures and Use of Prospective Memory Aiding Strategies Across the Three Checking Groups 63 Figure 2.3: Self-Rated Frequency of Prospective and Retrospective Memory Failures Across the Three Checking Groups 64 Figure 3.1: Self-Rated Frequency of Prospective Memory Failures and Use of Prospective Memory Aiding Strategies by the Two Checking Groups 81 Figure 3.2: Self-Rated Frequency of Prospective and Retrospective Memory Failures by the Two Checking Groups 81 Figure 4.1: Mean Performance (+SE) on the Habitual Prospective Memory Task in Each Session for Low Checkers (n = 88) and High Checkers (n = 99) 109 Figure 4.2: Failure Rates (+SE) on the Time- and Event-Cued Episodic Prospective Memory Tasks for Low Checkers (n = 88) and High Checkers (n = 98) 111 Figure 4.3: Failure Rates (+SE) on the Personally Relevant and Personally Irrelevant Episodic Prospective Memory Tasks for Low Checkers (n = 88) and High Checkers (n = 98). 111 Figure 4.4: Mean Delay in Providing Reminders (+SE) for the Time- and Event-Cued Episodic Prospective Memory Tasks for Low Checkers (n = 77) and High Checkers (n = 76) 112 Figure 4.5: Mean Number (÷SE) of Clock Checks during the 35 Minute Interval of the Time- Cued Prospective Memory Tasks, Broken Down into 5-Minute Intervals for Low Checkers (n 90) and High Checkers (n = 100) 112 Figure 4.6: Mean Ratings (+SE) on the Three Subscales of the Time-Cued Prospective Memory Questionnaire (TCPMQ) for Low Checkers (n = 90) and High Checkers (n = 103) 113 Figure 4.7: Mean Ratings (+SE) on the Four Subscales of the Prospective Memory Questionnaire (PMQ) for Low Checkers (n = 90) and High Checkers (n = 103) 114 Figure 4.8: Mean Ratings (+SE) on the Two Subscales of the Prospective and Retrospective Memory Questionnaire (PRMQ) for Low Checkers (n = 90) and High Checkers (n 103) 114 vii PREFACE The following is a manuscript-based rather than traditional dissertation. It begins with a brief introduction and overview which are followed by four self-contained manuscripts, and it concludes with a general discussion of the entire body of work. Two of the manuscripts have been published and two are currently under review. As a result of the choice to use a manuscript-based dissertation and to provide a general context for each manuscript several sections of this document will appear similar and perhaps redundant. For instance, there is a great deal of similarity and overlap in the first few paragraphs of the introduction to Chapter 1 and the introductions of the various manuscripts presented throughout the dissertation as well as between the instruments sections of the manuscripts presented in Chapters 2, 3 and 4. viii ACKNOWLEDGMENTS There are many people who guided and supported me during the past four years of my doctoral studies who I would like to acknowledge. First and foremost I would like to thank my supervisor, Peter Graf, who made my graduate experience truly unforgettable. I feel honored and privileged to have had the opportunity to work with one of the most respected and influential memory researchers in the world. Peter challenged me to think, write and express myself more clearly. His seemingly everlasting patience and his generosity with his time and resources allowed me to produce work I can truly be proud of and to travel around the world to share that work with an international audience. His unique sense of humor made the lab a fun place to work and made the challenges of graduate school far easier to bear. Peter taught me more than every teacher and professor I have ever had combined. I could not have wished for a better supervisor. I would also like to thank my supervisory committee, Eric Eich and Ralph Hakstian, for their careful consideration of my work, thought provoking questions, numerous reference letters and advice. I feel fortunate to have had the opportunity to work with Eric and Ralph and look forward to working with each in the future. I would like to thank my lab mates, Jess Gao, Lynn Fontenelle, Daniel Siu and Hiroe Li for helping to make the lab a fun, supportive and intellectually stimulating environment. Lynn went above and beyond her duties as a research assistant, providing me not only with practical support for my research but also with emotional support during times of stress. Jess has helped me in more ways than I can express. She has always been happy to discuss my research ideas, to help me resolve research, technical and personal problems and to lift my spirits during times of challenge. She has become a dear friend who I respect and cherish. I would like to acknowledge, Tonia Relkov, Loretta Siu, Theresa Jubenville, Matthew Denny-Keys and the numerous other undergraduate students who volunteered their time to assist me with my various research projects. I would also like to thank the many undergraduate students who participated in my studies for their time and energy. I am grateful to the Natural Sciences and Engineering Council of Canada and the Michael Smith Foundation for Health Research for believing in my potential. The financial support that they provided allowed me to live comfortably and travel extensively during my years as a graduate student. ix I am grateful to Zehra LeRoy, Tracy Cassels, Amanda LaMarre, Katharine Yoshida, Shahab Ghorashi and Jennifer Corbett, fellow graduate students and friends who made my graduate years both fun and memorable. I would like to thank two of my oldest and dearest friends, Susan Bell and Andrea Halvorsen who not only supported me during my years as a graduate student but for most of my life. I would also like to thank my family for cheering me on and for continually reminding me of the immensity of the goal that I was pursuing. Finally I would like to thank my partner Ryan McLaughlin for supporting me in every way possible. Ryan has always been willing to share with me his expertise in neuroscience and to help me in the writing process. I will never forget the care he gave me when I was writing my comps with a full body rash, the patience and understanding he shows me when I have to take time away from him to focus on my work and the peace and happiness that he brings me always. x STATEMENT OF CO-AUTHORSHIP The manuscripts presented in this dissertation were all co-authored by Dr. Peter Graf, who supervised me in each stage of the research process. I developed the theory that is examined in the dissertation as well as the idea for the review paper. Under Peter’s supervision I designed each of the studies and I independently tested all of the research participants. With some guidance from Peter and a member of my supervisory committee, Dr. Ralph Hakstian, I performed all of the statistical analyses. Peter and I worked closely together to prepare each of the manuscripts and his time and dedication to writing process should be noted. To recognize these collaborate efforts plural pronouns (e.g., we, our, us) are used throughout the manuscripts and for the sake of consistency they are also used throughout the remainder of the dissertation. xi CHAPTER 1 GENERAL INTRODUCTION1 Obsessive-Compulsive Disorder (OCD) is a debilitating anxiety disorder that afflicts approximately 3% of the population (Kolada, Bland & Newman, 1994; Stein, Forde, Anderson & Walker, 1997). The disorder is characterized by obsessions and compulsions. Obsessions are intrusive and persistent thoughts, images or impulses that are unwanted and cause anxiety. Compulsions are repetitive behaviours or mental acts such as washing, counting or checking which are typically initiated in an attempt to reduce the anxiety generated by obsessions (American Psychiatric Association, 2000). OCD can be extremely distressful, can occupy a large proportion of individuals’ time and energy and as such can interfere with daily routines, functioning at work, social activities and relationships with others. The causes of OCD have been difficult to determine in part because it is a heterogeneous disorder with many different subtypes (Fontenelle, Mendlowicz & Versiani, 2005; Hasler et al., 2005; Henderson & Pollard, 1988; Mataix-Cols, Rosario-Campos & Leckman, 2005; McKay et al., 2004). The most common subtype, characterized by checking compulsions, occurs in over 50% of OCD patients (Henderson & Pollard, 1988; Ramussen & Eisen 1988; Stein et al., 1997), with an additional 15% of the general population demonstrating sub-clinical checking compulsions (Stein et al., 1997). Individuals with checking compulsions have intrusive doubts that they failed to perform a task and/or failed to perform it properly. They worry that if they made an error, some harm will befall themselves or someone else, and then they feel compelled to check to ensure that the task was indeed performed and/or that it was performed correctly (Muller & Roberts, 2005; Rachman & Shafran, 1998; Woods, Vevea, Chambless, & Bayan, 2002). To illustrate, an individual with checking compulsions may have a recurring intrusive doubt that s/he forgot to unplug the iron, worry about the consequences of such a failure (e.g., starting a fire) and then feel compelled to return home to check the status of the iron at the expense of attending to other activities. The mechanisms that contribute to compulsive checking are not well understood. However, according to one prominent theory — commonly referred to as the memory deficit theory — a deficit in memory contributes to the compulsion to check. The theory postulates that individuals’ compulsions to check are fueled by a deficit in memory which makes it difficult for them to remember performing a previous action (e.g., turning off an iron). A modified version of this theory — which we will refer to as the meta-memory deficit theory — postulates that a Parts of this chapter have been submitted for publication. Cuttler, C. & Graf, P. (in submission). Checking-in on the memory deficit and meta-memory deficit theories of checking compulsions. I subjective or perceived deficit in memory contributes to the compulsion to check, that it is individuals’ lack of confidence in their memory that contributes to the compulsion to check. In developing and testing the memory deficit and meta-memory deficit theories previous investigators have focused exclusively on the domain of retrospective memory, on the ability to remember previously learned information and events. This dissertation focuses on re contextualizing and examining the theories in the domain of prospective memory. According to our prospective memory deficit theory a deficit in prospective memory contributes to the development and maintenance of checking compulsions. Prospective memory is defined as the ability to formulate intentions and plans, to retain, to recollect and to execute them at the appropriate time or in the appropriate context (Einstein & McDaniel, 1996; Graf, 2005; Meacham & Dumitru, 1976). In other words, it is the ability to carry out actions in the future. Some everyday examples are remembering to take medication on schedule, to attend appointments, to lock the door upon leaving home and to turn off the iron and stove after use. It is somewhat surprising that previous investigators of the memory deficit and meta memory deficit theories have neglected to consider prospective memory because everyday-life and clinical observations show that checkers’ compulsions are related to this future-oriented aspect of memory. Checkers tend to become concerned with, and have intrusive doubts about, tasks whose performance depends on prospective memory. For instance, they tend to doubt and check whether or not the iron, stove, lights, water or gas are turned off, and whether doors or windows are properly shut and locked. Following this observation we developed the prospective memory deficit theory. We reasoned that if checkers have a deficit in prospective memory and they frequently experience prospective memory failures, they may begin to worry about and develop intrusive doubts concerning these failures. When the perceived consequences of a prospective memory failure are serious (e.g., the failure to unplug the iron after use might cause a fire), these intrusive doubts may lead to checking behaviour. Dissertation Overview and Objectives The overall goal of this dissertation is to examine the link between deficits in memory and checking compulsions. Towards this goal, the dissertation includes a comprehensive literature review, a description of three empirical studies and a general discussion. Following this overview, we present a literature review (Cuttler & Graf, in submission) which focuses on examining the evidence for the memory deficit and meta-memory deficit theories of checking compulsions. The review is divided into two sections, each of which 2 concerns a different domain of memory. The first section contains a review of the large body of literature that has focused on retrospective memory deficits in compulsive checkers and more generally in patients with OCD. This review is undertaken in an attempt to determine whether retrospective memory deficits are unique to checkers as opposed to being common to all OCD patients. The second section of the review summarizes the results of studies that have examined prospective memory in checkers and OCD patients more generally2.Since only a small number of studies have focused on this domain, this section also contains supplementary analyses in which we combine the data from two of the studies presented in this dissertation in an attempt to examine whether prospective memory deficits are unique to checkers. To summarize, the review paper has two primary objectives: 1. To examine whether there is evidence that retrospective memory deficits andlor meta-memory deficits are unique to individuals with checking compulsions. 2. To examine the evidence for the memory deficit and meta-memory deficit theories of checking compulsions in the domain of prospective memory. Chapter 2 (Cuttler & Graf, 2007) represents our first attempt to reframe the memory deficit and meta-memory deficit theories in the domain of prospective memory. It presents the first study to examine whether there is a link between deficits in prospective memory and checking compulsions. As such, the study described in Chapter 2 was intended to be exploratory and was guided by two objectives: 1. To examine whether checking compulsions are related to objective deficits in prospective memory. 2. To examine whether checking compulsions are related to subjective (i.e., self- reported) deficits in prospective memory. The study presented in Chapter 3 (Cuttler & Graf, 2008) was designed to replicate the results of the first study (Chapter 2) and to rule out competing explanations for the link between deficits in prospective memory and checking compulsions. Previous research has shown that checking compulsions are associated with elevated levels of depression and anxiety (Frost, Sher, & Geen, 1986; Hasler, et al., 2005), and in turn, both of these attributes are associated with impaired performance on prospective memory tasks (Harris, & Menzies, 1999; Kliegel & Jagar, 2006; Rude, Hertel, Jarrold, Covich, & Hedlund, 1999). Moreover, we hypothesized that 2 The literature review was prepared after the manuscripts presented in Chapters 2 and 3 were published and it includes the results of those studies. 3 distractibility is related to both checking compulsions and impaired prospective memory. Thus, the study presented in Chapter 3 had two main objectives: 1. To examine whether there is a direct relationship between deficits in prospective memory and checking compulsions or whether the relationship is secondary to checkers’ comorbid psychopathology. 2. To examine whether the link between deficits in prospective memory and checking compulsions is due to distractibility. Chapter 4 (Cuttler & Graf, in submission) presents a study designed to further examine the link between checking compulsions and deficits in prospective memory. Like retrospective memory, prospective memory is not a unitary construct; it includes clearly delineated components, including episodic and habitual prospective memory. The behavioural (i.e., objective) measures of prospective memory that were used in the first (Chapter 2) and second studies (Chapter 3) were episodic prospective memory tasks. The first study (Chapter 2) contained two different types of episodic prospective memory tasks, an event-cued task and a time-cued task. The results of the study were somewhat unexpected, they revealed a dissociation in checkers’ performance on the two types of tasks. There are a number of potential explanations for that dissociation which we examine in the study described in Chapter 4. While episodic prospective memory tasks are the most commonly used type of task in research on prospective memory, the tasks that checkers tend to become concerned with (e.g., turning off irons, locking doors) are best categorized as habitual prospective memory tasks. In view of the close theoretical link between checking compulsions and this aspect of prospective memory, the study described in Chapter 4 also focused on assessing whether checkers show deficits on a habitual prospective memory task. To summarize, the study described in Chapter 4 was undertaken with two primary objectives: 1. To examine the link between checking compulsions and performance on a habitual prospective memory task. 2. To better understand a previously reported dissociation between checkers’ performance on event- and time-cued episodic prospective memory tasks. Chapter 5 presents a general discussion of the entire body of work. This chapter offers a summary of the main findings presented throughout the dissertation, highlighting how they contribute to the field of prospective memory and to our understanding of checking compulsions. The chapter contains an attempt to rectify some discrepant findings in the second (Chapter 3) and third studies (Chapter 4), a section focused on describing the limitations of the 4 research as well as a section providing suggestions for future research. The chapter concludes with a discussion of the major contributions and implications of the research. LITERATURE REVIEW CHECKING-IN ON THE MEMORY DEFICIT AND META-MEMORY DEFICIT THEORIES OF CHECKING COMPULSIONS Obsessive-Compulsive Disorder (OCD) is a debilitating anxiety disorder that afflicts approximately 3% of the population (Kolada, Bland & Newman, 1994; Stein, Forde, Anderson & Walker, 1997). The disorder is characterized by obsessions and compulsions. Obsessions are intrusive and persistent thoughts, images or impulses that are unwanted and cause anxiety. Compulsions are repetitive behaviours or mental acts such as washing, counting or checking which are typically initiated to reduce the anxiety generated by a obsessions (American Psychiatric Association, 2000). OCD is a heterogeneous disorder with many different subtypes which have been delimited by means of both observation and empirical investigations (Fontenelle, Mendlowicz & Versiani, 2005; Hasler et al., 2005; Henderson & Pollard, 1988; Mataix-Cols, Rosario-Campos & Leckman, 2005; McKay et al., 2004). The most common subtype, characterized by checking compulsions, occurs in over 50% of OCD patients (Henderson & Pollard, 1988; Ramussen & Eisen 1988; Stein et al., 1997), with an additional 15% of the general population demonstrating sub-clinical checking compulsions (Stein et al., 1997). Individuals with checking compulsions have intrusive doubts about failing to perform a task and/or failing to perform it properly. They worry that if they made an error, some harm will befall themselves or someone else, and then they feel compelled to check to ensure that the task was indeed performed and/or that it was performed correctly (Muller & Roberts, 2005; Rachman & Shafran, 1998; Woods, Vevea, Chambless, & Bayan, 2002). Concretely, an individual with checking compulsions may have a recurring intrusive doubt that s/he forgot to unplug the iron, worry about the consequences of such a failure (e.g., starting a fire) and then feel compelled to return home to check on the status of the iron at the expense of attending to other activities. The causes of OCD are still unclear. Biological accounts point to evidence that OCD tends to run in families, that it is associated with fronto-straiatal abnormalities as well as with increased serotonergic responsiveness (see Tallis 1995a for a review). Behavioural accounts of OCD focus on patients’ dysfunctional beliefs and attitudes (e.g., an exaggerated sense of responsibility), as well as on memory deficits, attentional biases and problems with inhibition (see Muller & Roberts, 2005; Tallis, 1995a for reviews). In line with these accounts, the most common treatments for OCD are pharmacological and behavioral. Pharmacological 5 interventions involve serotonin transport inhibitors (Greist, Jefferson, Kobak, Katzelnick, & Serlin, 1995; Pigott & Seay, 1998), and the most common behavioral intervention, cognitive behavior therapy, tends to focus on exposure and response prevention (see Foa, Frankliin & Kozak, 1998; Rowa, Antony & Swinson, 2007 for reviews). In this article we focus on a behavioral account of checking compulsions which has important implications for treatment. Although foreshadowed in the writings of Janet (1903) and Freud (1909), Sher, Frost and Otto (1983) were the first to articulate and empirically examine its basic assumptions. Sher et al. suggested, first, that “compulsive checkers may simply be poorer at memory for prior actions, necessitating repeated checks” (pg. 358), second, that the compulsion to check may be driven by the failure to distinguish between memory for actual versus imagined events (e.g., having unplugged the iron versus thinking about unplugging it), and third, that the compulsion to check may arise from a lack of confidence in this ability. Since Sher et al.’s seminal contribution investigators have made a distinction between a memory deficit theory — the claim that checkers have a memory deficit which contributes to the compulsion to check, and a meta-memory deficit theory — the claim that it is checkers’ diminished confidence in their memory abilities which contributes to the compulsion to check. What evidence is available to support these theoretical claims? Our primary objective was to find out whether there is compelling evidence that memory deficits and/or meta-memory deficits are unique to individuals with checking compulsions. The focus on deficits unique to checkers was brought to the fore by Tallis, Pratt and Jamani (1999) who observed that “unlike checking, there is no obvious reason why... memory problems should be of aetiological significance with respect to other obsessional symptoms” (p. 165). Radomsky, Rachman and Hammond (2001) made the same point by noting that “Patients rarely say that they are washing their hands again and again because they don’t recall washing them earlier, but they do report an inability to remember if a check, say of the stove, was completed correctly” (p. 815). Consistent with such observations and the theoretical claims by Sher et al (1983), it follows that if memory and/or meta-memory deficits contribute to the compulsion to check then the evidence should show that these deficits are unique to individuals with checking compulsions. If all individuals with OCD were to exhibit similar deficits in memory or meta-memory, this finding would suggest that the deficits are a consequence of some factor associated with OCD (e.g., anxiety, depression, medication, brain abnormalities) rather than linked with the compulsion to check. At the very least, if memory or meta-memory deficits contribute to the compulsion to check then the evidence should show that such deficits are significantly more severe in checkers than in other individuals with OCD. 6 Our primary objective is relevant to a number of previous reviews. Among them are reviews on the neuropsychological status of individuals with OCD (e.g., Cox, 1997; Muller & Roberts, 2005; Otto, 1992; Savage, 1998; Tallis, 1995a, 1995b, 1997), and especially prominent, a relatively recent meta-analytic review by Woods et al. (2002) which focused specifically on the memory deficit and meta-memory deficit theories of checking compulsions. In combination, these reviews provide fairly compelling evidence that memory and other cognitive functions are impaired in OCD patients and in checkers. By contrast to our objective, however, the previous reviews were not specifically designed to investigate whether memory deficits are unique to checkers. That is, the previous reviews asked about the presence versus absence of memory deficits in OCD patients and checkers, but they did not compare and contrast, for example, the memory performance of OCD checkers with that of OCD non-checkers. We undertook the direct comparison of these subgroups of OCD patients in order examine the critical claim that a deficit in memory and/or meta-memory contributes to the compulsion to check. Our second objective was to examine the memory deficit and meta-memory deficit theories in the domain of prospective memory, which is the ability to formulate intentions, plans and promises, and to retain, recollect and carry them out at the appropriate time or in the appropriate context (Einstein & McDaniel, 1996; Graf, 2005; Meacham & Dumitru, 1976). Nearly every prior investigation of the memory deficit theory and meta-memory deficit theory has focused on retrospective memory, which is broadly defined as the ability to remember previously learned information, facts or events, while only a few studies have examined these theories in the context of prospective memory. It is somewhat surprising that previous investigators have not focused more on the ability to remember to perform intended actions because everyday-life and clinical observations show that checkers’ compulsions are related to this future-oriented aspect of memory. Checkers tend to be concerned with whether or not the iron or stove is turned off and whether doors or windows are properly shut and locked, that is with tasks whose proper execution depends on prospective memory. An additional reason for focusing on prospective memory is that deficits in this domain may provide an intuitive account of the origins of the intrusive doubts that instigate checking. If checkers have a deficit in prospective memory and they frequently experience prospective memory failures, they may begin to worry about and develop intrusive doubts concerning these failures. When the perceived consequences of a prospective memory failure are serious (e.g., the failure to unplug the iron after use might cause a fire), these intrusive doubts may lead to checking behaviour. 7 In the first section of this paper we examine the memory deficit and meta-memory deficit theories in the domain of retrospective memory. We review the large body of literature that has focused on retrospective memory deficits in checkers and COD patients more generally in order to determine whether retrospective memory deficits are unique to checkers as opposed to being common to all COD patients. In the second section, we examine the memory deficit and meta memory deficit theories in the domain of prospective memory. We review the few studies that have examined prospective memory in checkers and COD patients more generally and also provide some supplementary analyses in an attempt to gain insights into whether prospective memory deficits are unique to checkers. General Method Scope of Reviews To locate publications of relevance to our goals, we used each of the basic terms “memory”, “confidence” and “meta-memory” in combination with each of the following — “obsessive-compulsive disorder”, “obsessive compulsive disorder”, “COD”, “check” and “checking” — to search PsyclNFC and Google Scholar. Next, we did a follow up search of the reference section of each document which was produced by the initial search. From the results of these searches, we excluded all documents which did not report original empirical research. For the remaining documents, we reviewed the abstract and if necessary the method section in order to identify studies that focused on explicit episodic memory (Graf & Schacter, 1985), episodic prospective memory or habitual prospective memory (Graf & UttI, 2001; UttI, 2008). We excluded investigations on other aspects of memory (e.g., semantic memory, implicit memory, working memory, directed forgetting, memory biases) because although they are important for understanding COD, these aspects of memory are not directly implicated by either the memory deficit theory or the meta-memory theory of compulsive checking. Our search and selection process yielded a total of 67 empirical articles. Sixty-five of these articles focused on retrospective memory and three focused on prospective memory. Twenty-one of these articles included comparisons between various kinds of checkers (COD and sub-clinical) and healthy controls, 50 included comparisons between COD patients and healthy controls, 13 included comparisons between COD patients and psychiatric controls and 15 articles compared COD patients with checking compulsions and COD patients without checking compulsions. In total, the articles present data from over 1580 healthy control participants, 258 psychiatric control participants, over 236 sub-clinical checkers and 1519 COD patients. Cf the latter group, 385 were described as COD checkers, 219 were described as COD non-checkers and the remaining 915 were not described in detail. 8 Expectations Evidence relevant to the memory deficit and meta-memory deficit theories comes from investigations with a variety of populations, especially the following: The general population of OCD patients (i.e. individuals with an Axis I diagnoses of OCD who are not screened for specific symptoms), COD checkers (i.e. individuals with an Axis I diagnosis of COD who frequently experience the compulsion to check), and sub-clinical checkers (i.e. individuals — typically community living adults including undergraduate students — who do not have an Axis 1 diagnosis of COD although they frequently experience the compulsion to check). In previous investigations, these populations have been compared to OCD non-checkers (i.e., individuals with an Axis I diagnosis of COD who do not experience checking compulsions), to psychiatric controls, to healthy controls and to non-checkers. The latter two groups differ only by virtue of the fact that the non-checkers have been specifically screened for the absence of checking com Pu lsions. The availability and accessibility of these different groups of participants is likely to be one, probably the most important, reason for their differential use in previous investigations. However, the theoretical claim that memory deficits and meta-memory deficits contribute to the compulsion to check is addressed most directly and relevantly by investigations which have compared COD checkers to CCD non-checkers. These two patient groups are likely to be characterized by similar comorbidities (e.g., anxiety, depression), and to suffer from the same potentially negative consequences of medication and brain dysfunction. Oonsistent with the memory deficit theory and the meta-memory deficit theory, we expected the extant research either to reveal memory deficits and meta-memory deficits only in the CCD checkers, or to reveal significantly larger deficits in the CCD checkers than in the COD non-checkers. Cnly a relatively small number of investigations have directly compared CCD checkers and CCD non-checkers. However, additional pertinent evidence is available from a number of investigations that have compared the memory test performance and the meta-memory scores of sub-clinical checkers and non-checkers. Sub-clinical checkers are typically free of the medications that may confound the memory performance and meta-memory scores of COD checkers; moreover, they have fewer and less severe forms of the comorbidities which are known to afflict CCD patients (Gibbs, 1996). In addition, sub-clinical checking compulsions are much more common (prevalence in the general population is estimated to be about 15%) than COD checking compulsions (Sher et al., 1983; Stein et al., 1997), and thus, studies with sub clinical checkers have tended to involve larger subject groups. Finally, and most importantly, it is generally agreed that sub-clinical and clinical checking compulsions are not fundamentally different, that checking compulsions occur on a continuum which maps the frequency of 9 experiencing the compulsion to check, the degree to which the compulsions interfere with daily functioning and the level of distress that is caused by them (Gibbs, 1996; Rasmussen & Eisen, 2002). Consistent with this characterization of sub-clinical checkers we expected the extant research to show memory deficits and meta-memory deficits in the sub-clinical checkers. Organization of Result Tables To summarize the findings, we have created five tables, four of which pertain to retrospective memory and one (included in the next section of this report) that pertains to prospective memory. Each of the first four tables is dedicated to a content defined area of retrospective memory, specifically to memory for actions (Table 1.1), to memory for source information (Table 1.2), to memory for visual/spatial materials (Table 1.3), and to memory for verbal materials (Table 1.4). We arranged these tables in this manner at least in part because of theoretical considerations, because of the claim by Sher et al. (1983) that compulsive checking may be triggered either by difficulties with remembering prior actions (e.g., turning off the stove), or by the failure to distinguish between memory for actual versus imagined events (e.g., did I turn off the stove or only think about it?). Consistent with the first of these suppositions, we anticipated discovering unique checking-related deficits either only in investigations concerned with action memory or larger deficits in these investigations than in those focusing on memory for visual or verbal materials. Consistent with the second supposition, we anticipated discovering unique checking-related deficits in studies concerned with source memory. As clinical observations illustrate that checkers tend to be concerned with tasks that require prospective memory we also anticipated discovering unique checking-related deficits in this domain of memory (i.e. in the results summarized in Table 1.5). A comparison of Tables 1.1 through 1.4 shows that action and source memory have been investigated far less frequently than either verbal memory or visual memory, even though only the former two have been specifically linked to the memory and meta-memory deficit theories of checking compulsions. However, the preponderance of research related to verbal and visual memory should not be interpreted as an overgeneralization of these theories; rather it is a consequence of the fact that most of the available data have come from investigations on the neuropsychology of OCD rather than from studies specifically focused on the memory and meta-memory deficit theories. Each table is arranged into four main sections. The first section summarizes the results from studies that involved a direct comparison between OCD checkers and OCD non-checkers. As argued earlier in this report, research with these specific populations speaks most directly to the memory deficit theory and the meta-memory deficit theory. Despite their relevance, 10 however, Tables 1 .1 through 1 .5 show that only a relatively small number of studies of this type are currently available. The 2nd section of each table summarizes the results from studies that compared sub clinical checkers with non-checkers, that is, healthy controls who have been screened for checking compulsions. Consistent with the assumption that OCD checkers and sub-clinical checkers are not fundamentally different from each other, that they merely represent different points on the same continuum (Gibbs, 1996; Rasmussen & Eisen, 2002), it may be argued that comparisons between these groups are also directly relevant to the memory deficit theory and to the meta-memory deficit theory. We accept this argument but with reservations because of the evidence that sub-clinical checking is accompanied by many of the same afflictions known to occur in clinical checking (e.g., anxiety, depression) (Cuttler & Graf, 2008; Frost et al., 1986; Gibbs, 1996; Morris, Blashfield, Rankupalli, Bradley, & Goodman, 1996). As a consequence, the comorbid afflictions known to be present in sub-clinical checkers are a potentially serious confound in any comparison between sub-clinical checkers and non-checkers. Studies which have compared OCD checkers with healthy controls, listed in the 3ft1 section of each table, may be viewed as similar to those in the 2rc section, except that potential confounds are even more likely and thus more serious. OCD checkers and sub-clinical checkers differ in potentially important ways, including in their degree of comorbid anxiety and depression, their history (years of suffering) of checking compulsions and their use of medication or therapy for treating compulsive behaviors. All such potential confounds must be identified and their influence on cognition needs to be removed if memory differences from studies with OCD checkers and healthy controls are to be used in support of the memory deficit and meta-memory deficit theories of compulsive checking. In the 4th section of each table we list studies that focused on the neuropsychology of OCD rather than being specifically concerned with the memory deficit or meta-memory deficit theories of compulsive checking. Comparisons between OCD patients and healthy controls are not directly relevant to the theories because the memory deficit theory and the meta-memory deficit theory make claims which are specific to checkers. Nevertheless, studies comparing OCD patients with healthy controls are useful for evaluating the claims that memory and meta memory deficits are unique to checkers. On the assumption that these deficits are unique to checkers, we expected either to find deficits only in the studies summarized in the 3rd but not 4th section of our tables or at least to find larger deficits in the 3 than 4th section. By contrast, similar deficits in the studies listed in the 31(1 and 4th sections would indicate that memory deficits are a common feature of OCD (perhaps caused by anxiety, depression, medication or brain dysfunction) and thus without etiological significance for checking compulsions. 11 In the final section of each table we list studies that have compared OCD patients with psychiatric controls. Once again, rather than being specifically concerned with the memory deficit and meta-memory deficit theories of compulsive checking these studies focused on the neuropsychology of OCD. The authors of these studies recognized the many confounds which obscure direct comparisons of OCD patients and healthy controls and included a psychiatric control group in an attempt to control for these confounds. We provide a summary of these studies for a similar reason, to gauge the potential influence of confounds that afflict comparisons of OCD patients or OCD checkers with healthy controls. Each table has six columns showing: The authors of each study, the type of test or instrument used for the study, the subject target and comparison group, the findings relevant to the memory deficit theory and finally the findings relevant to the meta-memory deficit theory. Many authors or author groups appear more than once in the first column because many investigations have employed more than one assessment instrument and have focused and reported on more than one potential between group difference. The 2’ column of each table identifies the instruments or methods used to assess memory or meta-memory. All standardized, commercially available instruments are identified by their common names, and we have used the label proprietary to mark studies which used an in-house (i.e., non standardized) method. The results of our review are listed in the last two columns of each table. Because of the small number of studies in the most critical top section of each table (i.e., studies comparing OCD checkers with OCD non-checkers), and because many studies do not provide information on variability, it would not have been possible or meaningful to apply the same conventional meta-analytic techniques to each of our data sets. For this reason, we opted for a simple count of relevant significant differences that were reported by the authors. In many cases, the assessment instrument or method which was employed yielded a number of different sub-scores, and the authors reported the presence or absence of significant between group differences in these sub-scores. To accommodate this situation, the tables show the number of significant differences which favored the memory deficit theory or the meta memory deficit theory, over the total number of significance tests. For the purpose of computing this ratio, we classified as a failure each of the few significant findings which went against the outcome expected by the memory deficit theory or the meta-memory deficit theory. Our analysis does not include studies for which the results showed ceiling or floor effects. Specifically, we did not count any statistical test outcomes that involved a mean which was within one standard deviation of the minimum or maximum possible score. Only a small number of ceiling and floor effects were discovered (<1%) and most were associated with 12 recognition tests. We chose not to include these results because in such cases the failure to find group differences may reflect primarily a lack of sensitivity. Memory Deficits and Meta-Memory Deficits in Retrospective Memory Results Memory for Actions. Sher et al. (1983) proposed over two decades ago that compulsive checking may be triggered by difficulties with remembering intended actions (e.g., locking the door, turning off the iron), and yet, only a relatively small number of studies have examined OCD checkers’ and OCD patients’ ability to remember actions (see Table 1.1). In most of these studies, memory for actions was assessed by means of a proprietary test. To illustrate, Sher et al. (1983, 1984, 1989) assessed action memory by requiring participants to recall at the end of the experiment all of the different tasks they completed in the course of the experiment. In other studies where the focus was primarily on source memory, participants were presented with brief task descriptions (e.g., break the match, bend the paperclip) which had to be carried out, written on paper, imagined or simply observed (i.e., observe the action being performed by the experimenter). For the action memory test, participants were required to record all of the task descriptions and/or all of the tasks which they had performed. As shown on the top portion of Table 1.1, we found four statistical comparisons concerning the action memory test scores of COD checkers and COD non-checkers and none of these were significant. In contrast, we found 10 statistical comparisons between COD checkers’ and COD non-checkers’ action related meta-memory scores and four of these revealed greater deficits in the COD checkers. The 2nd portion of Table 1.1 lists 26 comparisons concerning the action memory test performance of sub-clinical checkers and non-checkers, with six showing a significant deficit in the sub-clinical checkers. Cur review revealed only one study which focused on the action related meta-memory of sub-clinical checkers and non-checkers; despite making 20 statistical comparisons, only one revealed a greater meta-memory deficit in the sub-clinical checkers. The 3rd section of Table 1.1 shows 13 statistical comparisons on the action memory test performance of COD checkers and healthy controls. Cnly one of these comparisons revealed a deficit in the COD checkers. The table also shows 11 comparisons of these groups’ action meta-memory, six of which showed a deficit in the COD checkers. The 41 section of Table 1.1 lists 24 comparisons on the action memory test performance of COD patients and healthy controls and only one of these comparisons showed a significant deficit in the COD patients. In addition, there have been 10 comparisons of COD patients’ and healthy controls’ meta-memory, with two of them revealing a deficit in COD patients. 13 T ab le 1. 1. A Su m m ar y o ft he R es ul ts o f R es ea rc h o n A ct io n M em or y . M em or y M eta -M em or y Te st OC D Gr ou p Co m pa ris on Gr ou p . De fic it De fic it 1. OC D Ch ec ke rs & OC D No n- Ch ec ke rs Co ug le et al. (20 07) M em or yf or Ac tio ns & Ev en ts Qu est ion nai re OC D Ch ec ke rs (n= 39) OC D No n- Ch ec ke rs (n= 20) n/a 4/8 He rm an s e ta l. (20 03) M eta -C og nit ion s Q ues tio nna ire OC D Ch ec ke rs (n= 9) OC D No n- Ch ec ke rs (n= 8) n/a 0/1 Je lin ek et al. (20 06) Ri ve rm ea d Be ha vio ur al M em or y— Ro ute OC D Ch ec ke rs (n= 23) OC D No n- Ch ec ke rs (n= 8) 0/2 n/a M cN all y& Ko hlb ec k (19 93) Pr op rie tar y OC D Ch ec ke rs (n1 2) OC D No n- Ch ec ke rs (n= 12) 0/2 0/1 Ov era ll 01 4 41 10 2. Su b- Cl in ica lC he ck er s& No n- Ch ec ke rs Ro th & Ba rib ea u (19 96) Se lf- Or de red Po int ing Su b-C lin ica lC he ck ers (n= 14) No n- Ch ec ke rs (n= 14) 1/2 n/a Ec ke r & En ge lka mp (19 95) Pr op rie tar y Su b-C lin ica lC he ck ers (n2 4) No n- Ch ec ke rs (n= 48) 0/9 n/a Ru be ns tei n et al. (19 93) Ex p la Pr op rie tar y Su b-C lin ica lC he ck ers (n= 20) No n- Ch ec ke rs (n= 20) 2/2 n/a Ru be ns tei n et al. (19 93) Ex p lb Pr op rie tar y Su b-C lin ica lC he ck ers (n= 19) No n- Ch ec ke rs (n2 0) 0/2 n/a Sh er et al. (19 83) Pr op rie tar y Su b-C lin ica lC he ck ers (n= 26) No n- Ch ec ke rs (n= 28) 1/1 n/a Sh er et al. (19 84) Pr op rie tar y Su b-C lin ica lC he ck ers (n= t) No n- Ch ec ke rs (n= t) 0/1 n/a Sh er et al. (19 89) Pr op rie tar y Su b-C lin ica lC he ck ers (n= 13) No n- Ch ec ke rs (n= 12) 1/2 n/a Ze tte rm an et al. (20 06) Pr op rie tar y Su b-C lin ica lC he ck ers (n= 19) No n- Ch ec ke rs (n3 5) 1/7 1/2 0 Ov era ll 6I2 6 11 20 3. OC D Ch ec ke rs & He alt hy Co nt ro ls Co ug le et al. (20 07) M em or yf or Ac tio ns & Ev en ts Qu est ion nai re OC D Ch ec ke rs (n= 39) He alt hy Co ntr ols (n= 69) n/a 4/8 He rm an se ta l. (20 08) Br ief Co gn itiv eC on fid en ce Qu est ion nai re OC D Ch ec ke rs (n1 6) He alt hy Co ntr ols (n1 6) n/a 2/2 Co ns tan se ta l. (19 95) Pr op rie tar y OC D Ch ec ke rs (n= 12) No n- Ch ec ke rs (n 7) 0/2 n/a Ec ke r & En ge lka mp (19 95) Pr op rie tar y OC D Ch ec ke rs (n= 24) No n- Ch ec ke rs (n= 48) 1/9 n/a M cN all y & Ko hlb ec k (19 93) Pr op rie tar y OC D Ch ec ke rs (n= 12) No n- Ch ec ke rs (n1 2) 0/2 0/1 Ov era ll 1I1 3 611 1 4. OC D Pa tie nt s& He alt hy Co nt ro ls Co ug le et al. (20 07) M em or y for Ac tio ns & Ev en ts Qu est ion nai re OC D No n- Ch ec ke rs (n= 20) He alt hy Co ntr ols (n6 9) n/a 1/8 Ga lde ris ie ta l. (19 95) Se lf- Or de red Po int ing OC D Pa tie nts (n2 2) He alt hy Co ntr ols (n= 21) 0/2 n/a He rm an se ta l. (20 03) M eta -C og nit ion s Q ues tio nn air e OC D Pa tie nts (n= 17) No n- Ch ec ke rs (n1 7) n/a 1/1 Je lin ek et al. (20 06) Ri ve rm ea d Be ha vio ur al M em or y— Ro ute OC D Pa tie nts (n3 1) He alt hy Co ntr ols (n= 33) 0/2 n/a M art in et al. (19 95) Se lf- Or de red Po int ing OC D Pa tie nts (n= 18) He alt hy Co ntr ols (n= 18) 0/1 8 n/a M cN all y & Ko hlb ec k (19 93) Pr op rie tar y OC D No n- Ch ec ke rs (n= 12) No n- Ch ec ke rs (n1 2) 1/2 0/1 Ov era ll 1I 24 21 10 Te st OC D Gr ou p Co m pa ris on Gr ou p M em or y M eta -M er no ry De fic it De fic it 5. OC D Pa tie nt s& Ps yc hi atr ic Co nt ro ls Co ug le et al. (20 07) M em or yf or Ac tio ns & Ev en ts Qu est ion nai re CO D No n- Ch ec ke rs (n= 20) Ps yc hia tri cC on tro ls (n= 22) n/a 0/4 He rm an s e ta l. (20 08) Br ief Co gn itiv eC on fid en ce Qu est ion nai re CO D Pa tie nts (n= 16) Ps yc hia tri cC on tro ls (n= 16) n/a 2/2 Ov era ll nla 21 6 td ata ar e un av ail ab le - As shown in the bottom section of Table 1.1 our search of the literature did not reveal any studies that have compared COD patients’ action memory test performance with psychiatric controls. However, we did locate studies that compared these two groups’ action meta-memory. As shown in the table, of the six comparisons of COD patients’ and psychiatric controls’ action meta-memory, two revealed a greater meta-memory deficit in CCD patients. Memory for Source Information. Memory for source information, hereinafter source memory (an ability which is assumed to be closely related to reality monitoring), concerns the ability to remember the source from which information has been obtained (Johnson, Hashstroudi, & Lindsay, 1993). Source memory is important for deciding, for example, whether news of a friend’s engagement was obtained first-hand from the friend or second-hand from an acquaintance of the friend. This type of memory is relevant to checking compulsions, in part, because it is required for distinguishing between imagined (internal) and real (external) events, for example, between thinking about turning off the iron and actually doing it. To our knowledge, there exist no standardized methods or instruments for assessing source memory, and thus, a wide variety of different methods have been used in the studies listed in Table 1.2. In Phase I of a typical experiment on source memory, participants may be presented with a series of simple action statements, some to be performed by the participants, others to be performed by the experimenter or to be imagined or simply to be written down. After a delay, the action statements are re-presented and the participants’ task is to recollect how the action statements were experienced in Phase 1 of the experiment. As shown in the top portion of Table 1.2, we discovered 64 statistical comparisons concerning the source memory test performance of COD checkers and COD non-checkers, none of which revealed a deficit in the COD checkers. In addition, there have been 10 comparisons of COD checkers’ and COD non-checkers’ source meta-memory, none of which revealed a deficit. The next segment of Table 1.2 lists 26 statistical comparisons of the source memory test performance of sub-clinical checkers and non-checkers and five of those showed a significant deficit in the sub-clinical checkers’ scores. Cf the five statistical comparisons of sub-clinical checkers’ and non-checkers’ source meta-memory, four showed a deficit in the sub-clinical checkers. The 3rd section of Table 1.2 lists 64 comparisons of COD checkers’ and healthy controls’ source memory test performance and only one showed a deficit in the COD checkers. There were also 24 comparisons involving meta-memory scores, and six of these showed a significant deficit in the COD checkers. 16 T ab le 1. 2. Su m ar y o ft he R es ul ts o f R es ea rc h o n S ou rc e M em or y . M em or y M eta -M em or y Te st OC D Gr ou p Co m pa ris on Gr ou p De fic it De fic it 1. OC D Ch ec ke rs & OC D No n- Ch ec ke rs H er m an s e ta l. (20 03 ) M eta -C og ni tio ns Qu est ion na ire OC D Ch ec ke rs (n= 9) OC D No n- Ch ec ke rs (n 4) n/ a 0/1 Br ow n et al. (19 94 ) Pr op rie tar y OC D Ch ec ke rs (n= 13) OC D No n- Ch ec ke rs (n= 9) 0/2 n/ a H er m an s e ta l. (20 03 ) Pr op rie tar y OC D Ch ec ke rs (n= 9) OC D No n- Ch ec ke rs (n= 8) 0/9 0/2 M cN all y & Ko hlb ec k (19 93 ) Pr op rie tar y OC D Ch ec ke rs (n 12 ) OC D No n- Ch ec ke rs (n 12 ) 0/7 0/6 M er ck elb ac h & W es se l( 20 00 ) Pr op rie tar y OC D Ch ec ke rs (n 7) OC D No n- Ch ec ke rs (n 12 ) 0/1 0/1 M or itz ,J ac ob se n et al. (20 06 ) Pr op rie tar y OC D Ch ec ke rs (n= 17 ) OC D No n- Ch ec ke rs (n= 10) 0/4 5 n/ a Ov er all 01 64 01 10 2. Su b- Cl in ica lC he ck er s& N on -C he ck er s Ec ke r & En ge lk am p (19 95 ) Pr op rie tar y Su b- Cl ini ca lC he ck er s ( n2 4) No n- Ch ec ke rs (n 48 ) 1/8 3/4 Ru be ns tei n et al. (19 93 )E xp Ia Pr op rie tar y Su b- Cl ini ca lC he ck er s ( n= 20) No n- Ch ec ke rs (n 20 ) ill n/a Ru be ns tei n et al. (19 93 ) Ex p 3 Pr op rie tar y Su b- Cl ini ca lC he ck er s (n= 19) No n- Ch ec ke rs (n= 20 ) 1/2 n/a Sh er et al. (19 83 ) Pr op rie tar y Su b- Cl ini ca lC he ck er s( n= 26) No n- Ch ec ke rs (n= 28) 0/2 1/1 Ze tte rm an et al. (20 06 ) Pr op rie tar y Su b- Cl ini ca lC he ck er s ( n= 19) No n- Ch ec ke rs (n= 35) 2/1 3 n/a Ov er all 51 26 41 5 3. OC D Ch ec ke rs & H ea lth y Co nt ro ls He rm an s et al. (20 08 ) M eta -C og ni tio ns Qu est ion na ire OC D Ch ec ke rs (n 16 ) He alt hy Co ntr ols (n 16 ) n/a 1/1 Br ow n et al. (19 94 ) Pr op rie tar y OC D Ch ec ke rs (n 13 ) He alt hy Co ntr ols (n 18 ) 0/2 n/a Co ns tan s e ta l. (19 95 ) Pr op rie tar y OC D Ch ec ke rs (n= 12 ) No n- Ch ec ke rs (n= 7) 0/2 0/2 Ec ke r & En ge lk am p (19 95 ) Pr op rie tar y OC D Ch ec ke rs (n= 24 ) No n- Ch ec ke rs (n= 48) 1/8 4/4 M cN all y & Ko hlb ec k (19 93 ) Pr op rie tar y OC D Ch ec ke rs (n= 12 ) No n- Ch ec ke rs (n= 12) 0/7 1/6 M or itz ,J ac ob se n et al. (20 06 ) Pr op rie tar y OC D Ch ec ke rs (n 17 ) He alt hy Co ntr ols (n= 51) 0/4 5 0/1 1 Ov er all 11 64 61 24 4. OC D Pa tie nt s & He alt hy Co nt ro ls H er m an s e ta l. (20 03 ) M eta -C og ni tio ns Qu est ion na ire OC D Pa tie nt s (n= 17 ) No n- Ch ec ke rs (n= 17 ) n/ a 1/1 Br ow n et al. (19 94 ) Pr op rie tar y OC D Pa tie nt s (n 26 ) He alt hy Co ntr ols (n= 1 8 ) 0/2 n/a H er m an s e ta l. (20 03 ) Pr op rie tar y OC D Pa tie nt s( n= 8) No n- Ch ec ke rs (n= 17 ) 0/9 0/3 M cN all y & Ko hlb ec k (19 93 ) Pr op rie tar y OC D No n- Ch ec ke rs (n= 12 ) No n- Ch ec ke rs (n= 12 ) 0/7 3/6 M er ck elb ac h & W es se l( 20 00 ) Pr op rie tar y OC D Pa tie nt s( n= 19 ) He alt hy Co ntr ols (n= 16 ) 0/1 1/1 M or itz , J ac ob se n et al. (20 06 ) Pr op rie tar y OC D Pa tie nt s (n 27 ) He alt hy Co ntr ols (n 51 ) 3/4 5 0/1 1 — 1 Te st OC D Gr ou p Co m pa ris on Gr ou p M em or y M eta -M em or y De fic it De fic it 4. OC D Pa tie nt s& He alt hy Co nt ro ls co nt . Ov era ll 31 64 51 22 5. OC D Pa tie nt s& Ps yc hi atr ic Co nt ro ls He rm an se ta l. (20 08) M eta -C og nit ion s Q ues tio nn air e OC D Pa tie nts (n= 16) Ps yc hia tri cC on tro ls (n= 16) n/a 1/1 Ov era ll nla Ill OD The 4th portion of Table 1.2 shows that there have been 64 comparisons of the source memory test performance of OCD patients and healthy controls, three of which were significant. The table also lists 22 comparisons related to the meta-memory scores of the same groups, with five showing a significant deficit in the OCD patients. As shown in the bottom section of Table 1.2 our literature search did not reveal any studies that have compared the source memory test performance of OCD patients and psychiatric controls. However, we did locate one study that compared these two groups’ source meta-memory. The study involved one statistical comparison which showed a greater deficit in the source meta-memory of OCD patients. Memory for Visual andlor Spatial Materials. Table 1.3 highlights the wealth of research on checkers’ and OCD patients’ memory for visual (i.e., non-verbal) and/or spatial materials. This fortunate situation is due to the fact that these aspects of memory are at the core of every neuropsychological test battery, and because almost every study with psychiatric patients involves at least one such battery. Because of the direct link to neuropsychological assessments, most of the data on visual and/or spatial memory have been collected by means of standardized methods, as opposed to the proprietary methods which were used for the studies listed in Tables 1.1 and 1.2. As shown in the top portion of Table 1.3 we located eight comparisons concerning the visual/spatial memory test performance of OCD checkers and OCD non-checkers, two of which revealed a deficit in the OCD checkers. The table also lists 18 comparisons of OCD checkers’ and OCD non-checkers’ visual/spatial meta-memory, one of which was significant. The 2 section of Table 1.3 shows four comparisons of the visual/spatial memory test performance of sub-clinical checkers and non-checkers, one of which was significant. As evident from Table 1.3, our literature search did not uncover any studies which have examined the visual/spatial meta-memory scores of sub-clinical checkers and non-checkers. The 3 section of Table 1.3 lists nine comparisons of the visual/spatial memory test performance of OCD checkers and healthy controls. Eight of these nine comparisons revealed a deficit in the OCD checkers. As shown in this section of the table, our search of the literature did not reveal any studies that have explored the visual/spatial meta-memory scores of QCD checkers and healthy controls. The 4th section of Table 1.3 shows that there have been 116 statistical comparisons of the visual/spatial memory test performance of OCD patients and healthy controls, 53 of which were significant. In addition, there have been 19 statistical comparisons of these two groups’ visual/spatial meta-memory, four of which were significant. 19 T ab le 1. 3. A Su m m ar y o ft he R es ul ts o f R es ea rc h o n V is ua l M em or y . M em or y M eta -M em or y Te st OC D Gr ou p Co m pa ris on Gr ou p . De fic it De fic it 1. OC D Ch ec ke rs & OC D No n- Ch ec ke rs Ch a e ta l. (20 07 ) Re y Co mp lex Fi gu res OC D Ch ec ke rs (n= 24) OC D No n- Ch ec ke rs (nz 23 ) 2/2 n/ a M or itz et al. (20 03 ) Re y Co mp lex Fi gu re s OC D Ch ec ke rs (n= 23 ) OC D No n- Ch ec ke rs (n= 9) 0/2 n/a M or itz et al. (20 03 ) Co rsi Bl oc k Ta pp ing OC D Ch ec ke rs (n= 23 ) OC D No n- Ch ec ke rs (n 9) 011 n/ a Om or i e ta l. (20 07 ) W M S- Vi su al OC D Ch ec ke rs (n= 27 ) OC D No n- Ch ec ke rs (n 26 ) 0/1 n/a Je lin ek et al. (20 06 ) RB M T — Pi ctu re s & Fa ce s OC D Ch ec ke rs (n= 23 ) OC D No n- Ch ec ke rs (n= 8) 0/2 n/ a To lin et al. (20 01 ) Pr op rie tar y OC D Ch ec ke rs (n= 5) CO D No n- Ch ec ke rs (n 9) n/a 1/1 8 Ov era ll 21 8 11 18 2. Su b- Cl in ica lC he ck er s& No n- Ch ec ke rs Ro th & Ba rib ea u (19 96 ) W M S- Vi su al Re pr od uc tio n Su b- Cl ini ca lC he ck er s( n= 14 ) No n- Ch ec ke rs (n= 14 ) 0/2 n/ a Sh er et al. (19 84 ) W M S- Vi su al Re pr od uc tio n Su b- Cl ini ca lC he ck er s ( nt ) No n- Ch ec ke rs (n= t) 0/1 n/a Sh er et al. (19 89 ) W M S- Vi su al Re pr od uc tio n Su b- Cl ini ca lC he ck er s( n= 13 ) No n- Ch ec ke rs (n 12 ) 1/1 n/ a Ov era ll 1I4 nl a 3. OC D Ch ec ke rs & He alt hy Co nt ro ls Ch a et al. (20 07 ) Re y Co mp lex Fi gu re s OC D Ch ec ke rs (n 24 ) He alt hy Co ntr ols (n= 20 ) 2/2 n/a Ta llis et al. (19 99 ) Co rsi Bl oc k Ta pp ing CO D Ch ec ke rs (n= 12) He alt hy Co ntr ols (n= 12 ) 0/1 n/ a Zi tte rl et al. (20 01 ) Co rsi Bl oc k Ta pp ing CO D Ch ec ke rs (n= 27 ) He alt hy Co ntr ols (n= 27) 1/1 n/a Ta llis et al. (19 99 ) Re cu rri ng Fi gu re s CO D Ch ec ke rs (n= 12 ) He alt hy Co ntr ols (n= 12 ) 2/2 n/a Ta lus et al. (19 99 ) Fi gu re Re ca ll OC D Ch ec ke rs (n= 12 ) He alt hy Co ntr ols (n= 12 ) 2/2 n/a Zi tte rl et al. (20 01 ) LG T- 3 OC D Ch ec ke rs (n= 27 ) He alt hy Co ntr ols (n= 27 ) 1/1 n/a Ov era ll 819 nla 4. OC D Pa tie nt s& He alt hy Co nt ro ls An dr és et al. (20 07 ) Re y Co mp lex Fi gu re s CO D Pa tie nt s( n= 35 ) He alt hy Co ntr ols (n= 35) 2/2 n/a Be ha r e ta l. (19 84 ) Re y Co mp lex Fi gu re s CC D Pa tie nt s (n= 16 ) He alt hy Co ntr ols (n= 16 ) 0/1 n/a Bo on e et al. (19 91 ) Re y Co mp lex Fi gu re s OC D Pa tie nt s (n= 20 ) He alt hy Co ntr ols (n= 16 ) 1/1 n/a Ch ae ta l. (20 07 ) Re y Co mp lex Fi gu re s OC D No n- Ch ec ke rs (n 23 ) He alt hy Co ntr ols (n= 20 ) 0/2 n/a Ch an g et al. (20 07 ) Re y Co mp lex Fi gu re s OC D Pa tie nt s (n 16 ) He alt hy Co ntr ols (n= 15 ) 0/2 n/a Co x et al. (19 89 ) Re y Co mp lex Fi gu re s OC D Pa tie nt s( n= 42 ) He alt hy Co ntr ols (n= 35 ) 2/2 n/a D ec ke rsb ac h et al. (20 00 ) Re y Co mp lex Fi gu re s CO D Pa tie nt s (n= 17 ) No rm aU ve Da ta 2/2 n/a M art ino t e ta l. (19 90 ) Re y Co mp lex Fi gu re s CC C Pa tie nt s( n= 14 ) He alt hy Co ntr ols (n= 8) 1/1 n/ a M ata ix- Co ls et al. (20 03 ) Re y Co mp lex Fi gu re s CC D Pa tie nt s (n= 30 ) He alt hy Co ntr ols (n= 30) 3/ 4 n/a 1’. ) 0 M ori tz et al. (20 03) M ori tz et al. (20 05) Pe na dé se ta l. (20 05) Ro h et al. (20 05) Sa va ge et al. (19 99) Sa va ge et al. (20 00) Se ga las (20 08) Sh in et al. (20 04) Si mp so n et al. (20 06) Ch ris ten se n et al. (19 92) Ga lde ris ie ta l. (19 95) M ori tz et al. (20 03) M ori tz et al. (20 05) Pu rce ll et al. (19 98) Zi eli ns ki et al. (19 91) Ar on ow itz et al. (19 94) Co he n et al. (19 96) Si mp so n et al. (20 06) An dr és et al. (20 07) Bo on e et al. (19 91) Ch ris ten se n et al. (19 92) Ch ris ten se n et al. (19 92) Pe na dé s e ta l. (20 05) Ra do ms ky & Ra ch ma n (19 99) Ai gn er et al. (20 07) Ch am be rla in et al. (20 07) Di rso n et al. (19 95) Je lin ek et al. (20 06) Sa va ge et al. (19 96) Zi eli ns ki et al. (19 91) Pu rce ll et al. (19 98) Pu rce ll et al. (19 98) Pu rce ll et al. (19 98) Pu rce ll et al. (19 98) M ori tz, Ku elz et al. (20 06) Re y Co mp lex Fi gu res Re y Co mp lex Fi gu res Re y Co mp lex Fi gu res Re y Co mp lex Fi gu res Re y Co mp lex Fi gu res Re y Co mp lex Fi gu res Re y Co mp lex Fi gu res Re y Co mp lex Fi gu res Re y Co mp lex Fi gu res Co rsi Bl oc kT ap pin g Co rsi Bl oc kT ap pin g Co rsi Bl oc kT ap pin g Co rsi Bl oc kT ap pin g Co rsi Bl oc kT ap pin g Co rsi Bl oc kT ap pin g Be nto n Vi su al Re ten tio n Be nto n Vi su al Re ten tio n Be nto n Vi su al Re ten tio n W M S- Vi su al Re pr od uc tio n W M S- Vi su al Re pr od uc tio n W M S- Vi su al Re pr od uc tio n W M S- Pa ire d- As so cia tes W M S- Fa ce s W M S- Vi su al M em or y Fa cia lR ec og nit ion Pa tte rn Re co gn iti on M em or y Si gn or et’ sM em or y Ef fic ien cy Ba tte ry RB M T — Pi ctu res & Fa ce s De lay ed Re co gn iti on Sp an Re cu rri ng Fi gu res Pr op rie tar y Pr op rie tar y Pr op rie tar y Pr op rie tar y Su bje ctiv eN eu ro co gn iti on In ve nto ry OC D Pa tie nts (n= 32) OC D Pa tie nts (n= 71) OC D Pa tie nts (n3 3) OC D Pa tie nts (n= 21) OC D Pa tie nts (n2 0) OC D Pa tie nts (n= 33) OC D Pa tie nts (n= 50) OC D Pa tie nts (n= 30) OC D Pa tie nts (n= 30) OC D Pa tie nts (n= 18) OC D Pa tie nts (n= 22) OC D Pa tie nts (n= 32) OC D Pa tie nts (n= 71) OC D Pa tie nts (n= 23) OC D Pa tie nts (n= 21) OC D Pa tie nts (n= 31) OC D Pa tie nts (n= 65) OC D Pa tie nts (n= 30) OC D Pa tie nts (n= 35) OC D Pa tie nts (n= 20) OC D Pa tie nts (n 18 ) OC D Pa tie nts (n= 18 ) OC D Pa tie nts (n= 33) OC D Pa tie nts (n= 10) OC D Pa tie nts (n4 0) OC D Pa tie nts (n= 20) OC D Pa tie nts (n= 26) OC D Pa tie nts (n3 1) OC D Pa tie nts (n2 0) OC D Pa tie nts (n2 1) OC D Pa tie nts (n= 23) OC D Pa tie nts (n2 3) OC D Pa tie nts (n2 3) OC D Pa tie nts (n= 23) OC D Pa tie nts (n= 67’ ) He alt hy Co ntr ols (n= 20) He alt hy Co ntr ols (nz 3O ) He alt hy Co ntr ols (n= 33) He alt hy Co ntr ols (n= 20) He alt hy Co ntr ols (n= 20) He alt hy Co ntr ols (n= 30) He alt hy Co ntr ols (n= 50) He alt hy Co ntr ols (n3 0) He alt hy Co ntr ols (n= 35) He alt hy Co ntr ols (n= 18) He alt hy Co ntr ols (n= 21) He alt hy Co ntr ols (n= 20) He alt hy Co ntr ols (n= 30) He alt hy Co ntr ols (n= 23) He alt hy Co ntr ols (n2 1) He alt hy Co ntr ols (n= 22) He alt hy Co ntr ols (n= 32) He alt hy Co ntr ols (n= 35) He alt hy Co ntr ols (n= 35) He alt hy Co ntr ols (n1 6) He alt hy Co ntr ols (n= 18) He alt hy Co ntr ols (n= 18) He alt hy Co ntr ols (n= 33) He alt hy Co ntr ols (nt 2O ) He alt hy Co ntr ols (n= 40) He alt hy Co ntr ols (n= 20) He alt hy Co ntr ols (n= 20) He alt hy Co ntr ols (n= 33) He alt hy Co ntr ols (n= 20) He alt hy Co ntr ols (n= 21) He alt hy Co ntr ols (n= 23) He alt hy Co ntr ols (n= 23) He alt hy Co ntr ols (n= 23) He alt hy Co ntr ols (n= 23) He alt hy Co ntr ols (n= 30 Te st OC D Gr ou p Co m pa ris on Gr ou p M em or y M eta -M em or y De fic it De fic it 4. OC D Pa tie nt s& He alt hy Co nt ro ls 0/2 0/2 1/1 6/6 3/4 2/2 3/3 3/4 0/2 0/1 0/1 1/1 0/1 0/1 1/1 2/2 2/2 2/2 1/2 0/2 1/2 0/1 0/1 0/1 4/6 1/2 3/6 0/2 1/6 2/8 1/1 1/1 1/1 0/1 n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/ a n/a n/a n/a n/a n/a n/a 0/1 r” ) . M em or y M eta -M em or y Te st OC D Gr ou p Co m pa ris on Gr ou p . De fic it De fic it 4. OC D Pa tie nt s& He alt hy Co nt ro ls To lin et al. (20 01) Pr op rie tar y OC D Pa tie nts (nz l4) He alt hy Co ntr ols (n= 14) 0/1 8 4/1 8 Ov era ll 53 11 16 41 19 5. OC D Pa tie nt s& Ps yc hi atr ic Co nt ro ls Ch am be rla in et al. (20 07) Pa tte rn Re co gn iti on M em or y OC D Pa tie nts (n= 20) Ps yc hia tri cC on tro ls (n= 20) 1/2 n/a Ch an g et al. (20 07) Re y Co mp lex Fi gu res OC D Pa tie nts (n= 16) Ps yc hia tri cC on tro ls (n= 15) 0/2 n/a Co he n et al. (19 96) Be nto n Vi su al Re ten tio n OC D Pa tie nts (n= 65) Ps yc hia tri c C on tro ls (n= 17) 0/2 n/a M ori tz et al. (20 05) Co rsi Bl oc kT ap pin g OC D Pa tie nts (n7 1) Ps yc hia tri cC on tro ls (n= 33) 0/1 n/a M ori tz, Ku elz et al. (20 06) Su bje ctiv eN eu ro co gn iti on In ve nto ry OC D Pa tie nts (n= 67) Ps yc hia tri cC on tro ls (n= 30) n/a 0/1 Ra do ms ky & Ra ch ma n (19 99) W M S- Vi su al M em or y OC D Pa tie nts (n= 10) Ps yc hia tri cC on tro ls (n= 10) 0/1 n/a To lin et al. (20 01) Pr op rie tar y OC D Pa tie nts (n= 14) Ps yc hia tri cC on tro ls (n= 14) 1/1 8 2/1 8 Ov era ll 21 26 2I1 9 td ata ar e no ta va ila ble “ 3 “ 3 The bottom section of Table 1.3 shows that of the 26 comparisons of the visual/spatial memory test performance of OCD patients and psychiatric controls, only two revealed a deficit in the OCD patients. Similarly, of the 19 comparisons of the visual/spatial meta-memory scores of these two groups only two were significant. Memory for Verbal Materials. For the same reasons as given in connection with memory for visual/spatial materials, we have an abundance of data on checkers’ and OCD patients’ memory for verbal materials. One curious difference between Tables 1.3 and 1.4 is the greater use of proprietary methods for assessing verbal than visual/spatial memory. However, this difference does not seem critical because the literature shows that most proprietary tests differ from the more familiar standardized tests primarily by virtue of to-be-remembered spoken words or sentences they employed. As shown in the top portion of Table 1.4, we identified 17 statistical comparisons concerning the verbal memory test performance of OCD checkers and OCD non-checkers, none of which showed a significant difference in favor of the memory deficit theory. In addition, we located four comparisons concerning the verbal meta-memory scores of OCD checkers and OCD non-checkers, and none of those was significant. The 2nd section of Table 1.4 lists 19 statistical comparisons of the verbal memory test performance of sub-clinical checkers and non-checkers, and only one of these reached significance. The table shows that there have been four statistical comparisons of these groups’ verbal meta-memory, none of which indicated a significant deficit in the sub-clinical checkers. The 3rd section of Table 1.4 enumerates 15 comparisons of the verbal memory test performance of OCD checkers and healthy controls, nine of which were significant. As shown in the table, our review revealed seven comparisons of the verbal meta-memory of these two groups, five of which indicated a meta-memory deficit in OCD checkers. The 4th section of Table 1.4 identifies 114 comparisons of the verbal memory test performance of the OCD patients and healthy controls and 29 of these were significant. In addition, there have been 15 comparisons focused on OCD patients’ and healthy controls’ meta-memory, five of which showed a significant deficit in the OCD patients. The final section of Table 1.4 lists 21 comparisons of the verbal memory test performance of OCD patients and psychiatric controls, none of which revealed a significant deficit. As shown in the table, our search of the literature only uncovered one comparison focused on the meta-memory scores of these two groups and it was not significant. 23 Ta bl e 1. 4. A Su m m ar y o ft he R es ul ts o fR es ea rc h o n V er ba l M em or y . M em or y M eta -M em or y Te st OC D Gr ou p Co m pa ris on Gr ou p . De fic it De fic it 1. OC D Ch ec ke rs & OC D No n- Ch ec ke rs Ce sc hi et al. (20 03) Ca lif orn ia Ve rba lL ea rn ing OC D Ch ec ke rs (n= 16) OC D No n- Ch ec ke rs (n= 16) 0/4 n/a Ch ae ta l. (20 07) Ho pk ins Ve rba lL ea rn ing OC D Ch ec ke rs (n= 24) OC D No n- Ch ec ke rs (n= 23) 0/2 n/a Je lin ek et al. (20 06) RB M T — Na me & Sto ry OC D Ch ec ke rs (n2 3) OC D No n- Ch ec ke rs (n 8) 0/4 n/a Om ori et al. (20 07) W M S- Ve rba l OC D Ch ec ke rs (n= 27) OC D No n- Ch ec ke rs (nt 26 ) 0/1 n/a Ka ra da g et al. (20 05) Pr op rie tar y OC D Ch ec ke rs (n2 2) OC D No n- Ch ec ke rs (n= 10 ) 0/4 0/4 M ac do na ld et al. (19 97) Pr op rie tar y OC D Ch ec ke rs (n= 10) OC D No n- Ch ec ke rs (n= 10) 0/2 n/a Ov era ll 01 17 014 2. Su b- Cl in ica lC he ck er s& No n- Ch ec ke rs Ro th & Ba rib ea u (19 96) W M S- Lo gic al M em or y Su b-C lin ica lC he ck ers (n— 14) No n- Ch ec ke rs (n= 14) 0/2 n/a Sh er et al. (19 84) W M S- Lo gic al M em or y Su b-C lin ica lC he ck ers (n t) No n- Ch ec ke rs (n= t) 1/1 n/a Sh er et al. (19 89) W M S- Lo gic al M em or y Su b-C lin ica lC he ck ers (n= 13) No n- Ch ec ke rs (n= 12) 0/1 n/a Sh er et al. (19 84) W M S- Pa ire d- As so cia tes Le arn ing Su b-C lin ica lC he ck ers (n= t) No n- Ch ec ke rs (n t) 0/1 n/a Sh er et al. (19 89) W M S- Pa ire d- As so cia tes Le arn ing Su b-C lin ica lC he ck ers (n= 13) No n- Ch ec ke rs (n= 12) 0/1 n/a Ru be ns tei n et al. (19 93) Ex p 2 Pr op rie tar y Su b-C lin ica lC he ck ers (n= 19) No n- Ch ec ke rs (n= 20) 0/3 n/a Ru be ns tei n et al. (19 93) Ex p 3 Pr op rie tar y Su b-C lin ica lC he ck ers (n= 19) No n- Ch ec ke rs (n= 20) 0/2 n/a Sh er et al. (19 83) Pr op rie tar y Su b-C lin ica lC he ck ers (n= 26) No n- Ch ec ke rs (n= 28) 0/2 0/1 Tu na et al. (20 05) Pr op rie tar y Su b-C lin ica lC he ck ers (n- 16) No n- Ch ec ke rs (n= 15) 0/6 0/3 Ov era ll 11 19 014 3. OC D Ch ec ke rs & He alt hy Co nt ro ls Ce sc hi et al. (20 03) Ca lif orn ia Ve rba lL ea rn ing OC D Ch ec ke rs (n1 6) He alt hy Co ntr ols (n= 16) 2/4 n/a Ch ae ta l. (20 07) Ho pk ins Ve rba lL ea rn ing OC D Ch ec ke rs (n2 4) He alt hy Co ntr ols (n= 20) 0/2 n/a Zit ter le ta l. (20 01) LG T- 3 OC D Ch ec ke rs (n= 27) He alt hy Co ntr ols (n= 27) 1/1 n/a M ac do na ld et al. (19 97) Pr op rie tar y OC D Ch ec ke rs (n1 0) No n- Ch ec ke rs (n= 10) 0/2 n/a M ac do na ld et al. (19 97) Pr op rie tar y OC D Ch ec ke rs (n= 10) No n- Ch ec ke rs (n= 20) n/a 2/4 Tu na et al. (20 05) Pr op rie tar y OC D Ch ec ke rs (n= 17) No n- Ch ec ke rs (n= 15) 6/6 3/3 Ov era ll 91 15 5I7 4. OC D Pa tie nt s & He alt hy Co nt ro ls Ce sc hi et al. (20 03) Ca lif orn ia Ve rba lL ea rn ing OC D No n- Ch ec ke rs (n= 16) He alt hy Co ntr ols (n= 16) 2/4 n/a Ch an g et al. (20 07) Ca lif orn ia Ve rba lL ea rn ing OC D Pa tie nts (n= 16) He alt hy Co ntr ols (n= 15) 0/3 n/a N ) 4 M em or y M eta -M em or y Te st OC D Gr ou p Co m pa ris on Gr ou p De fic it De fic it 4. OC D Pa tie nt s & He alt hy Co nt ro ls co nt . De ck ers ba ch et al. (20 00) De ck ers ba ch et al. (20 04) M art in et al. (19 93) Sa va ge et al. (20 00) Zi eli ns ki et al .( 199 1) An dr és et al. (20 07) Be ha re ta l. (19 84) Co xe ta l. (19 89) Ju ra do et al. (20 01) Ju ra do et al. (20 02) Ki tis et al. (20 07) M art ino te ta l. (19 90) Ro th et al. (20 04) Ch ae ta l. (20 07) An dr és et al. (20 07) Bo on e et al .( 199 1) Ch ris ten se n et al. (19 92) Ch ris ten se n et al. (19 92) Ra do ms ky & Ra ch ma n (19 99) Di rso n et al .( 199 5) Je lin ek et al. (20 06) Sa va ge et al. (19 96) Sa wa mu ra et al. (20 05) Se ga làs (20 08) Di rso n et al. (19 95) Ju ra do et al. (20 01) Ka ra da g et al. (20 05) M ac do na ld et al .( 199 7) M ori tz et al. (20 07) M ori tz, Ku elz et al. (20 06) Ca lif orn ia Ve rba lL ea rn ing Ca lif orn ia Ve rba lL ea rn ing Ca lif orn ia Ve rba lL ea rn ing Ca lif orn ia Ve rba lL ea rn ing Ca lif orn ia Ve rba lL ea rn ing Au dit ory Ve rba lL ea rn ing Au dit ory Ve rba lL ea rn ing Au dit ory Ve rba lL ea rn ing Au dit ory Ve rba lL ea rn ing Au dit ory Ve rba lL ea rn ing Au dit ory Ve rba lL ea rn ing Au dit ory Ve rba lL ea rn ing Au dit ory Ve rba lL ea rn ing Ho pk ins Ve rba lL ea rn ing W M S- Lo gic al M em or y W M S- Lo gic al M em or y W M S- Lo gic al M em or y W M S- Pa ire d- As so cia tes Le arn ing W M S- Ve rba lM em or y Si gn or ets M em or y Ef fic ien cy Ba tte ry Ri BM T — Na me an d Sto ry De lay ed Re co gn iti on Sp an Id do ns Ve rba lS tra teg y Sp ain -C om plu ten se Ve rba lL ea rn ing Pr op rie tar y Pr op rie tar y Pr op rie tar y Pr op rie tar y Pr op rie tar y Su bje ctiv eN eu ro co gn iti on In ve nto ry 5. OC D Pa tie nt s& Ps yc hi atr ic Co nt ro ls Ce sc hi et al. (20 03) Ch an g et al. (20 07) De ck ers ba ch et al. (20 04) Ca lif orn ia Ve rba lL ea rn ing Ca lif orn ia Ve rba lL ea rn ing Ca lif orn ia Ve rba lL ea rn ing OC D No n- Ch ec ke rs (n= 16) OC D Pa tie nts (n= 16) OC D Pa tie nts (n= 30) Ps yc hia tri cC on tro ls (n= 16) Ps yc hia tri cC on tro ls (n= 15) Ps yc hia tri cC on tro ls (n= 30) OC D Pa tie nts (n= 17) No rm ati ve Da ta 6/7 n/a OC D Pa tie nts (n3 0) He alt hy Co ntr ols (n= 30) 2/4 n/a OC D Pa tie nts (n= 17) He alt hy Co ntr ols (n= 16) 2/6 n/a OC D Pa tie nts (n3 3) He alt hy Co ntr ols (n= 30) 6/8 n/a OC D Pa tie nts (n2 1) He alt hy Co ntr ols (n= 21) 1/9 n/a OC D Pa tie nts (n= 35) He alt hy Co ntr ols (n= 35) 0/2 n/a OC D Pa tie nts (n= 16) He alt hy Co ntr ols (n= 16) 0/1 n/a OC D Pa tie nts (n= 42) He alt hy Co ntr ols (n= 35) 0/2 n/a QC D Pa tie nts (n2 7) He alt hy Co ntr ols (n= 27) 0/6 n/a OC D Pa tie nts (n= 28) He alt hy Co ntr ols (n= 28) 0/2 n/a OC D Pa tie nts (n2 3) He alt hy Co ntr ols (n2 2) 0/3 n/a OC D Pa tie nts (n= 14) He alt hy Co ntr ols (n 8) 1/1 n/a OC D Pa tie nts (n= 27) He alt hy Co ntr ols (n= 29) 0/6 n/a OC D No n- Ch ec ke rs (n= 23) He alt hy Co ntr ols (n= 20) 0/2 n/a OC D Pa tie nts (n= 35) He alt hy Co ntr ols (n= 35) 1/2 n/a OC D Pa tie nts (n= 20) He alt hy Co ntr ols (n= 16) 0/2 n/a OC D Pa tie nts (n= 18) He alt hy Co ntr ols (n= 18) 0/1 n/a OC D Pa tie nts (n1 8) He alt hy Co ntr ols (n1 8) 0/1 n/a OC D Pa tie nts (n= 10) He alt hy Co ntr ols (n= 20) 0/1 n/a OC D Pa tie nts (n= 26) He alt hy Co ntr ols (n= 20) 0/6 n/a OC D Pa tie nts (n= 31) He alt hy Co ntr ols (n= 33) 0/4 n/a OC D Pa tie nts (n= 20) He alt hy Co ntr ols (n= 20) 0/6 n/a OC D Pa tie nts (n= 16) He alt hy Co ntr ols (n= 16) 4/4 n/a OC D Pa tie nts (n= 50) He alt hy Co ntr ols (n= 50) 4/6 n/a OC D Pa tie nts (n2 6) He alt hy Co ntr ols (n= 30) 0/3 n/a OC D Pa tie nts (n= 27) He alt hy Co ntr ols (n= 27) 0/5 n/a OC D Pa tie nts (n= 32) He alt hy Co ntr ols (n= 31) 0/4 3/4 OC D No n- Ch ec ke rs (n= 10) No n- Ch ec ke rs (n= 10) 0/2 0/4 OC D Pa tie nts (n= 28) He alt hy Co ntr ols (n= 28) 0/1 2/6 OC D Pa tie nts (n6 7) He alt hy Co ntr ols (n= 30) n/a 0/1 Ov era ll 29 11 14 51 15 1’ ) (n 0/4 0/3 0/4 n/a n/a n/a . M em or y M eta -M em or y Te st OC D G ro up Co m pa ris on Gr ou p . . De fic it De fic it 5. OC D Pa tie nt s & Ps yc hi at ric Co nt ro ls M art in et al. (19 93) Ca lif orn ia Ve rba lL ea rn ing OC D Pa tie nts (n= 17 ) Ps yc hia tri c C on tro ls (n= 11 ) 0/6 n/a Ki tis et al. (20 07) Au dit ory Ve rba lL ea rn ing OC D Pa tie nts (n= 23) Ps yc hia tri c C on tro ls (n= 24) 0/3 n/a M ori tz, Ku elz et al. (20 06) Su bje ctiv eN eu ro co gn iti on In ve nto ry OC D Pa tie nts (n= 67) Ps yc hia tri cC on tro ls (n= 30) n/a 0/1 Ra do ms ky & Ra ch ma n (19 99) W M S- Ve rba lM em or y OC D Pa tie nts (n= 10) Ps yc hia tri cC on tro ls (n1 O) 0/1 n/a Ov er all 01 21 011 0) Summary and Discussion The main findings from Tables 1.1 through 1.4 are highlighted in Figure 1.1. To complement the tables where the results from each section are summarized as fractions, Figure 1.1 shows the same data as proportions, with the left panels of the figure showing the findings relevant to the memory deficit theory and the right panels summarizing the findings relevant to the meta-memory deficit theory. The figure highlights similarities and differences between memory deficits and meta-memory deficits. It also shows similarities and differences across the memory domains (e.g., action memory, source memory) as well as across the different subject group comparisons (e.g., OCD checkers vs. OCD non-checkers). A detailed account of similarities and differences in the findings seems premature because of the absence or dearth of data in some cells, and because of the vast differences in the number of attempts that have been made to address each memory domain or subject group difference. Nevertheless, the top two panels of Figure 1.1, concerning memory and meta memory related to action and source memory, show a fair number of statistical comparisons for most cells. For nearly all subject group comparisons (e.g., OCD checkers vs. OCD non- checkers), these panels show larger deficits in meta-memory than in memory, suggesting that the former may be more strongly affected in OCD, and underscoring the importance of distinguishing between the memory deficit theory and the meta-memory deficit theory. Our search of the literature revealed an adequate amount of data for nearly each cell of the panels on the left side of Figure 1.1, thereby permitting cautious performance comparisons across the different memory domains. The findings in Figure 1.1 highlight the similarity in the pattern of deficits related to action and source memory (the top two left panels), as well as the similarity in the pattern of deficits related to visual and verbal memory (the bottom two left panels). In addition, however, the figure also highlights the difference in the pattern of deficits and the consistently larger deficits related to visual and verbal memory compared to action and source memory. The different pattern and larger deficits concerning visual and verbal memory than action and source memory was unexpected, but it may be an artifact related to the reliability and/or sensitivity of the methods and instruments used for assessing performance in each memory domain. As described in Tables 1.1 through 1.4, most previous investigations have employed proprietary instruments for assessing action and source memory, whereas visual and verbal memory have typically been probed by means of standardized tests or by tests closely modeled on standardized tests. It is possible that the standardized tests are more reliable and sensitive than the proprietary tests (note: reliability data are available for most standardized tests but not for the proprietary action and source memory tests), and that these differences might account 27 for the different size and pattern of deficits we found concerning visual and verbal memory than action and source memory. A more provocative alternative possibility is that individuals with OCD may develop various techniques and strategies for supporting memory in the context of their everyday life, and these techniques and strategies might translate more readily to the types of naturalistic tasks that have been used for assessing action and source memory than to the more sterile and controlled demands of the standardized instruments which have been used for assessing visual and verbal memory. Future research is required to distinguish between these possibilities. Figure 1.1. Proportion of Comparisons Revealing a Memory Deficit (left side) and Meta-Memory Deficit (right side) in Each Domain of Memory and for Each Type of Subject Group Comparison. Note: the numbers above the bars indicate the total number of relevant comparisons made. The most important findings in Figure 1.1 concern the differences in the performance and in the meta-memory scores of OCD checkers versus QCD non-checkers, and the second most important findings concern the differences in sub-clinical checkers versus non-checkers. 26 Action Memcy 24 0 i 11 6 10201111 4-, 0 4- 0) U >, 0 E 0) Co C 4-, Co (I) C 0 E 0) 0 (I) C 0 (I) Co 0 E 0 C-) 4- 0 C 0 t 0 0 0 0 64 64 0 4. 0. ci) 00.6 0.40 2 0.2 0) 0 ci) 0.8 (0 0.6 .2 0.4 (0 02(I) C 0 2 ci) 008 (1) 0.6 04 Co 0.2E 0 C-) 4- 10 C 0.8 0 0.6 0 0.42 1 0,2 26 Source Memory 1[J Visual Memory 19 18 19 — 0 0 Verbal Memory 4 [1_L_ OCD Sub-Clinical OCD OCD OCD Checkers! Checkers! Checkers! Patients! Patients! OCD OCD Healthy Healthy Psychiatric Non-Checkers Non-Checkers Controls Controls Controls 19 — OCD Sub-Clinical OCD Checkers! Checkers! Checkers! OCD OCD Healthy Non-Checkers Non-Checkers Controls OCD OCD Patients! Patients! Heafthy Psychiatric Controls Controls 28 For the reasons enumerated and discussed earlier in this report (i.e. because comparisons between these groups implicate the fewest possible confounds), comparisons between these subgroups bear most directly on the specificity of the memory deficit theory and of the meta memory deficit theory. The findings — the two leftmost bars of each panel — either show no deficits in the memory or meta-memory scores, or they show only relatively small deficits in either the memory performance or meta-memory scores. Combined across all memory domains, a total of 93 comparisons of the memory test performance of OCD checkers and DOD non-checkers have been made and only two showed a deficit in DOD checkers. Similarly, a total of 75 statistical tests have been used to find memory deficits in sub-clinical checkers versus non-checkers, and only 13 showed a deficit. With respect to meta-memory there have been a total of 42 comparisons of OCD checkers and OCD non-checkers and only five revealed the predicted deficit. An additional 29 comparisons have addressed deficits in the meta-memory of sub-clinical checkers and non-checkers, five of which revealed the predicted deficits. This balance of findings does not bode well for either the memory deficit theory or the meta-memory deficit theory. A few cells in Figure 1.1 seem at odds with this description of the findings. The most notable exceptions are the high incidence of deficits in meta-memory for source information in the sub-clinical checkers, and the relatively high incidence of deficits in meta-memory for action in the COD checkers relative to COD non-checkers. Although these specific exceptions are provocative, they must be viewed cautiously because they are based on a relatively small number of comparisons (five and 10, respectively), and the bulk of the evidence which favors the theory was contributed by just two independent studies (see Tables 1.1 and 1.2). The higher incidence of memory performance deficits found in comparisons of sub clinical checkers and non-checkers than in comparisons of COD checkers and OCD non- checkers, depicted in the two leftmost columns of Figure 1.1, also must be interpreted cautiously. As discussed earlier in this report, comparisons between sub-clinical checkers and non-checkers are known to involve variables (e.g., distractibility, depression, anxiety) that have the potential to confound the results. We speculate that such confounds are responsible for the more frequent occurrence of deficits in comparisons between sub-clinical checkers and non- checkers than in comparisons between OCD checkers and OCD non-checkers. By stark contrast to the absence or infrequent deficits in memory and meta-memory found in studies which have compared OCD checkers with OCD non-checkers or sub-clinical checkers with non-checkers, Figure 1.1 underscores the far greater incidence of deficits found in studies which have compared COD checkers or COD patients with healthy controls. On the performance side of Figure 1.1, those deficits are especially pronounced in visual and verbal 29 memory, whereas on the meta-memory side, deficits are pronounced in action and verbal memory. A direct comparison of the findings in the 3 (those showing deficits in CCD checkers relative to healthy controls) and (those showing deficits in CCD patients relative to healthy controls) columns of bars seems, at first glance, to indicate that deficits are more common in QCD checkers than CCD patients. A comparison of the findings in the 1St (those showing deficits in COD checkers relative to COD non-checkers) and 3Id (those showing deficits in COD checkers relative to healthy controls) columns of bars in Figure 1.1 shows that deficits are much more commonly found when COD checkers are compared to healthy controls than when they are compared to COD non-checkers. This finding highlights the importance of using a matched control group and suggests that the deficits that have been discovered through comparisons of COD checkers and healthy controls are a function of the many potential confounds inherent in such comparisons (e.g., medication, brain dysfunction, comorbid psychopathology). The same methodological message about the selection of appropriate control groups is underscored by the difference in the findings summarized in the 4th (those showing deficits in COD patients relative to healthy controls) and 5th (those showing deficits in COD patients relative to psychiatric controls) columns of bars in Figure 1. Deficits are more commonly found when COD patients are compared to healthy controls than when they are compared to psychiatric controls. Together these findings suggest that the deficits that have emerged in some previous studies using healthy control groups are largely spurious and that they should not be used as evidence to support the memory deficit or meta-memory deficit theories. The research summarized in Figure 1.1 is in line with the recent review by Woods et al. (2002), showing that memory performance deficits and meta-memory deficits are a common occurrence in OCD checkers. More importantly, however, our review shows that such deficits are infrequent in studies which have compared COD checkers with COD non-checkers. Similarly, we have shown that such deficits are infrequent in studies that have compared sub clinical checkers with non-checkers. Therefore, the combined findings from our review argue against the theoretical claims that memory deficits or meta-memory deficits contribute to the compulsion to check. In our view, the evidence does not support either the memory deficit theory or the meta-memory deficit theory of checking. Is it time to abandon these theories? Despite the foregoing conclusion, we recommend against this kind of rash step, first, because to the best of our knowledge, a total of only 15 empirical investigations have specifically compared COD checkers with CCD non-checkers, and many of these studies involved relatively small samples. A second important reason against rejection of the memory deficit and meta-memory deficit theories comes from the fact these 30 theories have so far been examined in only a limited domain. Checking compulsions are typically initiated because of anxiety generated by intrusive doubts pertaining to the failure to perform an intended action, such as locking a door. While checkers’ ability to reach back and remember performing intended actions (i.e., action memory) has been the focus of previous investigations, checkers’ ability to reach forward and remember to execute intended actions (i.e., prospective memory) as not yet been adequately investigated. Memory Deficits and Meta-Memory Deficits in Prospective Memory Our search of the literature revealed only four studies that have examined prospective memory in checkers and/or OCD patients. One of these studies was excluded from our review (Jelinek, Moritz, Heeren & Naber, 2006) because of the presence of ceiling effects on each of the prospective memory tests. The results of our review of the remaining three studies are arranged as they were in the previous section on retrospective memory; they are summarized in a table (Table 1.5) which highlights results pertaining to memory and meta-memory. As so little research has been conducted in this domain, in an attempt to provide some preliminary insights into the specificity of checkers’ putative prospective memory deficits the results section also presents some supplementary analyses of data from two of our recent studies. Results The studies listed in Table 1.5 used proprietary tests to assess prospective memory. For instance, we (Cuttler & Graf, 2007, 2008) used a modified version of a test from the Rivermead Behavioral Memory Test (RBMT) battery (Wilson, Cockburn, & Baddeley, 1985) which requires participants to request the return of a personal belonging. Meta-memory for prospective memory has been assessed using proprietary tests where participants have to estimate how well they will perform on prospective memory tests and also by means of questionnaires which are designed to measure the frequency of experiencing prospective memory failures in everyday life. As shown on the top portion of Table 1.5, to date, there have been no comparisons of the prospective memory test performance of OCD checkers and OCD non-checkers. However, there has been one comparison of OCD checkers’ and OCD non-checkers’ meta-memory, and it showed no differences in the scores of the two groups. The 2 portion of Table 1.5 lists three comparisons of the prospective memory test performance of sub-clinical checkers and non-checkers and two of those showed a significant deficit in the sub-clinical checkers. The table also shows that there have been 12 comparisons of the meta-memory scores of sub-clinical checkers and non-checkers, 10 of which revealed a deficit in sub-clinical checkers. 31 T ab le I5 . Su m ar y o ft he R es ul ts o f R es ea rc h o n Pr os pe ct iv e M em or y . M em or y M eta -M em or y Te st OC D Gr ou p Co m pa ris on Gr ou p . De fic it De fic it 1. OC D Ch ec ke rs & OC D No n- Ch ec ke rs M ori tz, Ku elz et al. (20 06) Su bje ctiv eN eu ro co gn iti on In ve nto ry OC D Ch ec ke rs (n= 30) OC D No n- Ch ec ke rs (n= 37) n/a 0/1 Ov er all nl a Oi l 2. Su b- Cl in ica lC he ck er s& N on -C he ck er s Cu ttle r & Gr af (20 07) PM Q Su b-C lin ica lC he ck ers (n= 40) No n- Ch ec ke rs (n= 45) n/a 4/4 Cu ttle r& Gr af (20 08) PM Q Su b-C lin ica lC he ck ers (n= 64) No n- Ch ec ke rs (n= 62) n/a 4/4 Cu ttle r & Gr af (20 07) PR M Q Su b-C lin ica lC he ck ers (n4 0) No n- Ch ec ke rs (n= 45) n/a 1/1 Cu ttle r& Gr af (20 08) PR M Q Su b-C lin ica lC he ck ers (n= 64) No n- Ch ec ke rs (n= 62) n/a 1/1 Cu ttle r & Gr af (20 07) Pr op rie tar y Su b-C lin ica lC he ck ers (n= 40) No n- Ch ec ke rs (n= 45) 1/1 0/1 Cu ttle r & Gr af (20 07) Pr op rie tar y Su b-C lin ica lC he ck ers (n= 40) No n- Ch ec ke rs (n= 45) 0/1 0/1 Cu ttle r& Gr af (20 08) Pr op rie tar y Su b-C lin ica lC he ck ers (n= 64) No n- Ch ec ke rs (n= 62) 1/1 n/a Ov er all 21 3 10 I12 3. OC D Ch ec ke rs & H ea lth y Co nt ro ls n/a n/a Ov era ll nia nla 4. OC D Pa tie nt s& He alt hy Co nt ro ls M ori tz, Ku elz et al. (20 06) Su bje ctiv eN eu ro co gn iti on In ve nto ry OC D Pa tie nts (n= 67) He alt hy Co ntr ols (n= 40) n/a 0/1 Ov era ll 0I2 Oi l 5. OC D Pa tie nt s& Ps yc hi atr ic Co nt ro ls M ori tz, Ku elz et al. (20 06) Su bje ctiv eN eu ro co gn iti on In ve nto ry OC D Pa tie nts (n= 67) Ps yc hia tri cC on tro ls (n= 40) n/a 0/1 Ov era ll 012 OI l C A ) M The 3rd section of Table 1.5 shows that to date there have been no comparisons of the prospective memory or meta-memory of OCD checkers and healthy controls. Similarly, the 4th section of Table 1.5 shows that there have been no comparisons of the prospective memory test performance of OCD patients and healthy controls. However, there has been one comparison of the meta-memory of COD patients and healthy controls and it was not significant. The bottom section of Table 1.5 shows that there have also been no comparisons of the prospective memory test performance of COD patients and psychiatric controls. The table shows that there has been one comparison of the meta-memory of these two groups and it was not significant. Supplementary Analyses. The paucity of research on prospective memory provides little fuel for speculation about the specificity of prospective memory and meta-memory deficits to checkers. Thus, we conducted some supplementary analyses on the data from two of our studies (Cuttler & Graf, 2007, 2008) in order to determine whether prospective memory and meta-memory deficits are specific to checkers or are also present in washers. Washing compulsions are the second most prevalent form of compulsive behaviour (Henderson & Poflard, 1988; Rachman, 2002) and these compulsions commonly co-occur with checking compulsions (Rachman, 2002; Rachman & Hodgson, 1980). In our previous studies, we used the Padua Inventory (Sanavio, 1988) to assess obsessive-compulsive behaviours. Fortunately, this inventory contains a checking and a washing (contamination) subscale. By pooling our two samples and using the criteria presented in Table 1.6, we were able to obtain decent sized samples of relatively pure checkers (n = 26), washers (n = 36) and controls (n =97). Table 1.6. Classification Criteria and Mean Padua Inventory Subscale Scores for Checkers (n = 26), Washers (n = 36) and Controls (n 97). Checkers Washers Controls Criteria Mean (SD) Criteria Mean (SD) Criteria Mean (SD) Checking Score > 10 16.92 (3.49) < 10 5.47 (2.12) < 10 4.34 (2.89) Washing Score < 10 6.73 (2.31) > 10 16.53 (5.30) < 10 6.47 (2.46) As detailed in the original articles (Cuttler & Graf, 2007, 2008) participants in each study were assigned a prospective memory test which required them to request the return of a personal belonging when they were told “we are now finished all of the tests.” Participants who requested the return of their belonging upon the occurrence of this spoken cue were scored as 33 successful and participants who failed to give the reminder at the appropriate moment were scored as failing. As depicted in Figure 1.2, checkers performed significantly worse than controls on the prospective memory test, x2 (1) = 6.24, p < .05. However, washers’ performance on this test was similar to controls, x2 (1) = .00, p> .05. Of primary importance, checkers performed significantly worse than washers on this test, x2 (1) = 4.32, p < .05. Figure 1.2. Failure Rates (+SE) on the Prospective Memory Test for Controls (n = 97), Washers (n = 36) and Checkers (n = 26). 100 80 J . 601 201 0 To assess meta-memory, participants in each study completed the Prospective Memory Questionnaire (PMQ; Hannon, Adams, Harrington, Fries-Dias, & Gibson, 1995) The PMQ contains four subscales, three which focus on the frequency individuals experience different types of prospective memory failures (episodic, habitual, internally cued) and one which focuses on individuals’ use of prospective memory aiding strategies (e.g., reminders). Higher scores on the fist three subscales indicate more frequent prospective memory failures and higher scores on the memory aiding strategies subscale indicate more frequent use of memory aiding strategies. The results from the PMQ are shown in Figure 1.3. They show that compared to controls, checkers report significantly more failures on the episodic prospective memory scale, t (121) = -2.66, p < .05, the habitual prospective memory scale, t (121) = -5.17, p < .05, and the internally cued prospective memory scale, t (121) = -3.97, p < .05. Our previously reported finding that checkers give higher ratings on the memory aiding strategies scale (Cuttler & Graf, 2007, 2008) failed to reach significance in this reduced sample of checkers, t (121) = -1.40, p> .05. In contrast, washers did not report more failures than controls on the episodic prospective memory scale, t(131)= -.79, p> .05, the habitual prospective memory scale, t(131)= -.51, p> .05, the internally cued prospective memory scale, t(131)= -1.92, p> .05, or the memory aiding Controls Washers Checkers 34 strategies scale, t (131) = -1.48, p> .05. Moreover, compared to washers, checkers reported significantly more failures on the habitual prospective memory scale, t (60) = -.3.26, p < .05. The differences between checkers’ and washers’ ratings on the episodic prospective memory scale, t (60) = -.1.48, p> .05, the internally cued prospective memory scale, t (60) = -1.66, p> .05, and the memory aiding strategies scale, t (60) = -.09, p> .05, failed to achieve significance. Figure 1.3. Mean Ratings (+SE) on the Four Subscales of the Prospective Memory Questionnaire (PMQ) for Controls (n 97), Washers (n 36) and Checkers (n 26). 5 Controls Washers < 4J 21 LJ Cl) . 1 4- a)(UL.. 0 a LI Cl) As an additional measure of meta-memory participants in each study completed the Prospective and Retrospective Memory Questionnaire (PRMQ; Smith, Della Sala, Logie, & Maylor, 2000). The PRMQ contains two subscales, one which focuses on the frequency of experiencing prospective memory failures and one which focuses on the frequency of experiencing retrospective memory failures. Thus, for both subscales higher scores indicate more frequent memory failures. As illustrated in Figure 1 .4, the results from the prospective memory subscale of the PRMQ are consistent with the results of the PMQ. Checkers reported experiencing significantly more prospective memory failures than controls, t (121) = -3.38, p < .05, while washers reported experiencing these failures at a similar frequency as controls, t (131) = -.86, p> .05. The difference between checkers and washers fell just short of the conventional level of significance, t (60) = -1.96, p = .054. Consistent with the findings of our review of the literature on retrospective memory, checkers and washers both reported experiencing significantly more retrospective memory failures than controls, t (121) = -2.01, p < .05 and t (131) = -2.14, p < .05, respectively, and the ratings of checkers and washers did not differ, t (60) = -.38, p> .05. Episodic Habitual Internally Cued Memory Aiding Prospective Prospective Prospective Strategies Memory Memory Memory 35 Figure 1.4. Mean Ratings (+SE) on the Two Subscales of the Prospective and Retrospective Memory Questionnaire (PRMQ) for Controls (n = 97), Washers (n = 36) and Checkers (n = 26). 4 D Controls Washers •Checkers 3 ‘o) 2 LI 0) 1 4- 0 0 Summary and Discussion In connection with prospective memory, the literature pertaining to the memory deficit and meta-memory deficit theories of compulsive checking is too small to reach any strong conclusions. To date, only two studies have examined the memory deficit theory in the context of prospective memory and both used sub-clinical, rather than OCD, checkers. While all three of the reviewed studies examined meta-memory deficits, only one examined these deficits in OCD patients. Clearly, more research with OCD patients is required. Research comparing the prospective memory test performance and meta-memory scores of OCD checkers and OCD non-checkers is particularly needed. While scant, the existing evidence pertaining to the memory deficit theory is promising in the domain of prospective memory. The results of the two studies (listed in the 2nd section of Table 1.5) that focused on prospective memory test performance indicate that sub-clinical checkers have deficits in this domain. Moreover, the results of the supplementary analyses showing that checkers demonstrate greater deficits in prospective memory test performance than either controls or sub-clinical washers indicate that these deficits are unique to checkers. This initial evidence for the specificity of prospective memory deficits is encouraging and suggests that a deficit in prospective memory may hold the key to a better understanding of the compulsion to check. Indeed, checkers tend to become concerned with and check whether they performed tasks whose successful execution depends on prospective memory (e.g., locking a door). Thus, checking compulsions may develop, in part, as a compensatory reaction to a deficit in prospective memory. If individuals frequently forget to perform prospective memory tasks they may develop a strategy of checking to ensure that important tasks do not remain undone. Rospective Memory Retrospective Memory 36 Future research comparing the prospective memory test performance COD checkers and OCD non-checkers is clearly needed to further examine this possibility. The results pertaining to the meta-memory deficit theory in the domain of prospective memory are mixed. Moritz, Kuelz, et al. (2006) used the prospective memory subscale of the Subjective Neurocognition Inventory to assess meta-memory and failed to find any deficits in OCD patients relative to either healthy or psychiatric controls. Moritz, Kuelz, et al. also failed to find any differences in the scores of OCD checkers and CCD non-checkers. Together the findings of their study indicate that deficits in prospective related meta-memory are not present in, or unique to, checkers or OCD patients. However, the findings from our research and supplementary analyses conflict with this conclusion. We (Cuttler & Graf, 2007. 2008) used the PMQ and PRMQ to assess meta-memory in the domain of prospective memory and found that relative to non-checkers, sub-clinical checkers demonstrate deficits on each of the five subscales assessing meta-memory in this domain. The results of our supplementary analyses revealed no differences in the meta-memory scores of sub-clinical washers and controls. Moreover, sub-clinical checkers were found to exhibit greater meta-memory deficits than sub clinical washers, however only one of these effects reached conventional levels of significance. Thus, the findings from our research indicate that meta-memory deficits in the domain of prospective memory may be unique to checkers. The inconsistent results pertaining to meta-memory may reflect differences in the populations tested. While Moritz, Kuelz, et al. (2006) sampled from a population of OCD patients we sampled from a sub-clinical population. While it is generally agreed that clinical and sub-clinical compulsions do not differ in any fundamental way (Gibbs, 1996; Rasmussen & Eisen, 2002) it is possible that the mechanisms that underlie sub-clinical and clinical checking compulsions differ. It may be that a deficit in prospective memory contributes to the more common and less severe form of sub-clinical checking compulsions but that separate factors contribute to the less common, more severe and more dysfunctional form of clinical checking compulsions. Future research comparing COD patients with checking compulsions, COD patients without checking compulsions, sub-clinical checkers and non-checkers is needed to examine this possibility. General Discussion Various versions of the memory deficit and meta-memory deficit theories of checking compulsions have circulated in the literature for over a century (Janet, 1903; Freud, 1909), but it was not until Sher et al’s (1983) seminal contribution that researchers began empirically examining the theories’ assumptions. While previous reviewers have made the critical first step of determining whether memory and meta-memory deficits are present in individuals with 37 checking compulsions (Woods et al., 2002), the goal of our review was to examine whether there is evidence that memory deficits and/or meta-memory deficits are unique to individuals with checking compulsions. Our review, which to the best of our knowledge was comprehensive, in so far as it contains all studies which have directly compared the memory test performance or meta-memory scores of OCD checkers and COD non-checkers, sub-clinical checkers and non-checkers, and COD checkers and healthy controls, offers two critical insights. The first insight pertains to methodology and the relevance of various control groups. The second insight is theoretical and pertains to domain of memory in which the theories have been considered and investigated. Much of what we know about checkers’ memory deficits and meta-memory deficits comes from studies which have involved comparisons between COD checkers and healthy controls and between sub-clinical checkers and non-checkers. However, our review shows that the results of these comparisons can be misleading and difficult to interpret because they implicate a large number of potential confounds. While it is possible to identify many of these potential confounds and statistically control for them, future research focused on understanding checking compulsions may be served better by direct comparisons between OCD checkers and OCD non-checkers. The most important insight gleaned from this review is that the memory deficit and meta memory deficit theories may be too general. Little has been explicitly stated about the domain of memory that may be impaired in checkers. Rather, previous investigators have assumed that all there is to memory is retrospective memory and as a result they have focused almost exclusively on testing the theories in this domain of memory. However, the results of our review show that deficits in retrospective memory are not specific to checkers and therefore do not hold the power to explain the compulsion to check. It is not too surprising that previous investigators of the memory deficit and meta-memory deficit theories have failed to consider prospective memory as it is a relatively new and often neglected field of research. However, the examples that are commonly used to illustrate checking compulsions suggest that these compulsions are linked to prospective memory tasks. The results of our review highlight a need to refocus attention on this domain of memory, they suggest that deficits in prospective memory are specific to checkers and therefore may hold the key to a better understanding of the compulsion to check. One promise for contextualizing and examining the memory deficit and meta-memory deficit theories in the domain of prospective memory is that deficits in this domain may help to explain the intrusive doubts that typically instigate checking. In their usual formulation, the memory deficit and meta-memory deficit theories are unable to account for checkers’ intrusive 38 doubts about whether tasks were completed. The theories simply suggest that following these intrusive doubts, checkers have difficulty remembering performing the task or they do not trust their memory of performing the task. Extending the memory deficit and meta-memory deficit theories to the domain of prospective memory helps to explain the occurrence of these intrusive doubts. If individuals with checking compulsions have deficits in prospective memory and a history of prospective memory failures they may begin to worry about and to have intrusive doubts concerning prospective memory failures (i.e., “did I unplug the iron”?). When the perceived consequences of a prospective memory failure are serious (e.g., the failure to unplug the iron after use might cause a fire), the fear of another failure may trigger compulsive checking. Moreover, intrusive doubts may be triggered not only by the fear of objectively verifiable poor prospective memory test performance but also by a person’s lack of confidence in her/his prospective memory (i.e., by a meta-memory deficit). Prospective memory is an exciting and largely uncharted domain of memory and a focus on this domain opens a number of potential avenues for future research. Like retrospective memory, prospective memory is not a unitary construct; it includes clearly delineated components including episodic prospective memory, monitoring and habitual prospective memory (Graf & UttI, 2001; UttI, 2008). Episodic prospective memory tasks are those that need to be executed only once, after a delay during which the intended action is not actively held in conscious awareness (Graf& UttI, 2001; Graf, 2005; Kvavilashvili & Ellis, 1996; Meacham & Dumitru, 1976). An example is remembering to pick a friend up at the airport in two days. Monitoring tasks differ by virtue of the fact that the intended action remains in conscious awareness during the brief retention interval (Graf & UttI, 2001; UttI, 2008). An example is remembering to stop the bath water before the tub overflows. Habitual prospective memory tasks differ from the latter two in that they need to be performed on a routine, regular or habitual basis (Graf & UttI, 2001; Hannon, et al., 1995; Kvavilashvili & Ellis, 1996; Meacham & Dumitru, 1976). An example is remembering to lock the door upon leaving home. The existing evidence for the memory deficit theory in the domain of prospective memory has come exclusively from studies focused on episodic prospective memory. Future research is needed to examine checkers’ performance on a wider variety of prospective memory tasks. Research on habitual prospective memory may be particularly enlightening as the tasks that checkers tend to become concerned with are best categorized as habitual prospective memory tasks (e.g., individuals need to lock doors on a routine basis) and the results of our direct comparisons of sub-clinical checkers’ and washers’ meta-memory indicate that checkers have unique meta-memory deficits in this component of prospective memory. 39 Future research should also focus on examining how memory deficits and meta-memory deficits interact with other factors believed to contribute to OCD. For instance, prevailing cognitive-behavioural models have emphasized the role of dysfunctional beliefs in the etiology and maintenance of COD (Clark, 2004; Frost & Steketee, 2002; Obsessive Compulsive Cognitions Working Group, 2005; Salkovskis, 1996). COD patients have been shown to be perfectionists, to be intolerant of uncertainty, to have an inflated sense of responsibility and to overestimate the likelihood of threat (Obsessive Compulsive Cognitions Working Group, 2005). Not all COD patients hold these dysfunctional beliefs (Taylor, 2006) and not everyone who has a deficit in prospective memory develops checking compulsions. It seems unlikely that someone with a prospective memory deficit would develop checking compulsions if they did not worry about their frequent prospective memory failures, feel responsible for them or perceive their consequences to be serious. Rather, it seems likely that only those individuals who have a prospective memory deficit and who hold dysfunctional beliefs will develop checking compulsions. Concretely, if someone has frequent prospective memory failures which they feel are unacceptable due to perfectionistic standards and they believe that failures could result in catastrophic outcomes for which they would be responsible (e.g., “I will be responsible for the fire that will result if I left my iron on”) then checking compulsions seem highly likely to ensue. 40 References Aigner, M., Sachs, G., Bruckmuller, E., Winklbaur, B., Zitterl, W., Kryspin-Exner, I., Gur, R., & Katschnig, H. (2007). Cognition and emotion recognition deficits in obsessive- compulsive disorder. Psychiatry Research, 149, 121-128. American Psychiatric Association (2000). Diagnostic and Statistical Manual of Mental Disorders (4th Edition). Washington, DC: American Psychiatric Association. Andrés, S., Boget, T., Lázaro, L, Penadés, R., Morer, A., Salamero, M., & Castro-Fornieles, J. (2007). Neuropsychological performance in children and adolescents wth obsessive- compulsive disorder and influence of clinical variables. Biological Psychiatry, 61, 946- 951. Aronowitz, B. R., Hollander, E., DeCaria, C., Cohen, L., Saoud, J. B., Stein, D., Liebowitz, M. R., & Rosen, W. G. (1994). Neuropsychology of obsessive compulsive disorder. Neuropsychiatry, Neuropsychology, and Behavioural Neurology, 7(2), 81-86. Behar, D., Rapoport, J. L., Berg, C. J., Denckla, M. B., Mann, L., Cox, C., Fedio, P., Zahn, T., & Wolfman M. G. (1984). Computerized tomography and neuropsychological test measures in adolescents with obsessive-compulsive disorder. American Journal of Psychiatry, 141, 363-369. Boon, K. B., Ananth, J., Philpott, L., Kaur, A., & Djenderedjian, A. (1991). Neuropsychological characteristics of nondepressed adults with obsessive-compulsive disorder. Neuropsychiatry, Neuropsychology, and Behavioural Neurology,4(2), 96-1 09. Brown, H. D., Kosslyn, S. M., Breiter, H. C., Baer, L., & Jenike, M. A. (1994). Can patients with obsessive-compulsive disorder discriminate between percepts and mental images? A signal detection analysis. Journal of Abnormal Psychology, 103(3), 445-454. Ceschi, G., Van der Linden, M., Dunker, D., Perroud, A., & Brédart, S. (2003). Further exploration memory bias in compulsive washers. Behaviour Research and Therapy, 41, 737-748. Cha, K. R., Koo, M., Kim, C., Kim, J. W., Oh, W., Suh, H. S., & Lee, H. 5. (2007). Nonverbal memory dysfunction in obsessive-compulsive disorder patients with checking compulsions. Depression and Anxiety, 0, 1-6. Chamberlain, S. R., Fineberg, N. A., Blackwell, A. D., Clark, L., Robbins, T. W., & Sahakian, B. J. (2007). A neuropsychological comparison of obsessive-compulsive disorder and trichotillomania. Neuropsychoologia, 45, 654-662. 41 Chang, S. W., McCracken, J. T., & Piacentini, J. C. (2007). Neurocognitive correlates of child obsessive compulsive disorder and tourette syndrome. Journal of Clinical and Experimental Neuropsychology, 29 (7), 724-733. Christensen, K. J., Won Kim, S., Dysken, M. W., & Maxwell Hoover, K. (1992). Neuropsychological performance in obsessive-compulsive disorder. Biological Psychiatry, 31, 4-18. Clark, D. A. (2004). Cognitive-behavioral therapy for OCD. New York: Guilford Press. Cohen, L. J., Hollander, E., DeCaria, C. M., Stein, D. J., Simeon, D., Liebowitz, M. R., & Aronowitz, B. R. (1996). Specificity of neuropsychological impairments in obsessive- compulsive disorder. A comparison with social phobic and normal control subjects. Journal of Neuropsychiatry and Clinical Neuroscience, 8, 82-215. Constans, J. I., Foa, E. B., Frankin, M. E., Mathews, A. (1995). Memory for actual and imagined events in oc checkers. Behaviour Research and Therapy, 33(6), 665-671. Cougle, J. R., Salkovskis, P. M., & Wahl, K. (2007). Perception of memory ability and confidence in recollections in obsessive-compulsive checking. Journal of Anxiety Disorders, 21, 118-130. Cox, C. S. (1997). Neuropsychological abnormalities in obsessive-compulsive disorder and their assessments. International Review of Psychiatry, 9, 45-59. Cox, C. S., Fedio, P., & Rapoport J. L. (1989). Neuropsychological testing of obsessive- compulsive adolescents. In J. Rapoport (Ed), Obsessive-Compulsive Disorder in Childhood and Adolescents. Washington, DC: American Psychiatric Press. Crawford, J. R., Smith, G., Maylor, E. A., Della Sala, S., Logie, R. H. (2003). The prospective and retrospective memory questionnaire (PRMQ): Normative data and latent structure in a large non-clinical sample. Memory, 11(3), 261-275. Cuttler, C., & Graf, P. (2007) Sub-clinical compulsive checkers’ prospective memory is impaired. Journal of Anxiety Disorders, 21(3), 338-352. Cuttler, C., & Graf, P. (2008). Sub-clinical checking compulsions are related to impaired prospective memory independently of depression, anxiety and distractibility. Journal of Anxiety Disorders, 22(4), 642-654. Cuttler, C., & Graf, P. (in submission). Checking-in on the memory deficit and meta-memory deficit theories of checking compulsions. Clinical Psychology Review. 42 Deckersbach, T., Otto, M. W., Savage, C. R., Baer, L., & Jenike, M. A. (2000). The relationship between semantic organization and memory in obsessive-compulsive disorder. Psychotherapy and Psychosomatics, 69, 101-107. Deckersbach, T., Savage, C. R., Reilly-Harrington, N., Clark, L., Sachs, G., & Rauch, S. L. (2004). Episodic memory impairment in bipolar disorder and obsessive-compulsive disorder: The role of memory strategies. Bipolar Disorders, 6, 233-244. Dirson, S., Bouvard, M., Cottraux, J., & Martin, R. (1995). Visual memory impairment in patients with obsessive-compulsive disorder: A controlled study. Psychotherapy and Psychosomatics, 63, 22-31. Ecker, W., & Engelkamp, J. (1995). Memory for actions in obsessive-compulsive disorder. Behavioural and Cognitive Psychotherapy, 23, 349-371. Einstein, G. 0., & McDaniel, M. A. (1996). Retrieval processes in prospective memory: Theoretical approaches and some new empirical findings. In M. Brandimonte, G. 0. Einstein, & M. A. McDaniel (Eds.), Prospective memory: Theory and applications. New Jersey: Lawrence Eribaum Associates. Foa, E. B., Franklin, M. E., & Kozak, M. J. (1998). Psychosocial treatments for obsessive- compulsive disorder: Literature review. In R. P. Swinson, M. M. Antony, S. Rachman, & M. A. Richter (Eds.), Obsessive-compulsive disorder: Theory, research and treatment. New York: The Guilford Press. Fontenelle, L. F., Mendlowicz, M. V., & Versiani, M. (2005). Clinical subtypes of obsessive- compulsive disorder based on the presence of checking and washing compulsions. Revista Brasileira de Psiquiatria, 27(3), 201-207. Freud, 5. (1909). Notes upon a case of obsessional neurosis (the ‘rat man’). In A. Richards (Ed.), The Penguin Freud Library: Case Histroies II. Harmondsworth: Penguin. Frost, R. 0., Sher, K. J., & Geen, T. (1986). Psychopathology and personality characteristics of nonclinical compulisive checkers, Behaviour Research and Therapy, 24(2), 133-143. Frost, R. 0., & Steketee, G. (2002). Cognitive approaches to obsessions and compulsions: Theory, assessment and treatment. Oxford: Elsevier Galderisi, S., Mucci, A., Catapano, F., D’Amato, A. C., & Maj, M. (1995). Neuropsychological slowness in obsessive-compulsive patients. Is it confined to tests involving the fronto subcortical systems? British Journal of Psychiatry, 167, 394-398. 43 Gibbs, N. A. (1996). Nonclinical populations in research on obsessive-compulsive disorder: A critical review. Clinical Psychology Review, 16(8), 729-773. Graf, P. (2005). Prospective memory retrieval revisited. In N. Ohta, C. M. MacLead & B. UttI (Eds.), Dynamic cognitive processes (pp. 305-332). Tokyo: Springer. Graf, P., & Schacter, D. L.(1985). Implicit and explicit memory for new associations in normal and amnesic subjects. Journal of Experimental Psychology: Learning, Memory and Cognition 11(3), 501—518. Graf, P., & UttI, B. (2001). Prospective memory: A new focus for research. Consciousness and Cognition, 10, 437-450. Greist, J. H., Jefferson, J. W., Kobak, K. A., Katzelnick, D. J., & Serlin, R. C. (1995). Efficacy and tolerability of serotonin transport inhibitors in obsessive-compulsive disorder: A meta-analysis. Archives of General Psychiatry, 52(1), 53-60. Hannon, R., Adams, P., Harrington, S., Fries-Dias, C, Gibson, M. T. (1995). Effects of brain injury and age on prospective memory self-rating and performance. Rehabilitation Psychology, 40(4), 289-298. Hasler, G., LaSalle-Ricci, V. H., Ronquillo, J. G., Crawley, S. A., Cochran, L. W., Kazuba, D., Greenberg, B. D., & Murphy, D. L. (2005). Obsessive-compulsive disorder symptom dimensions show specific relationships to psychiatric comorbidity. Psychiatry Research, 135(2), 121-132. Henderson, J. G., & Pollard, C. A. (1988). Three types of obsessive compulsive disorder in a community sample. Journal of Clinical Psychology, 1988, 44(5), 747-752. Hermans, D., Engelen, U., Grouwels, L., Joos, E., Lemmens, J., Pieters, G. (2008). Cognitive confidence in obsessive-compulsive disorder: Distrusting perception, attention and memory. Behaviour Research and Therapy, 46, 98-1 13. Hermans, D., Martens, K., De Cort, K., Pieters, G., & Eelen, P. (2003). Source memory and metacognitive beliefs related to cognitive confidence in obsessive-compulsive disorder. Behaviour Research and Therapy, 41, 383-401. Janet, P. (1903). Les Obsessions et Ia Psychaesthenie. Paris: Alcan. Jelinek, L., Moritz, S., Heeren, D., & Naber, D. (2006). Everyday memory functioning in obsessive-compulsive disorder, 12, 746-749. Johnson, M. K., Hashtroundi, S., & Lindsay, 5. (1993). Source monitoring. Psychological Bulletin, 114,2-28. 44 Jurado, M. A., Junqué, C., Vallejo, J., Salgado, P., & Grafman, J. (2001). Impairment of incidental memory for frequency in patients with obsessive compulsive disorder. Psychiatry Research, 104, 213-220. Jurado, M. A., Junqué, C., Vallejo, J., Salgado, P., & Grafman, J. (2002). Obsessive-compulsive disorder (OCD) patients are impaired in remembering temporal order and in judging their own performance. Journal of Clinical and Experimental Neuropsychology, 24(3), 261- 269. Karadag, F., Oguzhanoglu, N., Ozdel, 0., Atesci, F. C., & Amuk, T. (2005). Memory function in patents with obsessive compulsive disorder and the problem of confidence in their memories: A clinical study. Croatian Medical Journal, 46(2), 282-287. Kitis, A., Akdede, B. B. K., Alptekin, K., Akvardar, Y., Arkar, H., Erol, A., & Kaya, N. (2007). Cognitive dysfunctions in patients with obsessive-compulsive disorder compared to patients with schizophrenia patients: Relation to overvalued ideas. Progress in Neuro Psychopharmacology & Biological Psychiatry, 31, 254-261. Kolada, J. L., Bland, R. C., & Newman, S. C. (1994). Epidemiology of psychiatric disorders in Edmonton. Obsessive-compulsive disorder. Acta Psychiatrica Scandinavica Supplementum, 376, 24-35. Kvavilashvili, L., & Ellis, J. (1996). Varieties of intentions: Some distinctions and classifications. In M. Brandimonte, G. 0. Einstein, & M. A. McDaniel (Eds.) Prospective memory: Theory and applications (pp. 23-51). New Jersey: Lawrence Erlbaum Associates. Ledley, D. R., Pai, A., & Franklin, M. E. (2007) Treating comorbid presentations: Obsessive- compulsive disorder, anxiety disorders, and depression. In M. M. Antony, C. Purdon, & L. J. Summerfeldt (Eds.), Psychological treatment of obsessive-compulsive disorder: Fundamentals and beyond. Washington: American Psychological Association. Macdonald, P. A., Antony, M. M., Macleod, C. M., Richter, M. A. (1997). Memory and confidence in memory judgments among individuals with obsessive compulsive disorder and non-clinical controls. Behaviour Research and Therapy, 35(6), 497-505. Martin, A., Pigott, T. A., Lalonde, F. M., Dalton, I., Dubbert, B., & Murphy, D. L. (1993). Lack of evidence for huntington’s disease-like cognitive dysfunction in obsessive-compulsive disorder. Biological Psychiatry, 33, 345-353. Martin, A., Wiggs, C. L., Altemus, M., Rubenstein, C., & Murphy, D. L. (1995). Working memory as assessed by subject-ordered tasks in patients with obsessive-compulsive disorder. Journal of Clinical & Experimental Neuropsychology, 17(5), 786-792. 45 Martinot, J. L., Allilaire, J. F., Mazoyer, B. M., Hantouche, D., Huret, J. D., Legaut-Demare, F., Deslauriers, A. G., Hardy, P., Pappata, D., Baron J. C., & Syrota, A. (1990). Obsessive- compulsive disorder: A clinical neuropsychological and positron emission tomography study. Acta Psychiatrica Scandinavica, 82, 233-242. Mataix-Colis, D., Alonso, P., Hernández, R., Deckersbach, T., Savage, C. R., Menchôn, J. M., & Vallejo, J. (2003). Relation of neurological soft signs to nonverbal memory performance in obsessive-compulsive disorder. Journal of Clinical and Experimental Neuropsychology, 25(6), 842-851. Mataix-Colis, D., Rosario-Campos, M. C., & Leckman, J. F. (2005). A multidimensional model of obsessive-compulsive disorder. American Journal of Psychiatry, 162(2), 228-238. McKay, D., Abramovitz, J. S., Calamari, J. E., Kyrios, M., Sookman, D., Taylor, S., &Wihelm, S. (2004). A critical evaluation of obsessive-compulsive disorder subtypes: symptoms versus mechanisms. Clinical Psychology Review, 24, 283-313. McNally, R. J., & Kohlbeck, P. A. (1993). Source memory in obsessive-compulsive disorder. Behaviour Research and Therapy, 3 1(3), 249-253. Meacham, J. A., & Dumitru, J. (1976). Prospective remembering and external-retrieval cues. JSAS Catalog of Selected Documents in Psychology, 6 no 65, 1-12. Merckelbach, H., & Wessel, I. (2000). Memory for actions and dissociation in obsessive- compulsive disorder. Journal of Nervous and Mental Disease, 188(2), 846-848. Moritz, S., Jacobsen, D., Willenborg, B., Jelinek, L., & Fricke, S. (2006). A check on the memory deficit hypothesis of obsessive-compulsive checking. European Archives of Psychiatry and Clinical Neuroscience, 256, 82-86. Moritz, S., Kioss, M., Jacobsen, D., Kellner, M., Andresen, B., Fricke, S., Kerkhoff, G., Siemen, C., & Hand, I. (2005). Extent, profile and specificity of visuospatial impairment in obsessive-compulsive disorder (OCD). Journal of Clinical and Experimental Neuropsychology, 27, 795-814. Moritz, S., Kloss, M., Jahn, H., Schick, M., & Hand, I. (2003). Impact of comorbid depressive symptoms on nonverbal memory and visuospatial performance in obsessive-compulsive disorder. Cognitive Neuropsychiatry, 8(4), 261-272. Moritz, S., Kuelz, A. K., Jacobsen, D., Kloss, M., & Fricke, S. (2006). Severity of subjective cognitive impairment in patients with obsessive-compulsive disorder and depression. Journal of Anxiety Disorders, 20, 427-443. 46 Moritz, S., WahI, K., Zurowski, B., Jelinek, L., Hand, I., Fricke, S. (2007). Enhanced perceived responsibility decreases metamemory but not memory accuracy in obsessive- compulsive disorder (COD). Behaviour Research and Therapy, 45, 2044-2052. Morris, M. R., Blashfield, R. K., Rankupalli, B., Bradley, M. M., & Goodman, W. K. (1 996/1 997). Sub-clinical obsessive-compulsive disorder in college students. Depression and Anxiety, 4, 233-236. Muller, J., & Roberts, J. E. (2005). Memory and attention in obsessive-compulsive disorder: A review. Journal of Anxiety Disorders, 19, 1-28. Obsessive Compulsive Cognitions Working Group (2005). Psychometric validation of the Obsessive Beliefs Questionnaire and the Interpretations of Intrusions Inventory: Part 2. Behaviour Research and Therapy, 43(11), 1527-1 542. Omori, I. M., Murata, Y., Yamanishi, T., Nakaaki, S., Akechi, T., Mikuni, M., & Furukawa, T. A. (2007). The differential impact of executive attendtion dysfunction on episodic memory in obsessive-compulsive diorder patients with checking symptoms vs. those with washing symptoms. Journal of Psychiatric Research, 41, 776-784. Otto, M. W. (1992). Normal and abnormal information processing: A neuropsychological perspective on obsessive compulsive disorder. Psychiatric Clinics of North America, 15(4), 825-848. Penadés, R., Catalán, R., Andrés, S., Salamero, M., & GastO, C. (2005). Executive function and nonverbal memory in obsessive-compulsive disorder. Psychiatry Research, 133(1), 81-90. Pigott, T. A., & Seay, S. (1998). Biological treatments for obsessive-compulsive disorder. In R. P. Swinson, M. M. Antony, S. Rachman, & M. A. Richter (Eds.), Obsessive-compulsive disorder: Theory, research and treatment. New York: The Guilford Press. Purcell, R., Maruff, P., Kyrios, M., & Pantelis. (1998). Cognitive deficits in obsessive-compulsive disorder on tests of frontal-straital function. Society of Biological Psychiatry, 43, 348-357. Rachman, S. (2002). A cognitive theory of compulsive checking. Behaviour Research and Therapy, 40(6), 625-639. Rachman, S., & Hoclgson, R. J. (1980). Obsessions and compulsions. New Jersey: Prentice Hall. Rachman, S., & Shafran, R. (1998). Cognitive and behavioral features of obsessive-compulsive disorder. In R. P. Swinson, M. M. Antony, S. Rachman, & M. A. Richter (Eds.), 47 Obsessive-compulsive disorder: Theory, research, and treatment (51-78). London: The Guilford Press. Radomsky, A. S., & Rachman, S. (1999). Memory bias in obsessive-compulsive disorder (OCD). Behaviour Research and Therapy, 37, 605-618. Radomsky, A. S., Rachman, S. & Hammond, D. (2001). Memory bias, confidence and responsibility in compulsive checking. Behaviour Research and Therapy, 39, 813-822. Rasmussen, S. A., & Eisen, J. A. (1988). Clinical and epidemiological findings of significance to neuropharmacologic rituals in OCD. Psychopharmacological Bulletin, 24, 466-470 Rasmussen, S. A., & Eisen, J. L. (2002). The course of clinical features of obsesssive compulsive disorder. In K. L. Davis, D. Charney, J. T., Coyle, & C. Nemeroff (Eds.), Neuropsychoparmacology: The fifth generation of progress. Philadelphia: American College of Neuropsychopharmacology. Roh, K. S., Shin, M. S., Kim, M., Ha, T., Shin, Y., Lee, K. J., & Kwon, J. S. (2005). Persistent cognitive dysfunction in patients with obsessive-compulsive disorder: A naturalistic study. Psychiatry & Clinical Neurosciences, 59(5), 539-545. Roth, R. M., & Baribeau, J. (1996). Performance of sub-clinical compulsive checkers on putative tests of frontal and temporal lobe memory functions. Journal of Nervous and Mental Disease, 184(7), 411-416. Roth, R. M., Baribeau, J., Milovan, O’Connor, K., & Todorov, C. (2004). Procedural and declarative memory in obsessive-compulsive disorder. Journal of the International Neuropsychological Society, 10, 647-654. Rowa, K., Antony, M. M., & Swinson, R. P. (2007). Exposure and response prevention. In M. M. Antony, C. Purdon, & L. J. Summerfeldt (Eds.), Psychological treatment of obsessive- compulsive disorder: Fundamentals and beyond. Washington: American Psychological Association. Rubenstein, C. S., Peynircioglu, Z. F., Chambless, D. L., & Pigott, T. A. (1993). Memory in sub clinical obsessive-compulsive checkers. Behaviour Research and Therapy, 31(8), 759- 765. Salkovskis, P. M. (1996). Cognitive-behavioral approaches to the understanding of obsessional problems. In R. M. Rapee (Ed.), Current controversies in the anxiety disorders (pp. 103- 134). New York: Guilford Press. 48 Sanavio, E. (1988). Obsessions and compulsions: The padua inventory. Behaviour Research and Therapy, 26(2), 169-1 77. Savage, C. R. (1998). Neuropsychology of obsessive-compulsive disorder: Research findings and treatment implications. In M. A. Jenike, L. Baer, & W. E. Minichiello (Eds.), Obsessive-Compulsive Disorders: Practical Management (3 ed.). St. Louis: Mosby. Savage, C. R., Baer, L., Keuthen, N. J., Brown, H. D., Rauch, S. L., & Jenike, M. A. (1999). Organizational strategies mediate nonverbal memory impairment in obsessive- compulsive disorder. Society of Biological Psychiatry, 45, 905-91 6. Savage, C. R., Deckersbach, T., Wilhelm, S., Rauch, S. L., Baer, L., Reid, T., & Jenike, M. A. (2000). Strategic processing and episodic memory impairment in obsessive-compulsive disorder. Neuropsychology, 14(1), 141-151. Savage, C. R., Keuthen, N. J., Jenike, M. A., Brown, H. D., Baer, L., Kendrick, A. D., Miguel, E. C., Rauch, S. L., & Albert, M. S. (1996). Recall and recognition memory in obsessive- compulsive disorder. Journal of Neuropsychiatry, 8(1), 99-1 03. Sawamura, K., Nakashima, Y., Inoue, M, Kurita, H. (2005). Short-term verbal memory deficits in patients with obsessive-compulsive disorder. Psychiatry and Clinical Neurosciences, 59, 527-532. Segalàs, C., Alonso, P., Labad, J., Jaurrieta, N., Real, E., Jiménez, S., Menchón, J. M., Vallejo, J. (2008). Verbal and nonverbal memory processing in patients with obsessive- compulsive disorder: Its relationship to clinical variables. Neuropsychology, 22(2), 262- 272. Sher, K., Frost, R. 0., & Otto, R. (1983). Cognitive deficits in compulsive checkers: An exploratory study. Behaviour Research and Therapy, 21(4), 357-363. Sher, K. J., Mann, B., Frost, R. 0. (1984). Cognitive dysfunction in compulsive checkers: Further explorations. Behaviour Research and Therapy, 22(5), 493-502. Sher, K. J., Frost, R. 0., Kusner, M., Crews, T. M., & Alexander, J. E. (1989). Memory deficits in compulsive checkers: Replication and extension in a clinical sample. Behaviour Research and Therapy, 27(1), 65-69. Shin, M. S., Park, S. J., Kim, M. S., Lee, Y. H., Ha, T. H., Kwon, J. S. (2004). Deficits of organizational strategy and visual memory in obsessive-compulsive disorder. Neuropsychology, 18(4), 665-672. 49 Simpson, H. B., Rosen, W., Huppert, J. D., Lin, S., Foa, E. B., & Liebowitz, M. R. (2006). Are there reliable neuropsycholog ical deficits in obsessive-corn pulsive disorder. Journal of Psychiatric Research, 40, 247-257. Smith, G. V., Della Salla, S, Logie, R. H, & Maylor, E. A. M. (2000). Prospective and retrospective memory in normal ageing and dementia: A questionnaire study. Memory, 8,311-321. Stein, M. B., Forde, D. R., Anderson, G., & Walker, J. R. (1997). Obsessive-compulsive disorder in the community: An epidemiologic survey with clinical reappraisal. American Journal of Psychiatry, 154(8), 1120-1126. Talus, F. (1995a). Obsessive compulsive disorder: A cognitive and neuropsychological perspective. Chichester: John Wiley & Sons. Talus, F. (1995b). The characteristics of obsessional thinking: Difficulty demonstrating the obvious? Clinical Psychology and Psychotherapy, 2(1), 24-39. Talus, F. (1997). The neuropsychology of obsessive-compulsive disorder: A review and consideration of clinical implications. British Journal of Clinical Psychology, 36, 3-20. Talus, F., Pratt, P., Jamani, N. (1999). Obsessive compulsive disorder, checking and non-verbal memory: A neuropsychological investigation. Behaviour Research and Therapy, 37, 161-166. Taylor, S. et al. (2006). Do dysfunctional beliefs play a role in all types of obsessive-compulsive disorder? Journal of Anxiety Disorders, 20, 85-97. Tolin, D. F., Abramowitz, J. S., Bartholomew, D. B., Amir, N., Street, G. P., & Foa, E. B. (2001). Memory and memory confidence in obsessive-compulsive disorder. Behaviour Research and Therapy, 39, 913-927. Tuna, S., Tekcan, A. L, Topçuolu, V. (2005). Memory and metamemory in obsessive- compulsive disorder. Behaviour Research and Therapy, 43, 15-27. UttIl, B. (2008). Transparent meta-analysis of prospective memory and aging. PLoS ONE, 3(2): el 568. doi: 10.1 371/journal.pone.0001 568. Weissman, M. M., Bland, R. C., Canino, G. J., Greenwald, S., Hwu, H. G., Lee, C. K., et al. (1994). The cross national epidemiology of obsessive compulsive disorder: The cross national collaborative group. Journal of Clinical Psychiatry, 55, 5-10. Wilson, B., Cockburn, J., & Baddeley, A. (1985). The Rivermead behavioral memory test manual. England: Thames Valley Test Company. 50 Woods, C. M., Vevea, J. L., Chambless, D. L., & Bayan, U. J. (2002). Are compulsive checkers impaired in memory? A meta-analytic review. Clinical Psychology: Science and Practice, 9(4), 353-366. Zermatten, A., Van der Linden, M., Larøi, F., Ceschi, G. (2006). Reality monitoring and motor memory in checking-prone individuals. Journal of Anxiety Disorders, 20, 580-596. Zielinski, C. M., Taylor, M. A., & Juzwin, K. R. (1991). Neuropsychological deficits in obsessive compulsive disorder. Neuropsychiatry, Neuropsychology, and Behavioural Neurology, 4(2), 110-126. Zitterl, W., Urban, C., Linzmayer, L., Aigner, M., Demal, U., Semler, B., Zitterl-Eglseer, K. (2001). Memory deficits in patients with DSM-IV obsessive-compulsive disorder. Psychopathology, 34, 113-117. 51 CHAPTER 2 SUB-CLINICAL COMPULSVIE CHECKERS’ PROSPECTIVE MEMORY IS IMPAIRED3 Obsessive-Compulsive Disorder is an anxiety disorder that is characterized by obsessions and compulsions. Obsessions are intrusive and persistent ruminations or impulses that cause anxiety, which is temporarily relieved by the execution of compulsions, repetitive behaviors or mental acts such as washing, counting or checking (American Psychiatric Association, 2000). While there are several different manifestations or subtypes of OCD, checking compulsions are the most common manifestation (Henderson & Pollard, 1988; Ramussen & Eisen 1988; Stein, Forde, Anderson, & Walker, 1997). Individuals with checking compulsions have intrusive doubts that they failed to do something or failed to do something correctly, and they feel compelled to check to ensure that a task was indeed completed and/or that it was completed properly (Muller & Roberts, 2005; Woods, Vevea, Chambless, & Bayan, 2002). For instance, an individual with checking compulsions may have the reoccurring intrusive doubt that s/he forgot to lock the front door of her/his home, worry about the consequences of such a failure (e.g., being robbed) and then feel compelled to return home to check the door. Very little is understood about the mechanisms that contribute to compulsive checking. However, according to one prominent theory the root cause of checking compulsions is an impairment of memory. This impairment may be real, marked by an objective deficit in remembering important activities, such as locking the front door, or it may be subjective and expressed primarily as a lack of confidence in the ability to remember (Sher, Frost, & Otto, 1983). Because checkers are either unable to recall performing an activity or because they do not trust their memory for performing the activity, they feel compelled to check. In a recent comprehensive review, Muller and Roberts (2005) argued that the memory deficit theory of compulsive checking receives at best only qualified support from the results of empirical investigations. However, this conclusion conflicts with the outcome of a prior comprehensive meta-analysis of research with both clinical and sub-clinical checkers by Woods et al. (2002). Based on the results of their analyses, Woods et al. (2002) concluded that checkers have an objectively verifiable impairment in short-term memory as well as in episodic memory. Moreover, they argued that checkers also have a subjective memory impairment, they lack confidence in their ability to remember (see also Hermans, Martens, De Cort, Pieters, & Eelen, 2003; Macdonald, Antony, Macleod, & Richter, 1997; Tuna, Tekan, & Topcuoglu, 2005). A version of this chapter has been published. Cuttler, C. & Graf, P. (2007) Compulsive checkers’ prospective memory is impaired. Journal of Anxiety Disorders, (3), 338-352. 52 To our knowledge, the memory deficit theory of compulsive checking has never been framed or tested in the domain of prospective memory, defined as the ability to formulate intentions, plans and promises, to retain them, to recollect and carry them out at the appropriate time or in the appropriate context (Einstein & McDaniel, 1996; Graf, 2005; Graf& UttI, 2001; Meacham & Dumitru, 1976). This ability to formulate and/or remember intentions seems intimately connected with the kinds of everyday tasks that frequently trigger checking compulsions, such as, remembering to turn off the stove or iron after use or to lock the front door upon leaving home. For this reason, the main goal of the present study was to explore the possibility that compulsive checkers have a cognitive deficit related to prospective memory. We suggest three possible ways in which checking compulsions might be related to prospective memory. First, individuals with checking compulsions may have difficulty de activating their intentions after executing them. Previous research has found that in normal healthy adults, once a prospective memory task has been completed, the memory representations of words related to that task are inhibited (Marsh, Hicks, & Binks, 1998). In contrast, words related to a task that someone else (e.g., the experimenter) had to perform and words that simply needed to be remembered for a later recall test show no such inhibition (Marsh et al., 1998). It has been argued that the already-executed-intention inhibition effect, or the mechanism(s) that mediates it, is responsible for preventing us from perseverating on already completed intentions (Ellis, 1996; Marsh et al., 1998). By this view, if the inhibition mechanism fails, a task may remain part of our ‘to-do-list’, and thereby contribute to compulsive checking behaviors. A second possibility is that individuals with checking compulsions fail to remember intentions that they have already executed, and consequently, they worry about still needing to execute them. By this view, the root cause of compulsive checking is a retrospective memory deficit for prospective memory tasks, a hypothesis that fits the traditional memory deficit theory of compulsive checking behavior (Sher, Frost, & Otto, 1983), except that it underscores amnesia for intentions rather than for previous events and experiences. The third and final possibility is that individuals with checking compulsions have generally poor prospective memory skills, have a long history of prospective memory failures, are aware of those failures and consequently they worry about future failures. Thus, when the perceived consequences of a failure are serious (e.g., the failure to unplug the iron after use might cause a fire), the fear of a future prospective memory failure may trigger compulsive checking behaviors. Moreover, checking behaviors may be triggered not only by the fear of objectively verifiable poor prospective memory task performance but also by a person’s subjective lack of confidence in his/her prospective memory ability. For a person who knows or 53 believes that his/her prospective memory is not reliable, checking is a reasonable strategy for dealing with the impairment. In other words, we suggest that checking compulsions are a compensatory reaction to an objective and/or subjective impairment in prospective memory. The main goal of the present study was to examine whether or not prospective memory task performance is impaired in individuals who are compulsive checkers. Although we have identified three potentially different prospective memory deficits related to compulsive checking, our study focused most directly on the last, the general possibility that individuals who are compulsive checkers suffer from an objective and/or a subjective impairment in prospective memory task performance. Study Overview Each participant in the present study completed a set of questionnaires, two laboratory- based episodic prospective memory tasks and a battery of standardized neuropsychological assessment tests. We used participants’ questionnaire responses as a subjective index of (i.e., a measure of confidence in) their prospective memory abilities and we used participants’ performance on the two laboratory tasks as an objective index of their prospective memory abilities. The questionnaire set included the Padua Inventory (Sanavio, 1988), an instrument that focuses on obsessive-compulsive behavior, and contains a detailed subscale for measuring checking behavior. We used participants’ scores on the checking subscale to classify them into one of three groups: High checkers, medium checkers and low checkers. The questionnaire set also included two inventories related to prospective memory: The Prospective Memory Questionnaire (PMQ; Hannon, Adams, Harrington, Fries-Dias, & Gibson, 1995) and the Prospective and Retrospective Memory Questionnaire (PRMQ; Smith, Della Sala, Logie, & Maylor, 2000). Each of these questionnaires requires participants to rate the frequency of experiencing various kinds of memory failures. Specifically, the PMQ probes failures connected with episodic prospective memory tasks, with habitual prospective memory tasks and with internally cued prospective memory tasks. Episodic prospective memory tasks are those where a to-be-performed plan is not in conscious awareness during the retention interval and needs to be executed only once (e.g., making an important phone call after dinner) (Graf, 2005; Kvavilashvili & Ellis, 1996; Meacham & Dumitru, 1976). Habitual prospective memory tasks require repeated execution of the same plan (e.g., removing the keys from the car before closing and locking the doors) (Graf & UttI, 2001; Hannon et al., 1995; Kvavilashvili & Ellis, 1996; Meacham & Dumitru, 1976). Internally cued prospective memory tasks are those where no salient external cue is provided for prompting the retrieval of the previously formed 54 plan (e.g., cashing a paycheck before running out of money) (Hannon et al., 1995). The PRMQ probes the frequency of both prospective and retrospective memory failures. We assumed participants’ ratings on the PMQ and the PRMQ would reflect both their perceptions of everyday experiences with prospective memory failures as well as confidence in their memory skills. Consistent with the further assumption that compulsive checking is a response to a subjective impairment in prospective memory related abilities we expected the high checkers’ questionnaire scores to evidence lower confidence. The assessment battery also included two laboratory-based episodic prospective memory tasks, one event-cued and the other time-cued (Einstein & McDaniel, 1990; Graf & Grodin, 2005). For the event-cued task, participants had to request the return of a personal belonging upon the occurrence of a specified spoken cue. For the time-cued task, participants were asked to remind the experimenter to place a phone call in exactly 30 minutes. In connection with these tasks, we also collected self-assessment data, first, by asking participants to rate how confident they felt about succeeding on the event-cued task, and second, by asking them to predict how close they would come (in seconds) to providing the phone-call reminder. For the time-cued task, participants had access to a clock, and we recorded clock-checking responses for an additional index about task-related planning and worrying. Finally, the assessment battery included several standardized tests for measuring attention, cognitive flexibility and verbal fluency. The primary purpose of these tests was to occupy participants during the retention intervals that were required for the objective episodic prospective memory tasks. Method Participants and Design Sub-clinical checking compulsions are fairly common among otherwise healthy and normal functioning undergraduate students (Sher et al., 1983), and compared to patients with clinical checking compulsions students are far more readily accessible and typically free of the comorbid afflictions (e.g., anxiety, depression) often present in patients. For these reasons, we used undergraduate students for the present study. We proceeded on the assumption that if a study with students shows evidence in support of our modified memory deficit theory, this evidence most likely would be even stronger in individuals with fully developed clinical checking corn pu Isions. We recruited 126 undergraduate students from the University of British Columbia Department of Psychology subject pool. In order to ensure our sample included individuals with a wide range of checking behaviors, we used two types of advertisements, one that called 55 specifically for participants with checking compulsions and another that made no mention of compulsions. One hundred participants were recruited by means of the latter, more general advertisement which made no mention of compulsions and 26 were recruited by means of the specific advertisement that called for individuals with checking compulsions. Participants ranged from 18 to 47 years of age with a mean of 20.95 years (SD = 3.47). Participant’s postsecondary education ranged from 1 to 6 years with a mean of 2.51 years (SD = 1.09). Forty of the participants were male and 86 were female. In order to address our main research objectives, the overall design of our study had to be quasi-experimental. We used participants’ scores on the checking subscale of the Padua Inventory to assign them to one of three groups: Low checkers, medium checkers and high checkers. Specifically, participants with checking scores between 0 and 4 were assigned to the low checkers group, participants with scores between 5 and 11 were assigned to the medium checkers group, and participants with scores of 12 or higher were assigned to the high checkers group. These cut-off points were chosen because they yielded approximately the same number of participants per group: 40 low checkers, 41 medium checkers and 45 high checkers. These criteria yielded groups with mean Padua Inventory checking scores of 2.52, 7.56, 18.02, respectively for the low, medium and high checkers. All of the low checkers, 92.7% of the medium checkers and 48.9% of the high checkers were recruited by means of our general advertisement; the remaining subjects were recruited by means of the advertisement that targeted individuals with checking compulsions. As one would expect based on the demographically homogeneous pool from which we drew our sample, the results of separate one-way ANOVAs showed that the three checking groups did not differ in age, F (2, 125) = 2.94, MSE = 34.37, p> .05, or in years of postsecondary education, F (2, 125) = 1.52, MSE = 1.80, p> .05. The three groups also did not differ in gender, x2 (2) = 3.24, p> .05. We conducted this research with the approval of the University of British Columbia Behavioral Ethical Review Board. Completing all components of the study required approximately one hour and participants received course credit in return for their participation. Instruments In the course of the experiment each participant was administered several questionnaires, prospective memory tasks and neurocognitive filler tests. Each of these study components is briefly described below. Padua Inventory. The Padua Inventory is a standardized self-report inventory developed by Sanavio (1988). The inventory contains 60 statements describing common obsessive-compulsive behaviors (e.g., I return home to check doors, windows, drawers, etc. to 56 make sure they are properly shut). For each item participants use a 5-point scale to indicate the degree of disturbance each behavior creates, with scale points marked: None at all, a little, quite a lot, a lot and very much. The inventory contains 4 subscales each of which measures different obsessive-compulsive concerns: Checking, impaired control of mental activities, contamination and worries about losing control over motor behaviors. The questionnaire is a reliable measure of obsessive-compulsive behavior, as evidenced by high test-retest reliability (.80) (Sanavio, 1988), and alpha coefficients (.94) (Sanavio, 1988). We computed participants’ checking scores according to Sanavio (1988). Possible scores on the checking subscale range from 0 to 32, with higher scores indicating a higher level of checking behavior. Prospective and Retrospective Memory Questionnaire (PRMQ). The PRMQ is a standardized self-report inventory developed by Smith et al. (2000). The scale contains 16 questions concerning various memory failures (e.g., Do you forget appointments if you are not prompted by someone else or by a reminder such as a calendar or diary?). For each item, participants indicate the frequency of experiencing the failure using a 5-point scale, with scale points marked: Never, a little, quite a lot, a lot and very often. The scale contains two subscales, one focusing on prospective memory and the other on retrospective memory. Both subscales yield high reliability coefficients (prospective memory = .84, retrospective memory = .80) (Crawford, Smith, Maylor, Della Sala, & Logie, 2003). We computed scores for each of the PRMQ subscales according to Crawford et al. (2003). Possible scores on each subscale range from 1 to 5, with higher scores indicating a greater frequency of memory failures. Prospective Memory Questionnaire (PMQ). The PMQ is a standardized self-report inventory developed by Hannon et al. (1995). The scale contains 52 statements concerning various prospective memory failures (e.g., I forgot to lock up my house, bike or car) and prospective memory aiding techniques (e.g., I write myself reminder notes). For each item participants rate the frequency of experiencing the failure or using the technique. They use a 9- point scale, with the scale points representing how many times in a given time period (e.g., a week, month or year) they experience each prospective memory failure or use each memory aiding technique. The questionnaire contains four subscales, three of which concern failures connected with different prospective memory tasks, specifically, episodic tasks, habitual tasks and internally cued tasks, and the fourth measures the frequency of using various prospective memory aiding strategies. Each subscale is reliable, with alpha coefficients ranging from .78 to .90, and the test-retest reliabilities ranging from .64 to .88 (Hannon et al., 1995). Possible scores on each subscale range from 1 to 9, with higher scores indicating a greater frequency of memory failures or use of memory aiding techniques. 57 Personal Belonging Task. We used a slightly modified version of the standardized belonging task (Wilson, Cockburn, & Baddeley, 1985) for assessing event-cued episodic prospective memory performance. Participants were told we were interested in their ability to remember to do things at a later time, and that we needed one of their personal belongings for this purpose. If the participant was wearing a watch then this was taken, thereby also ensuring that no participant was wearing a watch when completing the time-cued prospective memory task. If the participant was not wearing a watch then a cell phone or something of similar value was requested. We placed the personal belonging in a drawer and instructed participants to ask for its return upon being told later in the testing session: “we are now finished with all of the tests.” Participants were further informed that they would receive this prompt immediately before the debriefing. After ensuring that participants understood the instructions, we asked them for a confidence rating of their ability to remember to request back the personal belonging at the appropriate moment. We used a 10-point scale for the confidence ratings where I indicated extremely unconfident and 10 indicated extremely confident. Those participants who requested the return of their belonging at the appropriate moment were awarded a score of 1, indicating success; otherwise they received a score of 0. Phone Call Reminder Task. Inspired by Searleman (1996) and Kvavilashvili (1987) we used a Phone Call Reminder task for assessing time-cued episodic prospective memory task performance. Participants were told that in exactly 30 minutes the experimenter needed to leave the room to place an important phone call. They were instructed to tell the experimenter when exactly 30 minutes had passed, because we wished to find out “how good [they] are at giving this kind of reminder.” The instructions emphasized getting as close as possible to the 30 minute mark. Participants were given a stopwatch for keeping track of time and were told to look at it whenever they wished, but to keep it facedown whenever they were not checking it. Participants also were asked to predict how many seconds or minutes late (i.e., past 30 minutes) they would be on this task. The experimenter had a synchronized stopwatch, permitting her to record the precise time of each discrete stopwatch reference. The time of the reminder was also recorded. The delay (i.e., past 30 minutes) in providing the reminder served as our primary measure of time-cued prospective memory task performance accuracy. Neurocognitive Filler Tests. During the retention interval, we occupied participants by having them complete three blocks of a lexical decision test, Graf, UttI and Tuokko’s (1995) version of the Color-Word Stroop test (Stroop, 1935), a modified version of the standardized letter cancellation test (Diller, Ben Yishay, Gerstman, 1974), parts A and B of Reitan’s (1992) Trail Making test, and a modified version of the standardized FAS test of verbal fluency (Benton & Hamsher, 1989). These tasks were administered according to published instructions. 58 Procedure Table 2.1 shows the order in which the tasks and instruments were administered to each participant and the approximate time required for completing each. The episodic prospective memory tasks are highlighted. Table 2.1. A List of all Instruments and Tasks Assigned to Participants, Arranged According to When They Were Administered, Showing the Time Required to Complete Each. Approximate Instrument/Task Time Required (mm) Demographic Information 1 Assign Personal Belonging Prospective Memory Task and 1 Obtain Confidence Rating Questionnaires Padua Inventory 7 Prospective and Retrospective Memory Questionnaire 4 Prospective Memory Questionnaire 10 Assign Phone Call Reminder Prospective Memory Task and 1 Obtain Performance Prediction Neuropsychological Tests (Filler Activities) Lexical Decision Task — Set 1 6 Color-Word Stroop Test 5 Cancel H Test I Lexical Decision Task — Set 2 6 Trail Making Test Parts A and B 4 Verbal Fluency Test 2 Lexical Decision Task — Set 3 10 Debriefing 2 Participants were tested one at a time in a quiet room. Upon obtaining their written informed consent, we collected basic demographic information. We then assigned the event- cued Personal Belonging prospective memory task, and participants’ rated their confidence in being able to perform this task as instructed. Participants then completed the set of questionnaires listed in Table 2.1 according to their own pace. For the second phase of the experiment participants were assigned the time-cued Phone Call Reminder prospective memory task and participants made predictions about their performance on this task. Participants then completed the battery of neurocognitive filler tests 59 listed in Table 2.1. The filler tests, most of them commonly used neuropsychological tests, were administered in four 6-minute blocks and one 10-minute block. At the end of each block, participants were told to stop what they were doing and proceed to the next block of tests. We used this procedure to prevent the boredom that may have ensued if participants had been required to work on the same task for the entire 30 minute retention interval, and to ensure that all participants progressed through the neuropsychological tests at about the same rate. For the first 6-minute block, participants completed Block 1 of the lexical decision test. For the next 6- minute block participants were asked to complete the Color-Word Stroop test (Graf et al., 1995) and a modified version of the letter cancellation test (Diller et al., 1974). This was followed by Block 2 of the lexical decision test. During the next 6-minute block participants were given parts A and B of the Trail Making test (Reitan, 1992) and a modified version of the FAS test (Benton & Hamsher, 1989). During the final 10-minute block participants completed Block 3 of the lexical decision test. The end of the 30 minute retention interval fell 6 minutes into the final block, and thus, participants were expected to carry out the Phone Call Reminder prospective memory task (i.e., remind the experimenter to place the phone call) while occupied with making lexical decisions. After participants completed all of the filler tasks, the experimenter delivered the cue for the Personal Belonging prospective memory task with the phrase: “we are now finished with all of the tests.” To create a reasonable time-window for responding to this cue, the experimenter took about 10 seconds to tidy up and put the testing materials away and then informed participants that they were going to be debriefed. Then, participants received a full debriefing and their research participation credit. Resu Its Data Preparation All data were double checked for recording and transcription errors. Data accuracy was greater than 99%. There were five missing values on the questionnaires (i.e., < .001%), and each of these was replaced with the mean response from the respective checking group (i.e., high, medium, low checkers). In addition, one missing Phone Call Reminder prediction score (i.e., <.01) had to be replaced with the mean of the respective group. We examined all of the questionnaire data and the prospective memory data (i.e., confidence ratings, predictions, prospective memory success scores and clock-checking data) for univariate outliers, defined as scores falling more than three standard deviations away from their respective means. We discovered five outliers in the data from the low checkers, six in the data from the medium checkers and four in the data from the high checkers. These outlying 60 values were replaced with the nearest non-outlying value, specifically, a score either -3 or +3 standard deviations away from the corresponding group mean4. Objective Measures of Prospective Memory Performance One aim of the present study was to investigate whether high checkers have an objective impairment related to either event- or time-cued prospective memory tasks. For the event-cued task, participants had to request the return of a personal belonging when they were told: “we are now finished with all of the tests.” Our results showed that 55.6% of the high checkers, 58.5% of medium checkers and 80% of low checkers requested the return of their personal belonging. A chi-square analysis of participants’ performance on the Personal Belonging task showed a significant difference among the groups, x2 (2) 6.39, p < .05. Follow- up analyses revealed that the low checkers performed significantly better than the medium or high checkers, with 2(1) = 4.37, p < .05 and 2(1) 5.73, p < .05, respectively. The medium checkers’ performance was not different from that of the high checkers, x2 (1) = .08, p> .05. The correlation between checking scores from the Padua Inventory and performance on the Personal Belonging task was also significant, r (124) = -.15, p < .05 (one-tailed). For the time-cued prospective memory task participants had to remind the experimenter to make a phone call in exactly 30 minutes. High checkers provided this reminder on average 58.4 seconds late, medium checkers provided it 69.1 seconds late and low checkers provided it 49.8 seconds late. A one-way ANOVA of these reminder lateness scores showed no significant differences among the participant groups, F (2, 125) = .48, MSE = 3781.87, p> .05. The correlation between checking scores and performance on the Phone Call Reminder task was also not significant, r (124) = .02, p> .05. On the time-cued prospective memory task, we also recorded participants’ clock- checking behavior. This behavior is summarized in Figure 2.1, which shows the mean frequency of clock-checking by each participant group for each 5-minute retention-interval- period. A 3 x 6 ANOVA with Group (low, medium, high checkers) as a between subjects factor and the successive 5-minute retention-interval-period as a within subject factor supports the main message of the figure, namely, that the groups did not differ in their clock-checking strategies, F (2,123) = .99, MSE = 3.06, p> .05. The analysis showed a significant main effect due to the 5-minute retention-interval-periods, F (5, 615) = 59.03, MSE = 1.10, p <.05, but no other significant main or interaction effects. ‘ When we reran the analyses without the outlying scores replaced the results reported in this article remained the same. 61 Figure 2.1. The Frequency of Clock-Checking Responses on the Phone Call Reminder Task Across the Three Checking Groups. 3 0 02 C.) 0 0 0 I a, - E z 0 Subjective Measures of Prospective Memory Performance Do high checkers suffer from subjective impairments in prospective memory? To answer this question, we compared the three checking groups’ confidence ratings on the Personal Belonging task, their predicted lateness scores on the Phone Call Reminder task, and their ratings on the PMQ and the PRMQ. For the event-cued Personal Belonging task, we asked participants to indicate how confident they were about requesting the return of their personal belonging, and participants provided a rating out of 10, with 1 indicating extremely unconfident and 10 indicating extremely confident. The groups’ confidence ratings were similar, with means of 7.31, 7.80 and 7.38, respectively, for the high, medium and low checkers. A one-way ANOVA of participants’ confidence ratings, with Group (low, medium, high checkers) as a between subjects factor, was not significant, F (2, 125) = 1.32, MSE = 3.02, p> .05. The correlation between checking scores and confidence ratings also was not significant, r (124) = .01, p> .05. On the time-cued prospective memory task we asked participants to predict how late they would be in providing the phone call reminder. The mean prediction latencies were 174.4 seconds, 187.7 seconds and 195.6 seconds, respectively, for the high, medium and low checkers. A one-way ANOVA of the prediction latencies, with Group (low, medium, high checkers) as a between subjects factor, was not significant, F (2, 125) = .22, MSE 4890.22, p > .05. The correlation between checking scores and prediction latencies also was not significant, r (124) = .02, p> .05. • Low Checkers —ci—— Medium Checkers - -- - High Checkers 1-5 mins 5-10 mins 10-15 mins 15-20 mins 20-25 mins 25-30 mins Time Interval 62 The main findings from the PMQ are depicted in Figure 2.2. We conducted an ANOVA on the ratings concerning the frequency of experiencing different kinds of prospective memory failures (i.e., the left three sets of bars in Figure 2.2), with Group (low, medium, high checkers) as a between subjects factor and Prospective Memory Task Type (episodic, habitual, internally cued) as a within subject factor. The results showed significant main effects due to Group, F (2, 123) = 11.43, MSE = 1.96, p < .05, and due to Prospective Memory Task Type, F (2, 246) = 176.96, MSE .51, p < .05. The interaction between these factors did not reach significance, F (4, 246) 1.12, MSE = .51, p> .05. Follow-up one-way ANOVAs on the data from each prospective memory task type (i.e., on the left three sets of bars in Figure 2.2) showed significant main effects due to Group on the episodic prospective memory scale, F (2, 125) = 7.37, MSE = 9.30, p <.05, on the habitual prospective memory scale, F (2, 125) = 11.87, MSE = 3.82, p <.05, and on the internally cued prospective memory scale, F (2, 125) = 7.51, MSE 10.46, p < .05. In all cases, independent samples t-tests confirmed that high checkers gave significantly higher prospective memory failure ratings than the medium checkers, with t (84) = - 3.36, p <.05, t (84) = -3.52, p <.05, and t (84) = -2.99, p < .05, respectively, for the episodic, habitual and internally cued scales. The correlations between checking scores and failure ratings on each prospective memory task type were also significant, r (124) = .28, p < .05, r (124) = .44, p < .05, r (124) = .34, p <.05, respectively, for the episodic, habitual and internally cued scales. Figure 2.2. Self-Rated Frequency of Prospective Memory Failures and Use of Prospective Memory Aiding Strategies Across the Three Checking Groups. •D. Internally Cued ProspectRe Memory The rightmost set of bars in Figure 2.2 shows participants’ self-rated use of prospective memory aiding strategies. A one-way ANOVA on the corresponding data showed a significant Low Checkers Medium Checkers D High Checkers E Episodic Habitual Prospectie Prospecti’ve Memory Memory Prospecti’ve Memory Aiding Strategies 63 main effect due to Group, F (2, 125) = 4.44, MSE = 14.53, p < .05, with high checkers reporting using more prospective memory aiding strategies than the low checkers, t (83) = -2.97, p <.05. High and medium checkers did not differ significantly in their reported use of strategies, t (84) = -1.73, p> .05, and the medium checkers did not differ from the low checkers, t (79) = -1.17, p> .05. The correlation between checking scores and self-rated use of prospective memory aiding strategies was significant, r (124) = .30, p < .05. In a final ANOVA, we examined participants’ ratings on the PRMQ which explored the frequency of experiencing both prospective and retrospective memory failures. The first analysis had Group (low, medium, high checkers) as a between subjects factor and Memory Type (prospective memory, retrospective memory) as a within subject factor. The results, displayed in Figure 2.3, revealed a significant effect of Group, F (2, 123) = 7.52, MSE = .78, p < .05, and Memory Type, F (1, 123) = 76.02, MSE = .107, p < .05, as well as a significant interaction between these two factors, F (2, 123) = 3.15, MSE .107, p < .05. Follow-up ANOVAs revealed that the three groups gave significantly different ratings concerning their experience of prospective memory failures, F (2, 125) = 7.56, MSE = 4.40, p < .05, as well as their experiences of retrospective memory failures, F (2, 125) = 5.93, MSE = 1.83, p < .05. The correlations between checking scores and prospective and retrospective memory failure ratings were also significant, r (124) = .29, p < .05, r (124) = .34, p < .05, respectively. Figure 2.3. Self-Rated Frequency of Prospective and Retrospective Memory Failures Across the Three Checking Groups. Discussion According to one prominent theory, the root cause of compulsive checking is an impairment of memory. This impairment may be real, marked by an objective inability to • Low Checkers 1 Medium Checkers c High Checkers Prospecthe Memory Retrospecti’de Memory 64 remember important activities, such as turning the iron off after use, or it may be subjective and expressed primarily as a lack of confidence in the ability to remember (Sher et al., 1983). For the present study, we framed and tested this theory in the context of prospective memory, primarily because it appears that checking tendencies seem closely linked with activities that require execution after a delay, or rather, with the failure — or a lack of confidence in the ability — to carry out such activities. Put more concretely, if individuals frequently forget to carry out activities or if they believe that they frequently forget to carry out activities then checking may develop as a compensatory strategy for ensuring that important activities are completed. Our participants were undergraduate student volunteers whom we classified as high, medium or low checkers on the basis of their scores on the Padua Inventory. We required participants to complete two laboratory based prospective memory tasks, one time-cued (the Phone Call Reminder task) and the other event-cued (the Personal Belonging task), to make confidence ratings and performance predictions about these task and to complete two questionnaires on the self-assessed frequency of various kinds of memory failures and use of memory aiding strategies. The results from the study showed that high and medium checkers performed worse than low checkers on the event-cued episodic prospective memory task, but there was a complete absence of an effect of group on the time-cued task. The three participant groups were similarly confident in their ability to carry out the event-cued task, and they made comparable predictions about how late they would be on the time-cued task. By contrast, on the two questionnaires the high checkers reported experiencing more of all types of memory failures than either the medium checkers or the low checkers, and they also reported making more use of prospective memory aiding strategies. On the assumption that the questionnaire responses provide a reliable and valid confidence or self-assessment index, the questionnaire results are consistent with the claim that compared to low and medium checkers, high checkers suffer from a subjective memory impairment. This impairment seems broad-based and to cover all aspects of prospective memory that were probed by the PMQ, and the data from the PRMQ (see Figure 2.3) show that consistent with the traditional memory deficit theory this impairment also extends to retrospective memory. A surprising outcome in the present study is the absence of differences among the checking groups in the success confidence ratings on the event-cued task and in the prediction accuracy ratings about completing the time-cued task. This finding is surprising because both the success confidence ratings and the prediction accuracy ratings are self-produced, subjective indexes of performance, and consequently, we expected them to show the same 65 pattern of checking group differences as the questionnaire data. The failure to find the same group effects in both data sets may provide an important clue about the factors that determine participants’ self-ratings under different assessment conditions. For example, it is possible that when required to respond to the general questions about memory failures on the PMQ and the PRMQ, subjects’ ratings are filtered by an availability heuristic (Tversky & Kahneman, 1973), and consistent with the memory deficit theory of compulsive checking, when guided by this heuristic, high checkers have an easier time accessing memory failures than either medium checkers or low checkers. By contrast, when required to respond to the specific demands of a particular prospective memory tasks (e.g., the Personal Belonging task or the Phone Call Reminder task in the present experiment), self-ratings may be determined primarily by the real demands of the task and not by the availability or accessibility of similar previous experiences. Future research will need to explore this possibility. In the present study, we also sought out objective prospective memory task performance differences among the checking groups, and we found such differences on an event-cued task (i.e., the Personal Belonging task) but not on a time-cued task (i.e., the Phone Call Reminder task). The absence of an effect due to Group on the latter is interesting. Previous research reveals that performance differences between event- and time-cued prospective memory tasks are not uncommon (for example, Bastin & Meulemans, 2002; Einstein & McDaniel, 1995; Nigro, Senese, Natullo & Sergi, 2002) but they typically show the opposite to our findings, that is, effects on time- but not event-cued tasks. To explain this type of finding, it has been argued that compared to event-cued tasks, time-cued tasks are more difficult or resource demanding because they depend more heavily on subject-initiated processes (Craik, 1996; Einstein & McDaniel, 1990). Consistent with this task-difficulty view, we had expected larger performance differences among our checking groups on the time- than event-cued task, in keeping with the additional assumption that the checking compulsion itself is resource demanding (Eysenck & Calvo, 1992; Humphreys & Revelle, 1984) and thus depletes the resources that can be allocated to the prospective memory task. Thus, rather than reflecting a difference in the sensitivity of the event- and time-cued tasks, our finding of a group effect appears due to a more fundamental process that is implicated in compulsive checking. One possible explanation for why we found differences among the checking groups on the event- but not time-cued task focuses on the contexts that foster compulsive checking behaviors. It seems possible that compulsive checking becomes evident, prominent and performance-limiting only in situations where there is uncertainty that cannot be reduced or removed by strategic (i.e., planned, systematic, skillful) behaviors. Time-cued tasks permit this kind of behavior; they supply a clear, familiar and prominent dimension (i.e., time) for 66 anticipating or predicting the context where a planned activity needs to be carried out (Graf & Grodin, 2005). By contrast, for event-cued tasks, the occurrence of a relevant cue may not be predictable. The greater uncertainty inherent in event-cued situations may fuel the compulsion to check. The possibility that checking compulsions only impact performance on event-cued prospective memory tasks is consistent with the examples of checking behaviors that tend to be given in the research literature (e.g., checking to ensure that the stove or iron was turned off after use). We are not aware of any literature references to checking behaviors that occur in connection with time-cued tasks. Indeed, it does not seem possible to use the same checking strategy for ascertaining the correct completion of time- and event-cued tasks. Once the time to perform a planned task (e.g., pick a friend up at the airport at 2pm) has passed, the opportunity to perform that task has expired and thus checking serves no purpose. However, if one fails to unplug her/his iron, returning home to correct the failure is potentially beneficial. Conclusion The results of the present study are consistent with the traditional memory deficit theory of compulsive checking, and more importantly, they support the broader theoretical claim that deficits in prospective memory and in retrospective memory work together to produce and maintain compulsive checking tendencies. We envisage the following vicious cycle. Individuals with a propensity for compulsive checking have an impaired prospective memory, causing them to experience frequent prospective memory failures. These failures undermine confidence in prospective memory and lead to intrusive doubts about similar failures. The intrusive doubts may lead individuals to rely on the retrospective recollection of tasks that have/have not been completed, but in so doing, they experience additional failures because of a deficit in retrospective memory. The individual then feels compelled to check to ensure that the task has been completed. Our findings must be interpreted cautiously because they were obtained from undergraduate students, that is, from sub-clinical checkers who manage to function effectively under demanding everyday circumstances. Concerns about the generalizability of our findings are not triggered by low checking scores in our sample; as described in the Method section, our high checkers obtained a mean checking score of 18.02 on the Padua Inventory, a value that is comparable to that reported in previous research with clinical checkers (Hermans et al., 2003; Tallis, Pratt, & Jamani, 1999). Rather, concerns about generalizability arise because in clinical samples, compulsive checking tendencies may potentially manifest with greater degrees of comorbid afflictions (e.g., depression, anxiety) which may alter the pattern of effects we observed in the present study. 67 References American Psychiatric Association (2000). Diagnostic and Statistical Manual of Mental Disorders (4th Edition). Washington, DC: American Psychiatric Association. Bastin, C, & Meulemans, T. (2002). Are time-based and event-based prospective memory affected by normal aging in the same way? Current Psychology Letters: Behavior, Brain & Cognition, 7, 105-121. Benton, A.L., & Hamsher, K. (1989). Multilingual aphasia examination. Iowa City: AJA Associates. Craik, F. I. M. (1986). A functional account of age differences in memory. In F. Klix & H. Hagendorf (Eds.), Human memory and cognitive capabilities: Mechanisms and performances (pp. 409-422). New York: Elsevier Science. Crawford, J. R., Smith, G., Maylor, E. A., Della Sala, S., Logie, R. H. (2003). The prospective and retrospective memory questionnaire (PRMQ): Normative data and latent structure in a large non-clinical sample. Memory, 11(3), 261-275. Diller, L., Ben-Yishay, Y., Gerstman, L. J. (1974). Studies in cognition and rehabilitation in hemiplegia (Rehabilitation monograph No. 50). New York: New York University Medical Center Institute of Rehabilitation Medicine. Einstein, G. 0., & McDaniel, M. A. (1990). Normal aging and prospective memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 16(4), 717-726. Einstein, G. 0., & McDaniel, M. A. (1995). Aging and prospective memory: Examining the influences of self-initiated retrieval processes. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21(4), 996-1007. Einstein, G. 0., & McDaniel, M. A. (1996). Retrieval processes in prospective memory: Theoretical approaches and some new empirical findings. In M. Brandimonte, G. 0. Einstein, & M. A. McDaniel (Eds.) Prospective memory: Theory and applications. New Jersey: Lawrence Erlbaum Associates. Ellis, J. (1996). Prospective memory or the realization of delayed intentions: A conceptual framework for research. In M. Brandimonte, G. 0. Einstein, & M. A. McDaniel (Eds.) Prospective memory: Theory and applications. New Jersey: Lawrence Erlbaum Associates. Eysenck, M. W., & Calvo, M. G. (1992). Anxiety and performance: The processing efficiency theory. Cognition and Emotion, 6(6), 409-434. 68 Graf, P. (2005). Prospective memory retrieval revisited. In N. Ohta, C. M. MacLeod & B. Uttl (Eds.), Dynamic cognitive processes (pp. 305-332). Tokyo: Springer. Graf, P., & Grodin, S. (2005). Time perception and time-based prospective memory. In J. Glicksohn & M. S. Myslobodsky (Eds.), Timing the future: The case for a time-based prospective memory (pp. 1-29). Singapore: World Scientific Publishing. Graf, P., & UttI, B. (2001). Prospective memory: A new focus for research. Consciousness and Cognition, 10, 437-450. Graf, P., UttI, B., & Tuokko, H. (1995). Color- and picture- word stroop tests: Performance changes in old age. Journal of Clinical and Experimental Neuropsychology, 17(3), 390- 415. Hannon, R., Adams, P., Harrington, S., Fries-Dias, C, Gibson, M. T. (1995). Effects of brain injury and age on prospective memory self-rating and performance. Rehabilitation Psychology, 40(4), 289-298. Henderson, J. G., & Pollard, C. A. (1988). Three types of obsessive compulsive disorder in a community sample. Journal of Clinical Psychology, 1988, 44(5), 747-752. Hermans, D., Martens, K., De Cort, K., Pieters, G., Eelen, P. (2003). Reality monitoring and metacognitive beliefs related to cognitive confidence in obsessive-compulsive disorder. Behavior Research and Therapy, 41, 383-401. Humphreys, M. S., & Revelle, W. (1984). Personality, motivation, and performance: A theory of the relationship between individual differences in information processing. Psychological Review, 91(2), 153-1 84. Kvavilashvili, L. (1987). Remembering intention as a distinct form of memory. British Journal of Psychology, 78, 507-518. Kvavilashvili, L, & Ellis, J. (1996). Varieties of intention: Some distinctions and classifications. In M. Brandimonte, G. 0. Einstein, & M. A. McDaniel (Eds.), Prospective memory: Theory and applications (pp. 23-51). New Jersey: Lawrence Erlbaum Associates. Macdonald, P. A., Antony, M. M., Macleod, C. M., & Richter, M. A. (1997). Memory and confidence in memory judgments among individuals with obsessive compulsive disorder and non-clinical controls. Behavior Research and Therapy, 35(6), 497-505. Marsh, R. L., Hicks, J. L., & Bink, M. L. (1998). Activation of completed, uncompleted, and partially completed intentions. Journal of Experimental Psychology: Learning Memory, and Cognition, 24(2), 350-361. 69 Meacham, J. A., & Dumitru, J. (1976). Prospective remembering and external-retrieval cues. JSAS Catalog of Selected Documents in Psychology, 6 no 65, 1-12. Muller, J., & Roberts, J. E. (2005). Memory and attention in obsessive-compulsive disorder: A review. Anxiety Disorders, 19, 1-28. Nigro, G., Senese, V. P., Natullo, 0., Sergi, I. (2002). Preliminary remarks on the type of task and delay in children’s prospective memory. Perceptual and Motor Skills, 95(2), 515- 519. Rasmussen, S. A., & Eisen, J. A. (1988). Clinical and epidemiological findings of significance to neuropharmacologic rituals in OCD. Psychopharmacological Bulletin, 24, 466-470. Reitan, R. M. (1992). Trail making test: Manual of administration and scoring. Tuscon, Arizona: Reitan Neuropsychology Laboratory. Sanavio, E. (1988). Obsessions and compulsions: The padua inventory. Behavior Research and Therapy, 26(2), 169-1 77. Searleman, A. (1996). Personality variables and prospective memory performance. In D. J. Herrmann, C. McEvoy, C. Hertzog, P. Hertel, & M. K. Johnson (Eds.), Basic and applied memory research: Practical applications Volume 2 (pp. 111-119). New Jersey: Lawrence Erlbaum Associates. Sher, K., Frost, R. 0., & Otto, R. (1983). Cognitive deficits in compulsive checkers: An exploratory study. Behavior Research and Therapy, 2 1(4), 357-363. Smith, G. V., Della Salla, 5, Logie, R.H, & Maylor, E.A.M. (2000). Prospective and retrospective memory in normal ageing and dementia: A questionnaire study. Memory, 8, 311-321. Stein, M. B., Forde, D. R., Anderson, G., & Walker, J. R. (1997). Obsessive-compulsive disorder in the community: An epidemiologic survey with clinical reappraisal. American Journal of Psychiatry, 154(8), 1120-1126. Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18, 643-662. Tallis, F., Pratt, P., Jamani, N. (1999). Obsessive compulsive disorder, checking, and non verbal memory: A neuropsychological investigation. Behaviour Research and Therapy, 37, 161-166. Tuna, S., Tekcan, A. I., & Topcuoglu, V. (2005). Memory and metamemory in obsessive compulsive disorder. Behavior Research and Therapy, 43, 15-27. 70 Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive Psychology, 5(2), 207-232. Wilson, B., Cockburn, J., & Baddeley, A. (1985). The Rivermead behavioral memory test manual. England: Thames Valley Test Company. Woods, C. M., Vevea, J. L., Chambless, D. L., & Ute, J. B. (2002). Are compulsive checkers impaired in memory? A meta-analytic review. Clinical Psychology: Science and Practice, 9(4), 353-366. 71 CHAPTER 3 SUB-CLINICAL CHECKING COMPULSIONS ARE RELATED TO IMPAIRED PROSPECTIVE MEMORY INDEPENDENTLY OF DEPRESSION, ANXIETY AND DISTRACTIBILITY5 Obsessive-Compulsive Disorder is a debilitating anxiety disorder that affects approximately 3% of the population (Kolada, Bland, & Newman, 1994; Stein, Forde, Anderson, & Walker, 1997). It is characterized by obsessions — defined as persistent and intrusive, thoughts, images or impulses that are unwanted and cause anxiety, and by compulsions or repetitive behaviours or mental acts that typically occur as an attempt to reduce the anxiety generated by the obsessions (American Psychiatric Association). OCD is a heterogeneous disorder with many different subtypes (Henderson & Pollard, 1988; Mataix-Colis, Rosario-Campos, Leckman, 2005). The most common subtype, characterized by checking compulsions, occurs in over 50% of OCD patients (Henderson & Pollard, 1988; Ramussen & Eisen 1988; Stein et al., 1997), with an additional 15% of the general population demonstrating sub-clinical checking compulsions (Stein et al., 1997). Individuals with checking compulsions have intrusive doubts that they failed to perform a task and/or failed to perform it properly, and they worry that their failure may cause harm to themselves or others. They then feel compelled to check whether the task was indeed performed and/or that it was performed properly (Muller & Roberts; 2005; Rachman & Shafran, 1998; Woods, Vevea, Chambless, & Bayen, 2002). Concretely, an individual with checking compulsions may have the reoccurring intrusive doubt that s/he forgot to unplug the iron, worry about the consequences of such a failure (e.g., starting a fire) and then feel compelled to return home to check the status of the iron at the expense of performing other daily activities. The mechanisms that underlie compulsive checking are not yet understood, but one view, known as the memory deficit theory, provides intuitive and promising leads (Sher, Frost, & Otto, 1983). According to this theory, checkers have a memory dysfunction which makes it difficult for them to remember whether they completed a task and/or completed it properly, and their inability to recall performing the task drives the compulsion to check. The results of empirical investigations generally support the memory deficit theory, revealing that checkers perform worse on many tests of episodic memory, especially when required to remember actions and other non-verbal materials (Woods et al., 2002). However, the evidence for a verbal A version of this chapter has been published. Cuttler, C. & Graf, P. (2008). Sub-clinical checking compulsions are related to impaired prospective memory independently of depression, anxiety and distractibility.Journal of Anxiety Disorders, 22(4), 642-654. 72 memory deficit is far less consistent (Rubenstein, Peynircioglu, Chambless, & Pigott, 1993; Sher et al., 1983, see Talus, 1995 for a review). One version of the memory deficit theory which illuminates the existing findings has emphasized its subjective rather than objective basis, suggesting that the compulsion to check may be driven by a lack of confidence in memory ability rather than by an objectively verifiable performance deficit (Macdonald, Antony, Macleod, & Richter, 1997; Sher et al., 1983). Consistent with this view, checkers have been found to exhibit subjective but not objective impairments on some tests that required, for example, reality monitoring or recollection of verbal materials (Hermans, Martens, De Cort, Pieters, Eelen, 2003; Macdonald et al., 1997; Sher et al., 1983). More recently, we have investigated the view that checking compulsions are a compensatory reaction to an objective and subjective impairment in prospective memory (Cuttler & Graf, 2007). Prospective memory is the ability to formulate intentions, plans and promises, to retain them, to recollect and to carry them out at the appropriate time or in the appropriate context (Einstein & McDaniel, 1996; Graf, 2005; Meacham & Dumitru, 1976). Everyday-life and clinical observations show that checkers’ compulsions are related to this future-oriented aspect of memory. Checkers tend to become concerned with tasks that require prospective memory, such as whether or not the iron, stove, lights, water or gas are turned off and whether doors or windows are properly shut and locked. We speculated that checkers’ increased experience with prospective memory failures undermines confidence in the ability to perform prospective memory tasks, thus leading them to worry and have intrusive doubts about similar failures. We believe the compulsion to check may be triggered by the perceived seriousness of the potential consequences of a prospective memory failure (e.g., a fire occurs because of the failure to unplug the iron). In support of this general proposal, our results showed that checking is related to worse performance on an episodic event-cued prospective memory task which required participants to request the return of a personal belonging. In addition, we found that on two different questionnaires, higher checking scores were related to reporting experiencing more of a variety of prospective memory failures in everyday life (Cuttler & Graf, 2007). The main objective of the present study was to examine more closely the nature of the link between checking compulsions and prospective memory. Previous research has shown that checking compulsions typically do not occur in isolation. Rather, individuals with checking compulsions also tend to show elevated levels of depression, as well as elevated anxiety scores (Frost, Sher, & Geen, 1986; Hasler et al., 2005), and in turn, both of these attributes are associated with reduced performance on retrospective memory tests (Brand & Jolles, 1987; 73 Darke, 1988; Golinkoff, & Sweeny, 1989; Hill, & Vandervoort, 1992) and on prospective memory tasks (Harris, & Menzies, 1999; Rude, Hertel, Jarrold, Covich, & Hedlund, 1999). In view of this evidence, the specific objective of the present study was to examine whether, or to what extent, there exists a specific direct relationship between checking compulsions and prospective memory, as compared to a relationship that reflects the comorbid psychopathology of checkers. Previous research also points to the prevalence of executive dysfunctions in COD (Tallis, 1995). Checking has been associated with higher total scores on the Cognitive Failures Questionnaire (CFQ) (Sher, Mann, & Frost, 1984), an inventory which contains a subscale for measuring distractibility (Wallace, Kass, & Stanny, 2002). It seems likely that individuals who are distractible would experience more failures on prospective memory tasks (because such tasks depend on efficient executive functions or attention management skills), and for this reason, a second objective of the present study was to examine the role of distractibility in the link between checking compulsions and prospective memory. Study Overview Each participant completed a battery of instruments including one that focuses on obsessive-compulsive behaviour and contains a detailed subscale for measuring checking behaviour. We used participants’ checking subscale scores to classify them either as high checkers or low checkers. The battery also included a depression inventory, a state and trait anxiety inventory, and a questionnaire subscale for assessing distractibility. Previous research has documented increased levels of depression, state and trait anxiety in checkers (Frost et al., 1986; Hasler et al., 2005), and it has shown their association with impairments in prospective memory (Harris, & Menzies, 1999; Rude et al., 1999). We expected to replicate these findings. More importantly however, we hypothesized an additional direct link between checking compulsions and prospective memory that is independent of this common comorbid psychopathology and OCD associated attention management deficits. In order to assess subjective prospective memory, participants completed two inventories that required rating the frequency of experiencing various kinds of memory failures, including episodic prospective memory failures, habitual prospective memory failures, internally cued prospective memory failures and retrospective memory failures. Episodic prospective memory tasks are those where a to-be-performed plan is not in conscious awareness during the retention interval and needs to be executed only once (e.g., making an important phone call after dinner) (Graf, 2005; Kvavilashvili & Ellis, 1996; Meacham & Dumitru, 1976). Habitual prospective memory tasks require repeated execution of the same plan (e.g., removing the keys 74 from the car before closing and locking the doors) (Graf & Uttl, 2001; Hannon, Adams, Harrington, Fries-Dias, & Gibson, 1995; Kvavilashvili & Ellis, 1996; Meacham & Dumitru, 1976). Internally cued prospective memory tasks are those where no salient external cue is provided for prompting the retrieval of the previously formed plan (e.g., cashing a paycheck before running out of money) (Hannon et al., 1995). Retrospective memory tasks are those where past events or previously learned information needs to be recalled. In line with our previous findings (Cuttler & Graf, 2007), we expected checking to be associated with reporting more experiences with each variety of memory failure. Finally, the assessment battery also included one episodic prospective memory task which required participants to request the return of a personal belonging upon the occurrence of a specified spoken cue. As a further index of subjective prospective memory we also asked participants to rate how confident they felt about succeeding on this episodic prospective memory task. In line with our previous research (Cuttler & Graf, 2007), we expected to find a positive relationship between checking and failure on the prospective memory task. Method Participants and Design Sub-clinical checking compulsions are fairly common among otherwise healthy and normal-functioning individuals (Sher et al., 1983; Stein et al., 1997), and thus, we were able to use undergraduate students for the present study. We proceeded on the assumption that if a study with student participants shows evidence in support of our hypotheses, this evidence most likely would be even stronger in individuals with fully developed clinical checking compulsions. We recruited 126 undergraduate students from the University of British Columbia Department of Psychology subject pool. In order to ensure our sample included individuals with a wide range of checking behaviours, we used two types of advertisements, one that called specifically for participants with checking tendencies and another that made no mention of checking compulsions. Participants ranged from 17 to 44 years of age with a mean of 20.43 years (SD 4.03). Participants’ postsecondary education ranged from 1 to 8 years with a mean of 2.25 years (SD = 1.14). Forty-two of the participants were male and 84 were female. The overall design of the study was quasi-experimental. We used participants’ checking scores, assessed by means of the Padua Inventory described in the Instruments section, to assign them to one of two groups: High checkers or low checkers. Specifically, participants with checking scores between 11 and 30 were assigned to the high checkers group and participants with checking scores between 0 and 10 were assigned to the low checkers group. These cut-off 75 points yielded approximately the same number of participants per group: 64 high checkers and 62 low checkers. The mean Padua Inventory checking scores were 17.97 and 4.90, respectively, for the high and low checkers6.The two groups did not differ in age, t (123) = - 1.96, p> .05, or in gender, x2 (1) = .02, p> .05, but they did differ in years of postsecondary education, t (123) = -2.17, p < .05. High checkers had on average of 0.4 fewer years of post- secondary education7. We conducted this research with the approval of the University of British Columbia Behavioural Ethical Review Board. Completing all components of the study required approximately one hour, and participants received course credit in return for their participation. Instruments In the course of the experiment each participant was administered several questionnaires and a prospective memory task. Each of these study components is briefly described below. Padua Inventory. The Padua Inventory is a standardized self-report inventory developed by Sanavio (1988). The inventory contains 60 statements describing common obsessive-compulsive behaviours (e.g., I return home to check doors, windows, drawers, etc. to make sure they are properly shut). For each item participants use a 5-point scale to indicate the degree of disturbance the behaviour creates, with scale points marked: 0 = none at all, I = a little, 2 = quite a lot, 3 = a lot and 4 = very much. The inventory contains four subscales each of which measures different obsessive-compulsive concerns: Checking, impaired control of mental activities, contamination and worries about losing control over motor behaviours (Sanavio, 1988). The questionnaire is a reliable measure of obsessive-compulsive behaviour, as evidenced by high test-retest reliability (.80) and a high reliability coefficient (.94) (Sanavio, 1988). We computed participants’ checking scores by summing the ratings they gave on the items contained on the checking subscale of the inventory. Possible scores on the checking subscale range from 0 to 32, with higher scores indicating a higher level of checking compulsions. Prospective and Retrospective Memory Questionnaire (PRMQ). The PRMQ is a standardized self-report inventory developed by Smith, Della Sala, Logie and Maylor (2000). 6 The high checkers’ mean Padua Inventory checking subscale score is comparable to those reported in previous research with clinical checkers (Hermans et al., 2003; Talus, Pratt, & Jamani, 1999). Because of this difference in education, we performed regression analyses using education to predict each of the dependent variables explored in the study. Significant relationships were revealed only between education and the episodic prospective memory and internally cued prospective memory subscales of the PMQ. Hierarchical regression analyses revealed that the relationships found between checking and ratings on these subscales are not a result of differences in education. 76 The scale contains 16 questions concerning various memory failures (e.g., Do you forget appointments if you are not prompted by someone else or by a reminder such as a calendar or diary?). For each item, participants indicate the frequency of experiencing the failure on a 5- point scale, with scale points marked: 1 = never, 2 a little, 3 quite a lot, 4 = a lot and 5 = very often. The scale contains two subscales, one focusing on prospective memory and the other on retrospective memory. Both subscales yield high reliability coefficients (prospective memory = .84, retrospective memory = .80) (Crawford, Smith, Maylor, Della Sala, & Logie, 2003). We computed participants’ prospective memory and retrospective memory scores by averaging their ratings on the prospective memory and the retrospective memory items, respectively. Possible scores on each subscale range from 1 to 5, with higher scores indicating a greater frequency of memory failures. Prospective Memory Questionnaire (PMQ). The PMQ is a standardized self-report inventory developed by Hannon et al. (1995). The scale contains 52 statements concerning various prospective memory failures (e.g., I forgot to lock up my house, bike or car) and prospective memory aiding techniques (e.g., I write myself reminder notes). Typically participants have to rate how commonly they experience each failure in a given time period. However, we used a slightly modified version of this inventory which required participants to rate on a 5-point scale the frequency of experiencing each failure. The rating scale points were marked: I never, 2 = a little, 3 = quite a lot, 4 = a lot and 5 = very often. The questionnaire contains four subscales, three of which concern failures connected with different prospective memory tasks, specifically, episodic tasks, habitual tasks and internally cued tasks, and the fourth measures the frequency of using various prospective memory aiding strategies. Each subscale is reliable, with alpha coefficients ranging from .78 to .90, and test-retest reliabilities ranging from .64 to .88 (Hannon et al., 1995). We computed participants’ scores on each subscale by averaging the ratings they gave on the items contained within each subscale. Possible scores on each subscale range from 1 to 5, with higher scores indicating a greater frequency of memory failures or use of memory aiding strategies. Cognitive Failures Questionnaire (CFQ). The CFQ is a standardized self-report inventory developed by Broadbent, Cooper, FitzGerald and Parkes (1982). The inventory is used for assessing the frequency of experiencing everyday cognitive slips or errors, in other words, the frequency with which individuals make mistakes on simple tasks. The CFQ contains 25 questions concerning various action slips or errors (e.g., Do you start doing one thing at home and get distracted into doing something else unintentionally?). For each item, participants indicate the frequency of experiencing the failure on a 5-point scale, with scale points marked: 1 = never, 2 = very rarely, 3 = occasionally, 4 = quite often and 5 = very often. The questionnaire 77 contains four subscales: Distractibility, memory, blunders and memory for names (Wallace et al., 2002). Each subscale is reliable, with reliability coefficients ranging from .76 to .86 (Wallace et al., 2002). We computed participants’ scores on the distractibility subscale by summing the ratings they gave on the relevant items. Possible scores on the distractibility subscale range from 0 to 45, with higher scores indicating a greater degree of distractibility. Beck Depression Inventory (BDI). The BDI is a self-report inventory for measuring depression (Beck, Ward, Mendelson, Mock, & Erbaugh, 1961; Beck, 1987). The inventory contains 21 groups of four statements related to symptoms of depression (e.g., 0 = I don’t feel disappointed in myself, I = I am disappointed in myself, 2 = I am disgusted with myself, 3 = hate myself). Participants are instructed to circle the number corresponding to the one statement in each group which best describes how they have been feeling during the past week. The inventory is reliable with split-half reliabilities ranging from .78 to .93 and test-retest reliabilities ranging from .48 for psychiatric patients to .74 for undergraduate students (Beck, 1987). We computed participants’ depression scores by summing up all of the items they circled in the inventory. Possible scores range from 0 to 63, with higher scores reflecting more severe depression. State-Trait Anxiety Inventory (STAI). The STAt is an inventory for measuring state and trait anxiety (Spielberger, Gorsuch, Lushene, Vagg, & Jacobs, 1983). The inventory contains two parts, one to assess state anxiety or “feelings of apprehension, tension, nervousness and worry” (Spielberger et al., 1983, p.2), and another to assess trait anxiety or “relatively stable individual differences in anxiety-proneness” (Spielberger et al., 1983, p.1). The state anxiety section contains 20 statements that assess how individuals feel at the moment of completing the inventory. The trait anxiety section contains 20 statements that assess how individuals generally feel. For each item participants use a 4-point scale to indicate how they feel, with scale points marked: I = not at all, 2 = somewhat, 3 = moderately so and 4 = very much so. Both parts of the inventory yield high reliability coefficients for college students (state anxiety .92, trait anxiety = .91) (Spielberger et al., 1983). Test-retest reliability in college students is high on the trait anxiety section (.73 to .86) (Spielberger et al., 1983). As expected, test-retest reliability in college students is much lower on the state anxiety section (.16 to .54) (Spielberger et al., 1983). We computed state and trait anxiety scores according to Spielberger et al. (1983). Possible scores on each section range from 20 to 80, with higher scores reflecting more anxiety. Personal Belonging Task. We used a slightly modified version of the standardized belonging task (Wilson, Cockburn, & Baddeley, 1985) for assessing episodic prospective memory performance. Participants were told we were interested in their ability to remember to 78 do things at a later time, and that we needed one of their personal belongings for this purpose. If the participant was wearing a watch then this item was taken; otherwise a cell phone or something of similar value was requested. We placed the personal belonging in a drawer and instructed participants to ask for its return upon being told later in the testing session: “we are now finished with all of the tests.” Participants were further informed that they would receive this prompt immediately before the debriefing. After ensuring that participants understood the instructions, we asked them to rate their confidence in the ability to remember to request the return of their personal belonging at the appropriate moment. We used a 5-point scale for making these confidence ratings, where I indicated extremely unconfident and 5 indicated extremely confident. Those participants who requested the return of their belonging at the appropriate time were awarded a score of 1, indicating success; otherwise they received a score of 0. Lexical Decision Task. Participants also completed a lexical decision task. They were presented with 312 letter-strings on a computer monitor, one at a time, and for each made a decision about whether the letter-string was a word (e.g., famous) or a non-word (e.g., safoum) as quickly and as accurately as possible. The primary purpose of the lexical decision task was to distract participants and to occupy them during the retention interval required for the Personal Belonging prospective memory task, and thus, we do not report performance on this task. Procedure Participants were tested one at a time in a quiet room. Upon obtaining their written informed consent, we collected basic demographic information. We then assigned the event- cued Personal Belonging prospective memory task, and participants’ rated their confidence in being able to perform this task as instructed. Participants then completed the questionnaires, according to their own pace, in the following order: the Padua Inventory, the PRMQ, the PMQ, the CFQ, the BDI and the STAI. Following the questionnaires, participants completed the Lexical Decision task, at the end of which the experimenter cued the Personal Belonging task with the phrase: “we are now finished with all of the tests.” To create a reasonable time-window for responding to this cue, the experimenter took about 10 seconds to tidy up and put the testing materials away and then informed participants that they were going to be debriefed. After this cue, participants received a full debriefing and their research participation credit. Results Data Preparation All data were double checked for recording and transcription errors. Data accuracy was greater than 99%. Ten missing values on the questionnaires (i.e., <.001%) were replaced with 79 their corresponding mean responses. One participant failed to complete the state anxiety portion of the STAI and another failed to complete the trait anxiety portion. These participants’ data were not used in any analyses involving the relevant scores. We examined all of the questionnaire data and confidence ratings for univariate outliers, defined as scores falling more than three standard deviations away from their respective means. We discovered seven outliers in the data (i.e., < .001%), and they were replaced with the nearest non-outlying value, specifically, a score either -3 or +3 standard deviations away from the corresponding mean. Subjective Prospective Memory Performance Do checkers have faith in their prospective memory skills, or do they believe their prospective memory is impaired? To answer these questions, we examined the confidence ratings participants produced in connection with the Personal Belonging task, as well as the ratings they produced on the PMQ and the PRMQ. On the Personal Belonging task, participants rated, on a 5-point scale, how confident they were about requesting the return of their personal belonging at the appropriate moment. Consistent with Cuttler and Graf (2007), the groups’ confidence ratings were similar, with high checkers giving a mean rating of 3.57 and low checkers giving a mean rating of 3.79. An independent samples t-test of these ratings was not significant, t (124) = 1.42, p> .05. The main findings from the PMQ are depicted in Figure 3.1. We explored these data with an ANOVA that had the three different types of prospective memory tasks (episodic, habitual, internally cued, respectively, depicted by the three bars on the left side of the figure) as a within subject factor, and Group (high checkers, low checkers) as a between subjects factor. The results showed significant main effects due to Group, F (1, 124) = 12.15, MSE = .52, p <.005, and due to the type of prospective memory task, F (2, 248) = 188.09, MSE = .10, p < .001. The interaction between these factors did not reach significance, F (2, 248) = 2.38, MSE = .10, p> .05. Follow-up independent samples t-tests showed that the two groups gave significantly different ratings on the scales for episodic prospective memory, t (124) = -2.30, p < .05, habitual prospective memory, t (124) = -3.49, p < .005, and internally cued prospective memory, t (124) = -3.72, p <.001. The rightmost set of bars in Figure 3.1 shows participants’ self-rated use of prospective memory aiding strategies. An independent samples t-test revealed that high checkers reported using more prospective memory aiding strategies than low checkers, t(124) = -4.76, p < .001. 80 Figure 3.1. Self-Rated Frequency of Prospective Memory Failures and Use of Prospective Memory Aiding Strategies by the Two Checking Groups. D Low Checkers . I • High Checkers 3 ‘H 2- C Di ___ , __ 0- In a final ANOVA, we examined participants’ ratings on the PRMQ. The first analysis had Memory Type (prospective memory, retrospective memory) as a within subject factor and Group (high checkers, low checkers) as a between subjects factor. The results, highlighted by Figure 3.2, revealed a significant effect of Group, F (1, 124) = 5.25, MSE = .78, p < .05, and of Memory Type, F (1, 124) = 52.78, MSE = .10, p < .05. The interaction between these two factors was not significant, F (1, 124) = .01, MSE = .107, p> .05. Follow-up independent samples t-tests showed that the two groups differed significantly in their experiences of prospective memory failures, t (124) = -2.07, p < .05, as well as in their experiences of retrospective memory failures, t (124) = -2.20, p < .05. Figure 32. Self-Rated Frequency of Prospective and Retrospective Memory Failures by the Two Checking Groups. Low Checkers • High Checkers 2 Episodic Habitual Internally Cued Prospective Prospective Prospective Prospective Memory Aiding Memory Memory Memory Strategies -r Prospective Memory Retrospective Memory 81 To complement these findings of group differences on the PMQ and PRMQ, we also explored participants’ ratings by means of regression analyses. The results, summarized in the top row of Table 3.1, showed that checking was a significant predictor of each set of ratings. The Influence of Depression, Anxiety and Distractibility. The main objective of the present study was to examine whether, or to what extent, there exists a specific direct relationship between checking compulsions and prospective memory, as compared to a relationship that reflects the comorbid psychopathology of checkers and/or distractibility. As a first step toward this goal, we computed the zero-order correlations — listed in Table 3.2— among checking, depression, state and trait anxiety and distractibility. Consistent with previous research, the correlations among checking, depression, state anxiety and trait anxiety were significant (Frost et al., 1986; Hasler, et al., 2005; Spielberger et al., 1983). Moreover, the analysis revealed significant correlations between distractibility and each of checking, depression, state anxiety and trait anxiety. We next examined the correlations between each of checking, distractibility, depression, state and trait anxiety and the ratings on the PMQ and PRMQ. As shown in Table 3.2, the analysis revealed that checking, distractibility, depression, state anxiety and trait anxiety were significantly correlated with ratings on all of the PMQ and PRMQ subscales, with one exception; depression was unrelated to the self-reported use of prospective memory aiding strategies. The pattern of correlations is consistent with the possibility that the link between checking and prospective memory reported by Cuttler and Graf (2007) is secondary to checkers’ comorbid psychopathology and/or distractibility. In order to examine this possibility more directly, we conducted the series of hierarchical linear regression analyses that are summarized in the middle part of Table 3.1. Each analysis focused on one of the PMQ or PRMQ questionnaire subscales, and each involved two distinct steps. In Step 1, we entered either the depression scores from the BDI, the state and trait anxiety scores from the STAI or the distractibility scores from the CFQ, and in Step 2, we entered the checking scores from the Padua Inventory. Consistent with the results of the correlation analyses reported in Table 3.2, the Step 1 results in Table 3.1 show that depression, anxiety and distractibility were significant predictors of self-rated prospective memory task performance. There were two exceptions which occurred in connection with the use of prospective memory aiding strategies where neither the depression nor the anxiety scores were significant predictors. The Step 1 results in Table 3.1 also reveal that on average, the depression and anxiety scores had less predictive power (indicated by smaller R2 values) than did the checking scores (when they were entered in Step 1; see the top row of Table 3.1). By contrast, the predictive power of distractibility was at least twice as large as the predictive power of the checking scores. 82 T ab le 31 . R es ul ts o f a Se ri es o f L in ea r a n d H ie ra rc hi ca l L in ea r R eg re ss io n A na ly se s U si ng C he ck in g, D ep re ss io n, St at e a n d Tr ai t A nx ie ty a n d D is tra ct ib ili ty to Pr ed ic t R at in gs o n th e PM Q a n d PR M Q Su bs ca le s. PM Q PR M Q Ep iso di c H ab itu al Pr os pe ct iv e In te rn al ly C ue d M em or y A id in g Pr os pe ct iv e R et ro sp ec tiv e Pr os pe ct iv e Pr os pe ct iv e Pr os pe ct iv e M em or y M em or y M em or y M em or y St ra te gi es R2 F R2A F R F R ’ F R2 F R2 El i St ep I Ch ec ki ng . 14 19 .44 ** . 13 18 .57 ** . 21 32 .1 7* * . 23 36 .3 8* * . 13 18 .4 0* * . 17 25 .0 4* * St ep I D ep re ss io n . 08 11 .4 3* . 09 11 .8 0* * . 19 29 .0 0* * . 02 2. 19 . 10 14 .2 3* * . 14 19 .3 7* * St ep 2 Ch ec ki ng . 08 11 .4 2* * . 07 10 .6 2* * . 10 16 .82 ** . 21 33 .8 9* * . 07 9. 78 * . 09 13 .5 0* * St ep I St at e& Tr ai tA nx ie ty . 11 74 5* * . 09 6. 22 * . 18 12 .8 8* * . 03 1. 76 . 09 5. 98 * . 18 13 .1 7* * St ep 2 Ch ec ki ng . 07 10 .2 2* . 08 11 .6 2* * . 09 15 .15 ** . 20 30 .3 9* * . 06 9. 29 * . 06 10 .1 4* St ep I D ist ra ct ib ili ty . 34 62 .9 1* * . 29 50 .2 0* * . 50 12 2. 62 ** . 08 10 .9 9* * . 36 68 .5 9* * . 32 56 .9 9* * St ep 2 Ch ec ki ng . 02 39 5* . 02 4. 31 * . 03 8. 64 * . 16 25 .0 3* * . 02 3.1 5* 1 . 04 7. 46 * St ep I D ep re ss io n, St at e & T ra itA nx ie ty an d . 35 15 .95 ** . 31 13 .2 6* * . 53 33 39 ** . 08 2. 44 *1 . 36 16 .9 2* * . 33 14 .8 3* * D ist ra ct ib ili ty St ep 2 Ch ec ki ng . 02 4. 60 * . 03 5. 76 * . 03 7. 63 * . 17 25 .7 6* * . 02 3. 60 *1 . 03 5. 10 * * 1 in di ca te s th e m o de l is si gn ifi ca nt w ith p . 05 (on e-t ail ed ), * in di ca te s th e m o de l is si gn ifi ca nt w ith p • Q5 ,* * in di ca te s th e m o de l is si gn ifi ca nt w ith p . 00 1. Ta bl e 3. 2. Ze ro -O rd er Co rre la tio ns A m on g Ch ec ki ng , D ep re ss io n, St at e A nx ie ty ,T ra it A nx ie ty , D ist ra ct ib ili ty an d R at in gs on th e PM Q an d PR M Q Su bs ca le s V ar ia bl e 2 3 4 5 6 7 8 9 10 11 1. Ch ec ki ng . 38 ** 33 ** 44 ** . 41 ** 37 ** 36 ** 45 ** 48 ** 36 ** . 41 ** 2. D ep re ss io n . 65 ** 73 ** 44 ** . 29 ** . 30 ** 44 ** . 13 . 32 ** 37 ** 3. St at e A nx ie ty . 70 ** . 41 ** . 29 ** . 30 ** . 29 ** . 15 *1 . 25 * 33 ** 4. Tr ai tA nx ie ty . 52 ** . 32 ** . 25 ** . 42 ** . 16 *1 . 29 ** . 42 ** 5. D ist ra ct ib ili ty . 58 ** 54 ** . 71 ** 29 ** . 60 ** 56 ** 6. PM Q — Ep iso di c Pr os pe ct iv e M em or y . 67 ** . 72 ** . 16 *1 76 ** . 68 ** 7. PM Q — H ab itu al Pr os pe ct iv e M em or y . 56 ** . 14 . 60 ** . 56 ** 8. PM Q — In te rn al ly Cu ed Pr os pe ct iv e M em or y . 24 * 74 ** . 68 ** 9. PM Q — Pr os pe ct iv e M em or y Ai di ng St ra te gi es . 29 ** . 18 * 10 . P RM Q — Pr os pe ct iv e M em or y 11 . PR M Q — R et ro sp ec tiv e M em or y * 1 in di ca te s p . 05 (on e-t ail ed ), * in di ca te s p . 05 ,* * in di ca te s p . 00 1 The Step 2 results in the middle part of Table 3.1 show that checking remained a significant predictor of subjective prospective memory task performance even when it was entered after the depression, anxiety or distractibility scores. Equally important, however, the R2 values associated with checking were consistently smaller when this variable was entered in Step 2, that is, after depression, anxiety or distractibility. The reduction in the R2 values was about the same when checking was entered after depression and anxiety, but it was substantially larger when it was entered after distractibility. This collection of findings indicates that depression, anxiety and distractibility associated with checking compulsions are important contributors to the relationship between checking compulsions and self-rated prospective memory performance. However, the findings also show that taken independently of each other, these variables do not exhaust the predictive power of the checking scores. We also explored whether checking scores would be able to predict self-rated prospective memory performance even after removal of the influence due to the combination of depression, anxiety and distractibility. For this purpose, we entered the depression, anxiety and distractibility scores in Step I of a hierarchical linear regression analysis, followed by checking scores in Step 2. The results, reported in the bottom row of Table 3.1, show that although the R2 values associated with checking were consistently smaller when this variable was entered after depression, anxiety and distractibility, checking remained a significant predictor of the ratings on each of the questionnaire subscales. Objective Prospective Memory Task Performance Our results show that checkers report experiencing more prospective memory failures in everyday life, but do they actually perform worse on real prospective memory tasks? To test this possibility, the present study included one objective prospective memory task that required participants to request the return of a personal belonging when they were told: “we are now finished with all of the tests.” Participants’ performance on this task was consistent with findings reported by Cuttler and Graf (2007), revealing that 48.4% of the high checkers and 72.6% of low checkers requested the return of their personal belonging. A chi-square analysis showed a significant difference between the groups, x2 (1) 7.67, p < .05. This finding was supported by the results of a logistic regression analysis in which checking scores from the Padua Inventory were found to significantly predict performance on the Personal Belonging prospective memory task, x2 (1) = 4.90, p < .05 The Influence of Depression, Anxiety and Distractibility. In order to explore whether the association between checking and impaired prospective memory task performance is a result of comorbid depression, state and trait anxiety or distractibility we conducted a series of three logistic regression analyses using participants’ depression scores, state and trait anxiety 85 scores and distractibility scores to predict performance on the Personal Belonging prospective memory task. None of these models were significant, x2 (1) = 2.16, p> .05, x2 (2) = 3.41, p> .05 and x2 (1) = .97, p> .05, respectively for depression scores, state and trait anxiety scores and distractibility scores. The finding that depression, state and trait anxiety and distractibility scores offer no power to predict performance on the Personal Belonging Task suggests that the relationship between objective prospective memory performance and checking is not a result of increased levels of depression, anxiety or distractibility associated with checking. Discussion In a recent paper (Cuttler & Graf, 2007) we expanded the memory deficit theory to the domain of prospective memory by demonstrating links between checking compulsions and subjective and objective impairments in prospective memory. For the present study we set out to rule out competing explanations for these links. First, we wished to evaluate the relationships between checking and prospective memory after controlling for the elevations in depression and anxiety which are often associated with checking compulsions. Second, recognizing the possibility that problems with attention management — specifically with distractibility — may be responsible for the impairments in prospective memory that we previously discovered, we also wished to explore whether the relationships between checking and impaired prospective memory are secondary to increased levels of distractibility. The results from the present study were consistent with those from our previous study (Cuttler & Graf, 2007), showing that checking is associated with objective and subjective impairments in prospective memory. Specifically, we replicated our findings that checking is associated with worse performance on an episodic prospective memory task, requiring participants to request the return of a personal belonging as well as with reporting more of various types of everyday life prospective memory failures. As expected and consistent with previous research we found that depression and anxiety are associated with checking compulsions (Frost et al., 1986; Hasler, et al., 2005). Moreover, we showed that depression and anxiety are associated with self-rated prospective memory performance in everyday life. While controlling for the elevations in depression and anxiety associated with checking reduced the predictive power of checking, checking remained a significant predictor of subjective prospective memory impairments. This outcome suggests that the comorbid psychopathology associated with checking compulsions contributes to, but does not completely explain, the link between checking compulsions and subjective prospective memory impairments. 86 In line with one of the intuitions that motivated the presented study, we also found that checking compulsions are associated with increased levels of distractibility. Moreover, we demonstrated strong relationships between distractibility and self-rated prospective memory performance. This finding suggests that attention management skills are a key ingredient to successful prospective memory performance in everyday life, a factor that has been largely overlooked in previous prospective memory research. Controlling for the increased levels of distractibility associated with checking led to a sizable reduction in the ability of checking to predict self-rated prospective memory; nevertheless, checking remained a significant predictor. This finding suggests that although distractibility plays a crucial role in checkers’ subjective prospective memory impairment, it does not completely account for the link between checking and impaired prospective memory. Future researchers should measure attention management skills more broadly than distractibility to explore whether some of the remaining variance can be explained by other aspects of attention management. Future research could also focus on differences in rehearsal and other individual differences related to prospective memory performance, such as personality, as possible mediators of the link between checking and impaired prospective memory. Some of the strongest evidence for a link between checking compulsions and subjective prospective memory comes from the data on the use of prospective memory aiding strategies. The regression analyses showed that checking is associated with using more prospective memory aiding strategies, and controlling for depression, anxiety and distractibility had little influence on this relationship. This finding implies that checkers may be aware of their prospective memory deficit and are attempting to compensate for it through the use of strategies for improving prospective memory performance. Similarly, their increased use of memory aiding strategies may reflect checkers’ heightened concerns with having prospective memory failures and with the potential consequences of these failures. Despite their increased use of strategies, however, checkers continue to report worse performance on prospective memory tasks, possibly because checkers are not using strategies in an effective manner. Formal training on the use of prospective memory aiding strategies may help in treating checking compulsions. Future research might focus on the utility of training in prospective memory strategies as well as on confidence boosting exercises in the treatment of checking compulsions. In contrast to our other measures of subjective prospective memory, depression was unrelated to the self-reported use of prospective memory aiding strategies. The lack of relationship between depression and use of prospective memory aiding strategies may reflect the lack of motivation and pessimistic outlook associated with depression. Planning is an 87 effortful process and depressed individuals may lack the motivation to take the extra steps required for improving prospective memory. They may also believe that planning is futile, that they are bound to fail regardless of any attempts to succeed. While the relationship between distractibility and self-reported use of prospective memory aiding strategies was significant in our results, it was relatively small in comparison to the other subjective measures of prospective memory. Moreover, when the influence of distractibility was removed in the first step of the regression analysis, checking retained most of its predictive power. It may be that distractibility relates more to the retrieval phase of prospective memory, to the ability to notice cues and recognize their significance to a previously formed plan, than to the encoding phase and planning. It may also indicate that distractible individuals do not recognize their prospective memory deficit or that they are easily distracted from arranging their environment in such a way to maximize the probability of success. Our failure to find significant associations between objective prospective memory task performance and each of depression, anxiety and distractibility was unexpected, but it suggests that the relationship between checking compulsions and objective prospective memory task performance is not confounded by these factors. Rather there appears to be a direct link between checking compulsions and objective prospective memory task performance that is independent of depression, anxiety and distractibility. However, it is also possible that the failure to find significant relationships between objective prospective memory task performance and each of depression, anxiety and distractibility reflects the reduction in power associated with using a dichotomous as opposed to continuous outcome variable (Glass & Hopkins, 1996). Future research should focus on extending the findings from the present study to a sample of clinical compulsive checkers. Although our sample of high checkers had significantly higher depression and anxiety scores than our sample of low checkers, their scores are likely not as high as would be expected in a sample of clinical compulsive checkers. And while the range of comorbid psychopathology present in our sub-clinical sample was sufficient to reveal associations with the more powerful continuous subjective measures of prospective memory, this possible reduced range may have contributed to the failure to find a relationship between comorbid psychopathology and the dichotomous prospective memory task scores. Finally, our failure to find a significant relationship between distractibility and objective prospective memory task performance may be linked to the relatively distraction-free laboratory environment in which participants were tested. In contrast, the questionnaires which we used to assess subjective prospective memory were designed to assess prospective memory performance in the real world where we are typically faced with multiple sources of distraction. 88 Conclusions Extending the memory deficit theory of compulsive checking to the domain of prospective memory overcomes some of the limitations of the traditional memory deficit theories. One limitation of the traditional memory deficit theories is the failure to find correlations between checking and retrospective memory impairments. This failure “to find any systematic relationship between memory performance and checking is a considerable embarrassment to mnnestic deficit accounts of compulsive checking... suggest[ing] that memory impairment has little aetiological significance with respect to checking behaviour” (Talus et al., 1999, p 165). However, by reframing the memory deficit theories in the domain of prospective memory, we were able to demonstrate significant correlations between checking compulsions and both objective and subjective prospective memory performance. Another problem with the traditional memory deficit theories which is overcome by focusing on prospective memory is the ability to explain why checkers become concerned that they may have failed to perform a task and/or failed to perform it properly. That is, the traditional theories do not explain the intrusive doubts regarding the failure to perform a task. They simply suggest that after the question of whether or not they correctly completed a task intrudes upon their thoughts, checkers have difficulty remembering performing the task or they do not trust their memory of performing the task. However, the more fundamental issue seems to be, why checkers question whether or not they performed some task in the first place? Why do they not trust their ability to perform these banal tasks? Extending the memory deficit theories to the domain of prospective memory helps to explain this process. We suggest that checkers’ increased experience with prospective memory failures leads them to distrust their ability to perform certain tasks and therefore the intrusive doubts that a specific task was not completed. When the individual attempts to recall retrospectively performing the task they cannot or do not trust the memory trace, ultimately resulting in the compulsion to check. In this way prospective and retrospective memory deficits likely work together to produce and maintain checking corn pulsions. 89 References American Psychiatric Association (2000). Diagnostic and Statistical Manual of Mental Disorders (4th Edition). Washington, DC: American Psychiatric Association. Beck, A. T. (1987). Beck depression inventory (BDI). In K. Corcoran, & J. Fisher (Eds.), Measures for clinical practice: A sourcebook (pp. 107-110). New York: The Free Press. Beck, A.T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961) An inventory for measuring depression. Archives of General Psychiatry 4, 561-571. Brand, N., & Jolles, J. (1987). Information processing in depression and anxiety. Psychological Medicine, 17(1), 145-153. Broadbent, D. E., Cooper, P. F., FitzGerald, P., & Parkes, K. R. (1982). The cognitive failures questionnaire and its correlates. British Journal of Clinical Psychology, 21, 1-16. Crawford, J. R., Smith, G., Maylor, E. A., Della Sala, S., Logie, R. H. (2003). The prospective and retrospective memory questionnaire (PRMQ): Normative data and latent structure in a large non-clinical sample. Memory, 11(3), 261-275. Cuttler, C., & Graf, P. (2007) Sub-clinical compulsive checkers’ prospective memory is impaired. Journal of Anxiety Disorders, 21(3), 338-352. Darke, S. (1988). Anxiety and working memory capacity, Cognition and Emotion, 2(2), 145-1 54. Einstein, G. 0., & McDaniel, M. A. (1996). Retrieval processes in prospective memory: Theoretical approaches and some new empirical findings. In M. Brandimonte, G. 0. Einstein, & M. A. McDaniel (Eds.) Prospective memory: Theory and applications (pp. 115-141). New Jersey: Lawrence Erlbaum Associates. Frost, R. 0., Sher, K. J., & Geen, T. (1986). Psychopathology and personality characteristics of nonclinical compulisive checkers, Behaviour Research and Therapy, 24(2), 133-143. Glass, G. V., & Hopkins, K. D. (1996). Statistical methods in education and psychology. (gth ed). Boston: Allyn and Bacon. Golinkoff, M., & Sweeny, J. A. (1989). Cognitive impairments in depression. Journal of Affective Disorders, 17(2), 105-112. Graf, P. (2005). Prospective memory retrieval revisited. In N. Ohta, C. M. MacLeod, & B. UttI (Eds.), Dynamic cognitive processes (pp. 305-332). Tokyo: Springer. Graf, P., & UttI, B. (2001). Prospective memory: A new focus for research. Consciousness and Cognition, 10, 437-450. 90 Hannon, R., Adams, P., Harrington, S., Fries-Dias, C, Gibson, M. T. (1995). Effects of brain injury and age on prospective memory self-rating and performance. Rehabilitation Psychology, 40(4), 289-298. Harris, L. M., & Menzies, R. G. (1999). Mood and prospective memory. Memory, 7(1), 117-127. Hasler, G., LaSalle-Ricci, V. H., Ronquillo, J. G., Crawley, S. A., Cochran, L. W., Kazuba, D., Greenberg, B. D., & Murphy, D. L. (2005). Obsessive-compulsive disorder symptom dimensions show specific relationships to psychiatric comorbidity. Psychiatry Research, 135, 121-132. Henderson, G., & Pollard, A. (1988). Three types of obsessive compulsive disorder in a community sample. Journal of Clinical Psychology, 44(5), 747-752. Hermans, D., Martens, K., De Cort, K., Pieters, G., Eelen, P. (2003). Reality monitoring and metacognitive beliefs related to cognitive confidence in obsessive-compulsive disorder. Behaviour Research and Therapy, 41, 383-401. Hill, R. D., & Vandervoort, D. (1992). The effects of state anxiety on recall performance in older learners. Educational Gerontology, 18(6), 597-605. Kolada, J. L., Bland, R. C., & Newman, S. C. (1994). Epidemiology of psychiatric disorders in Edmonton. Obsessive-compulsive disorder. Acta Psychiatrica Scandinavica Supplementum, 376, 24-35. Kvavilashvili, L., & Ellis, J. (1996). Varieties of intentions: Some distinctions and classifications. In M. Brandimonte, G. 0. Einstein, & M. A. McDaniel (Eds.) Prospective memory: Theory and applications (pp. 23-51). New Jersey: Lawrence Erlbaum Associates. Macdonald, P. A., Antony, M. M., Macleod, C. M., & Richter, M. A. (1997). Memory and confidence in memory judgments among individuals with obsessive compulsive disorder and non-clinical controls. Behaviour Research and Therapy, 35(6), 497-505. Mataix-Colis, D., Rosario-Campos, M. C., & Leckman, J. F. (2005). A multidimensional model of obsessive-compulsive disorder. American Journal of Psychiatry, 162(2), 228-238. Meacham, J. A., & Dumitru, J. (1976). Prospective remembering and external-retrieval cues. JSAS Catalog of Selected Documents in Psychology, 6 no 65, 1-12. Muller, J., & Roberts, J. E. (2005). Memory and attention in obsessive-compulsive disorder: A review. Journal of Anxiety Disorders, 19, 1-28. Rachman, S., & Hodgson, R. J. (1980). Obsessions and compulsions. New Jersey: Prentice Hall. 91 Rachman, S., & Shafran, R. (1998). Cognitive and behavioral features of obsessive-compulsive disorder. In R. P. Swinson, M. M. Antony, S. Rachman, & M. A. Richter (Eds.) Obsessive-compulsive disorder: Theory, research, and treatment (51-78). London: The Guilford Press. Ramussen, S. A., & Eisen, J. A. (1988). Clinical and epidemiological findings of significance to neuropharmacologic rituals in OCD. Psychopharmacological Bulletin, 24, 466-470. Rubenstein, C. S., Peynircioglu, Z. F., Chambless, D. L., & Pigott, T. A. (1993). Memory in sub clinical obsessive-compulsive checkers. Behaviour Research and Therapy, 31(8), 759- 765. Rude, S. S., Hertel, P. T., Jarrold, W., Covich, J., & Hedlund, S. (1999). Depression-related impairments in prospective memory. Cognition and Emotion, 12(3), 267-276. Sanavio, E. (1988). Obsessions and compulsions: The padua inventory. Behaviour Research and Therapy, 26(2), 169-1 77. Sher, K., Frost, R. 0., & Otto, R. (1983). Cognitive deficits in compulsive checkers: An exploratory study. Behaviour Research and Therapy, 21(4), 357-363. Sher, K. J., Mann, B., Frost, R. 0. (1984). Cognitive dysfunction in compulsive checkers: Further explorations. Behaviour Research and Therapy, 22(5), 493-502. Smith, G. V., Della Salla, S, Logie, R. H, & Maylor, E. A. M. (2000). Prospective and retrospective memory in normal ageing and dementia: A questionnaire study. Memory, 8,311-321. Spielberger, C. D., Gorsuch, R. L., Lushene, R., Vagg, P. R., & Jacobs, G. A. (1983). Manual for the state-trait anxiety inventory (form Y). California: Consulting Psychologists Press. Stein, M. B., Forde, D. R., Anderson, G., & Walker, J. R. (1997). Obsessive-compulsive disorder in the community: An epidemiologic survey with clinical reappraisal. American Journal of Psychiatry, 154(8), 1120-1126. Tallis, F. (1995). Obsessive compulsive disorder: A cognitive and neuropsychological perspective. Chichester: John Wiley & Sons. Tallis, F., Pratt, P., Jamani, N. (1999). Obsessive compulsive disorder, checking and non verbal memory: A neuropsychological investigation. Behaviour Research and Therapy, 37, 161-166. Wallace, J. C., Kass, S. J., & Stanny, C. J. (2002). The cognitive failures questionnaire revisited: Dimensions and correlates. Journal of General Psychology, 129(3), 238-256. 92 Wilson, B., Cockburn, J., & Baddeley, A. (1985). The Rivermead behavioral memory test manual. England: Thames Valley Test Company. Woods, C. M., Vevea, J. L., Chambless, D. L., & Bayen, U. (2002). Are compulsive checkers impaired in memory? A meta-analytic review. Clinical Psychology: Science and Practice, 9(4), 353-366. 93 CHAPTER 4 SUB-CLINICAL COMPULSIVE CHECKERS SHOW IMPAIRED PERFORMANCE ON HABITUAL, EVENT- AND TIME-CUED EPISODIC PROSPECTIVE MEMORY TASKS8 Obsessive-Compulsive Disorder (OCD) is an anxiety disorder that afflicts approximately 3% of the population (Kolada, Bland, & Newman, 1994; Stein, Forde, Anderson, & Walker, 1997). The disorder is characterized by obsessions and compulsions. Obsessions are unwanted, intrusive and persistent thoughts, images or impulses that cause anxiety. Compulsions are repetitive behaviours or mental acts such as washing, counting or checking which are typically initiated in an attempt to reduce the anxiety generated by obsessions (American Psychiatric Association, 2000). OCD is a heterogeneous disorder with many different subtypes (Henderson & Pollard, 1988; Mataix-Colis, Rosario-Campos, Leckman, 2005). The most common subtype, characterized by checking compulsions, occurs in over 50% of OCD patients (Henderson & Pollard, 1988; Ramussen & Eisen 1988; Stein et al., 1997), with an additional 15% of the general population demonstrating sub-clinical checking compulsions (Stein et al., 1997). Individuals with checking compulsions have intrusive doubts that they failed to perform a task or failed to perform it properly. They worry that if they made an error some harm will befall themselves or someone else, and then they feel compelled to check to ensure that the task was performed properly (Muller & Roberts, 2005; Rachman & Shafran, 1998; Woods, Vevea, Chambless, & Bayen, 2002). To illustrate, a compulsive checker may have a recurring intrusive thought that s/he forgot to unplug the iron, worry about the consequences of such a failure (e.g., starting a fire), and then feel compelled to return home to check the status of the iron at the expense of attending to other activities. The causes of compulsive checking are not well understood, but the memory deficit theory provides some promising leads. Although foreshadowed in the writings of Janet (1903) and Freud (1909), Sher, Frost and Otto (1983) were the first to articulate and empirically examine its basic assumptions. Specifically, they suggested that “if an individual has difficulty recalling whether an intended action had been executed, they may be inclined to engage in checking behavior to insure the intended action had been carried out” (pg. 361). The existing evidence for the memory deficit theory is not very compelling however. A recent review revealed that OCD patients with checking compulsions do not demonstrate greater deficits in memory 8 A version of this chapter has been submitted for publication. Cuttler, C. & Graf, P. (in submission). Sub clinical compulsive checkers show impaired performance on habitual, event- and time-cued episodic prospective memory tasks. 94 than OCD patients without checking compulsions (Cuttler & Graf, in submission). The evidence suggested that memory deficits may be related to some factor connected with COD (e.g., depression, anxiety, medication, brain dysfunction) rather than contributing to or causing the compulsion to check. Until recently, the memory deficit theory was tested exclusively in the domain of retrospective memory, that is, in connection with the ability to remember recent events and experiences. This selective focus is somewhat surprising, in light of the examples which are typically used to illustrate checking compulsions: the tendency to check doors, windows and drawers to make sure they are properly shut and locked, and to check and recheck gas, water taps and light switches to ensure that they are turned off. What is common among these examples is that each concerns intended actions; they are all examples of activities which require prospective memory. Prospective memory is our ability to execute intended actions. Formally, prospective memory is defined as the ability to formulate intentions and plans, to retain, recollect and carry them out at the appropriate time or in the appropriate context (Einstein & McDaniel, 1996; Graf, 2005; Meacham & Dumitru, 1976). A potential link between prospective memory and the compulsion to check was recognized by Sher et al. (1983) when they observed that checkers may have problems remembering whether an “intended action had been executed” (p. 361). Nevertheless, Sher et al. (1983) focused on checkers’ retrospective memory, more specifically, they focused on checkers’ ability to look back and remember whether or not they had previously performed an intended action. By contrast, the focus of the present study was on prospective memory, on checkers’ ability to look forward and to remember at the right place or time to perform an intended action. In two recent investigations, we examined the possibility that the compulsion to check is a compensatory reaction to deficits in prospective memory (Cuttler & Graf, 2007, 2008). We reasoned that if individuals frequently experience prospective memory failures (e.g., they frequently forget to attend appointments, to pay bills and/or to turn off the iron), they may begin to worry and have intrusive doubts concerning prospective memory failures (e.g., did I turn off the iron?). Following the intrusive doubt the individual may begin to worry about the consequences of a failure and then feel compelled to check to ensure that the task was indeed completed. In other words, if people frequently forget to complete intended tasks they may develop a strategy of checking to ensure that important tasks do not remain undone. Our recent investigations revealed that checkers indeed report experiencing more problems with various aspects of prospective memory and they exhibit higher failure rates on a behavioural measure of prospective memory (Cuttler & Graf, 2007, 2008). We have demonstrated that checkers’ prospective memory deficits are not due to the elevations in depression, anxiety and 95 distractibility which are commonly associated with checking compulsions (Cuttler & Graf, 2008). Moreover, we have demonstrated that prospective memory deficits are specific to checkers, and that individuals with washing compulsions do not show the same deficits in prospective memory (Cuttler & Graf, in submission). The overall goal of the present study was to examine the link between checking compulsions and deficits in two specific components of prospective memory. Like retrospective memory, prospective memory is not a unitary construct; it includes clearly delineated components which have been labeled, habitual prospective memory, episodic prospective memory and monitoring (Graf & UttI, 2001; UttI, 2008). There remains some debate about whether the tasks used to assess monitoring are tracking a real component of prospective memory, as opposed to measuring working memory or more generally the resource demands of dual task processing (Uttl, 2008), and for this reason, monitoring was not a target of the present study. Instead, our focus was on episodic and habitual prospective memory. Episodic prospective memory tasks are those that need to be executed only once, after a delay during which the intended plan is not actively held in conscious awareness (Graf, 2005; Graf & UttI, 2001; Kvavilashvili & Ellis, 1996; Meacham & Dumitru, 1976). An example is remembering to pick a friend up at the airport in a week. This type of memory differs from habitual prospective memory by virtue of the fact that the former is concerned only with one-off tasks, tasks that are planned and executed only once, whereas the latter is required for tasks which need to be performed on a routine, regular or habitual basis (e.g., locking the door upon leaving home) (Graf & UttI, 2001; Hannon, Adams, Harrington, Fries-Dias, & Gibson, 1995; Kvavilashvili & Ellis, 1996; Meacham & Dumitru, 1976). The tasks that checkers tend to become concerned with (e.g., remembering to turn off the iron or stove or to lock the door) are best categorized as habitual prospective memory tasks, and for this reason, the primary objective of the present study was to examine the link between checking compulsions and habitual prospective memory. In two recent studies we demonstrated that checkers report experiencing more habitual prospective memory failures in everyday life (Cuttler & Graf, 2007, 2008). However, such self-reported failures must be interpreted cautiously because if checkers’ have a memory problem, they may not be able to provide reliable self-reports of their previous experiences with prospective memory failures. Therefore, in view of the potentially close theoretical link between checking compulsions and habitual prospective memory, it is important to assess in a more direct manner checkers’ performance on a habitual prospective memory task, and toward this goal, the present study included a behavioural measure of habitual prospective memory. 96 The present study also focused on episodic prospective memory. Our motivation for focusing on episodic prospective memory stems, in part, from a dissociation we found previously between checkers’ performance on two different types of episodic prospective memory tasks (Cuttler & Graf, 2007), which are often labeled event-cued and time-cued tasks (Einstein & McDaniel, 1990; Sellen, Louie, Harris & Wilkins, 1997). For event-cued prospective memory tasks the plan or intention requires execution upon the occurrence of a specific event. An example is remembering to relay a message to a colleague upon the next encounter with him/her. The next encounter with the colleague is the event that should trigger or cue prospective memory. In contrast, for time-cued prospective memory tasks the plan or intention requires execution either at a specific time or after a specific amount of time has elapsed. For example, remembering to attend a meeting at 1pm. In our previous study (Cuttler & Graf, 2007) checkers’ demonstrated a deficit on an event- but not on a time-cued episodic prospective memory task. The second objective of the present study was to examine various possible reasons for this dissociation. In our original article we suggested that checkers may have a specific deficit in event- but not time-cued episodic prospective memory (Cuttler & Graf, 2007), and that this dissociation may reflect a difference in the reparability of event- and time-cued tasks. Checking may be used to repair an event-cued prospective memory task failure. Concretely, if one fails to unplug her/his iron, s/he can always return home to unplug it. However, checking cannot be used to repair a time-cued prospective memory task failure. Once the time to perform a task has passed (e.g., pick a friend up at the airport at 2pm), no amount of repair work can transport us back in time. The possibility that the results of our previous study were due to this difference in the reparability of different kinds of prospective memory failures is consistent with the examples of checking behaviors that tend to be given in the research literature (e.g., checking to ensure that the stove or iron was turned off after use). We are not aware of any literature references to checking behaviors that occur in connection with time-cued tasks. A closely related but distinct explanation for the dissociation we found between checkers’ performance on the event- and time-cued episodic prospective memory tasks relates to the predictability of the cues used for these tasks. Time-cued tasks tend to be predictable; they involve a clear, familiar and prominent dimension (i.e., time) for anticipating or predicting the context for performing the planned activity and as such individuals can monitor/check an external device (i.e., clock) to support performance (Graf & Grodin, 2005). By contrast, for many event-cued tasks, the occurrence of a relevant cue is often unpredictable. It is possible that checkers only have difficultly or have greater difficulty with tasks that are unpredictable. To 97 examine this possibility, the present study included a behavioural measure of episodic event- cued prospective memory which was easily predictable. Yet a different explanation for the dissociation in checkers’ time- and event-cued episodic prospective memory task performance is related to the methods commonly used for scoring event- and time-cued prospective memory task performance. Event-cued prospective memory task performance is typically assessed using success/failure rates, by computing the percentage of participants in each group who completed or failed to complete an assigned task. In contrast, time-cued prospective memory task performance is typically scored by examining the delay in performing the task (i.e., the timeliness of task performance). We used these scoring methods in our previous study (Cuttler & Graf, 2007), but upon further reflection realized that assessing the delay in task performance is more of an indicator of punctuality than of the success or failure of prospective memory. Therefore, it is possible that the performance dissociation we observed between event- and time-cued tasks is due to the different methods used for scoring performance (i.e. success/failure focused vs. punctuality focused)9.To explore this possibility, we used both methods to score prospective memory task performance in the present study. A final possible explanation for the dissociation we found between checkers’ performance on the event- and time-cued episodic prospective memory tasks concerns the relevance of the tasks to the participants. In our previous study there was a difference in the personal relevance of the activities that participants had to perform for the event- and time-cued tasks. The event-cued task was personally relevant, because it required participants to request the return of a personal belonging (Cuttler & Graf, 2007). In contrast, the time-cued task was not personally relevant, and simply required participants to remind the experimenter to make a phone call (Cuttler & Graf, 2007). We see this as the least likely explanation for the performance dissociation because checking is often initiated in an attempt to prevent harm to others (Rachman & Hodgson, 1980). Nevertheless, to test the possibility that the dissociation reflects personal relevance rather than the nature of the cue (time vs. event), we also manipulated the relevance of the time- and event-cued prospective memory tasks in the present study. Study Overview Each participant was invited to the lab for two one hour testing sessions which were separated by one week. During each session participants were required to carry out a habitual prospective memory task, an episodic prospective memory task and a set of questionnaires. Unfortunately, we are not able to examine this possibility by means of the data from our previous study because in that study performance was at the ceiling. 98 Each of the sessions began with the assignment of an episodic prospective memory task which required participants to give the experimenter a reminder later in the session. As detailed in the Instruments section there were four versions of this task — a personally relevant event-cued task, a personally irrelevant event-cued task, a personally relevant time-cued task and a personally irrelevant time-cued task — permitting us to examine the influence of personal relevance and the nature of the cue (time vs. event) on checkers’ prospective memory task performance. In the first session, participants were also assigned a habitual prospective memory task which had to be performed numerous times in each session. Finally, participants also completed an identical set of questionnaires in each session, permitting us to examine the test-retest reliabilities of each of the questionnaires. In order to examine the relationship between the severity of checking compulsions and self-reported prospective memory performance, the set of questionnaires contained an inventory focused on obsessive-compulsive behaviour, with a detailed subscale for measuring checking behaviour, as well as three inventories designed to assess various aspects of prospective memory in everyday life. The latter questionnaires required participants to rate the frequency of using prospective memory aiding strategies, the frequency of completing tasks early/late (i.e., punctuality) and the frequency of experiencing various kinds of memory failures including episodic prospective memory failures, habitual prospective memory failures, internally cued prospective memory failures, time-cued prospective memory failures and retrospective memory failures. The set of questionnaires also included a depression inventory, an anxiety inventory and a cognitive failures inventory for measuring distractibility. Consistent with evidence that checkers have increased levels of depression, anxiety and distractibility (Cuttler & Graf, 2008; Frost, Sher & Geen, 1986; Hasler et al., 2005), and that these variables are associated with impairments in prospective memory (Cuttler & Graf, 2007; Harris, & Menzies, 1999; Kliegel & Jagar, 2006; Rude, Hertel, Jarrold, Covich, & Hedlund, 1999), these inventories were included so that we could rule out competing explanations for the relationship between checking compulsions and prospective memory, Several measures of personality were also included in the questionnaire booklet. However, to keep our message clear and focused on the relationship between checking compulsions and prospective memory, results related to personality differences among checkers are not reported here. Finally at the very end of each session, following completion of the questionnaires and prospective memory tasks, participants were asked to make four ratings with respect to their perception of the episodic prospective memory task assigned at the beginning of the session. 99 The primary purpose of these ratings was to provide a manipulation check, to ensure that the personal relevance manipulation worked, that participants perceived the personally relevant task to be more important to themselves and the personally irrelevant task to be more important to the experimenter. Method Participants and Design We recruited 193 undergraduate students from the University of British Columbia Department of Psychology subject pool. In order to ensure our sample included individuals with a wide range of checking behaviours, we used two types of advertisements, one that called specifically for participants with checking tendencies and another that made no mention of checking compulsions. Participants ranged from 17 to 45 years of age with a mean of 20.67 years (SD = 4.2). Participants’ postsecondary education ranged from I to 11 years with a mean of 2.56 years (SD = 1.43). Fifty-three of the participants were male and 140 were female. The overall design of the study was correlational/quasi-experimental. We correlated participants’ checking scores, assessed by means of the Padua Inventory described in the Instruments section, with their prospective memory task performance and self-reported problems related to prospective memory. We also used participants’ checking scores to assign them to one of two groups: High checkers and low checkers. Participants with checking scores above 10 were assigned to the high checkers group and participants with checking scores between 0 and 10 were assigned to the low checkers group. These cut-off points yielded a similar number of participants per group: 103 high checkers and 90 low checkers. The mean Padua Inventory checking scores were 18.47 (SD = 6.10) and 4.19 (SD = 2.79), respectively, for the high and low checkers. The high checkers’ mean Padua Inventory checking subscale score is comparable to those reported in previous research with clinical checkers (Hermans, Martens, De Cort, Pieters, Eelen, 2003; Tallis, Pratt, & Jamani, 1999). The high and low checkers did not differ in age, t(191) = 1.04, p> .05, gender, 2(1) = .05, p> .05, or years of postsecondary education, t(191) = 1.74, p> .05. Since we were interested in exploring whether either the nature of the cue or the personal relevance of the task would influence checkers’ prospective memory task performance, the design also included a 2 (time, event) x 2 (personally relevant, personally irrelevant) manipulation. All participants completed a time-cued task, an event-cued task, a personally relevant task and a personally irrelevant task, but the manner in which these tasks were combined and their order of administration varied across participants, resulting in four different conditions. Each of the 193 participants was randomly assigned to one of these 100 conditions. However, six participants failed to return for their second session10,thus leaving us with complete data from only 187 participants Table 4.1 outlines the task performed in each session for each of the four conditions as well as the total number of high and low checkers in each condition. Checking status was not related to condition, x2 (3) = .73, p> .05 and the individuals in the four conditions did not differ in age, F (3,192) = .56, p> .05, in gender, x2 (3) = 1.83, p> .05, or in years of education, F (3,192) = .33, p> .05. Table 4.1. Description of the Episodic Prospective Memory Tasks Assigned in Each Session for Each Condition and the Total Number of Low and High Checkers who Completed Each Task. Session I Session 2 (n=193) (n=187) Condition Task n of Each Group Task n of Each Group Personally Personally Condition Irrelevant Low Checkers n = 26 Relevant Low Checkers n = 26 A Event-Cued High Checkers n = 25 Time-Cued High Checkers n = 23 Task Task Personally Personally Condition Irrelevant Low Checkers n = 21 Relevant Low Checkers n = 19 B Time-Cued High Checkers n = 27 Event-Cued High Checkers n = 26 Task Task Personally Personally Condition Relevant Low Checkers n = 22 Irrelevant Low Checkers n = 22 C Time-Cued High Checkers n = 24 Event-Cued High Checkers n = 24 Task Task Personally Personally Condition Relevant Low Checkers n = 21 Irrelevant Low Checkers n = 21 D Event-Cued High Checkers n = 27 Time-Cued High Checkers n = 26 Task Task We conducted this research with the approval of the University of British Columbia Behavioural Ethical Review Board. Completing all components of the study required approximately two hours, and participants received course credit in return for their participation. Instruments In the course of the experiment each participant was administered a number of questionnaires and prospective memory tasks. Each of these study components is described below. 10 Interestingly, four of the six (66.67%) participants who failed to show up for their second session (a task requiring prospective memory) were high checkers. 101 Habitual Prospective Memory Task. To assess habitual prospective memory, participants were asked to write their subject ID code on the top right hand corner of every page of the questionnaire booklet. The instructions emphasized that ID codes were to be written on each page immediately before beginning that page. Participants were not permitted to record ID codes on previously completed pages or to record codes on yet to be completed pages. Participants were further instructed that when they returned for the second session they would be asked to fill out a similar booklet of questionnaires and that they should write their ID code on every page of that questionnaire booklet as well. They were told that they would not be reminded to perform this task when they returned; that it would be up to them to remember to perform this task on their own. Participants were further instructed that if “you are unable to remember your ID code when you return for the second session you can request it from the experimenter.” Finally, participants were told their subject ID code. Each questionnaire booklet contained 21 pages and for each session the critical dependent variable was the percentage of pages with a missing ID code. Episodic Prospective Memory Tasks. We developed a series of episodic prospective memory tasks for the present study. Each of the tasks required participants to provide a reminder. One of the tasks was personally relevant; it was designed to be important and relevant to participants. The task required participants to remind the experimenter to submit their research participation credit online. The other task was personally irrelevant; it was designed to be important and relevant to the experimenter but not to participants. The task required participants to remind the experimenter to send an email to her supervisor. We constructed two versions of each of these tasks, an event-cued version and a time- cued version. For the event-cued version participants were told that they would be asked to complete a series of questionnaires and that the last questionnaire would be labeled the NEO Five Factor Inventory. They were asked to give the experimenter the assigned reminder immediately before filling out this questionnaire. The title of each questionnaire, including the NEO Five Factor Inventory, appeared at the top left corner of the questionnaire in 14 point bold font. Performance was scored in both a binary and continuous manner. The binary measure was designed to reflect failure rates. Participants who gave the experimenter a reminder received a score of 0 and participants who failed to give the reminder received a score of 1. The continuous measure reflected the delay in providing the reminder (i.e., the timeliness of the reminder). Participants who reminded the experimenter to submit their credit before beginning the 60 item NEO Five Factor Inventory received a score of 0, those who gave the reminder after completing one of the items on the inventory received a score of 1, those who gave the 102 reminder after completing two of the items received a score of 2 and so on. Participants who gave the reminder after completing all 60 inventory items received a score of 60. For the time cued version of the episodic prospective memory tasks participants were told to give the experimenter the reminder in exactly 35 minutes; they were instructed to try to give the reminder as close as possible to this 35 minute mark. Participants were asked to set aside all personal time keeping devices (i.e., watches, cell phones) for the duration of the experiment. We ran a clock program on a desktop computer so that participants could keep track of the time. Any time they wished, participants could press the spacebar to reveal, for exactly one second, the total number of minutes and seconds that had passed. The program recorded the number of times the participants checked the time as well as the time of each check. Performance on the task was scored in both a binary and continuous fashion. The binary method was designed to reflect failure rates. Participants who gave a reminder after 35 minutes received a score of 0 and those who completely failed to give a reminder received a score of 1. The continuous measure was used to track the delay in providing the reminder (i.e., the number of minutes past the 35 minute mark). Time-Cued Prospective Memory Questionnaire (TCPMQ). The currently available self-report instruments for assessing prospective memory primarily tap event-cued prospective memory task performance and they contain too few items to meaningfully assess everyday life time-cued prospective memory task performance. Therefore, in order to complement the existing inventories, we developed a new inventory for assessing various aspects of time-cued prospective memory. The new inventory has three sections. The first section consists of 39 statements each describing a time-cued prospective memory failure (e.g., “I forget to go to a class”, “I forget to take food out of the oven”), and participants are required to rate by means of a 6-point scale how frequently they tend to experience each type of failure. The rating scale points are marked: n/a = not applicable, 1 = never, 2 = seldom, 3 = sometimes, 4 = often, 5 = very often. Participants are instructed to mark not applicable if they never perform the task described or if they never perform the task at specific times. The second section of the questionnaire assesses participants’ punctuality. Participants are presented with the same 39 time-cued prospective memory task descriptions (e.g., “I arrive at class”, “I take food out of the oven”) and are asked to rate with the aid of a 6-point scale how punctual they usually are when they perform the activity. The punctuality scale points are marked: n/a = not applicable, I = very early, 2 = a bit early, 3 = on time, 4 = a bit late, 5 = very late. Once again participants are asked to mark not applicable if they never perform the task described or if they never perform the task at specific times. The third section of the questionnaire assesses participants’ use of time-cued prospective memory aiding strategies. Participants are presented with 11 statements that refer 103 to various strategies that could be used facilitate time-cued prospective memory task performance (e.g., “I write notes on my hand to help me remember to do things on time”, “I use the alarm function on my cell phone to help me remember to do things on time”). Participants are provided with a 6-point scale for rating how often they use each strategy, with the scale points marked: n/a = not applicable, I = never, 2 = seldom, 3 = sometimes, 4 = often, 5 = very often. Participants are asked to respond with n/a if they do not have access to the listed device (e.g., a cell phone). The results from the present study revealed high test-retest reliability coefficients for each section or subscale of the instrument, with coefficients of .87 for the time- cued prospective memory section, .80 for the punctuality section and .81 for the time-cued prospective memory aiding strategies section. For each participant, we computed subscale scores based on the average of all items from each section and then computed the average of their session one and session two subscale scores. Thus, subscale scores could range from 1 to 5, with higher scores indicating a greater frequency of time-cued prospective memory failures, of being late or of using time-cued prospective memory aiding strategies. Prospective Memory Questionnaire (PMQ). The PMQ is a standardized self-report inventory developed by Hannon et al. (1995). The scale contains 52 statements concerning various prospective memory failures (e.g., “I forgot to lock up my house, bike or car”) and prospective memory aiding strategies (e.g., “I write myself reminder notes”). Typically participants have to rate how commonly they experience each failure in a given time period. However, we used a slightly modified version of this inventory which required participants to rate on a 6-point scale the frequency of experiencing each failure. The rating scale points were marked: n/a = not applicable, 1 = never, 2 rarely, 3 sometimes, 4 often, 5 = very often. Participants were instructed to mark not applicable if the item does not apply to them because they never perform the task described. The questionnaire contains four subscales, three of which concern failures connected with different prospective memory tasks, specifically, episodic tasks, habitual tasks and internally cued tasks. Internally cued tasks are those where no salient external cue is provided for prompting the retrieval of the previously formed plan, for example, remembering to cash a paycheck before running out of money (Hannon et al., 1995). The fourth subscale measures the frequency of using various prospective memory aiding strategies. Each subscale is reliable, with reliability coefficients ranging from .78 to .90, and test-retest reliabilities ranging from .64 to .88 (Hannon et al., 1995). In the present sample, test-retest reliabilities were .88 for the episodic subscale, .81 for the habitual subscale, .81 for the internally cued subscale and .87 for the memory aiding strategies subscale. We computed participants’ scores on each subscale by averaging the ratings they gave on the items contained within each subscale. We then took the average of participants’ session one and session two subscale scores. Thus, 104 possible scores on each subscale range from I to 5, with higher scores indicating a greater frequency of prospective memory failures or use of prospective memory aiding strategies. Prospective and Retrospective Memory Questionnaire (PRMQ). The PRMQ is a standardized self-report inventory developed by Smith, Della Sala, Logie and Maylor (2000). The scale contains 16 questions concerning various memory failures (e.g., “Do you forget appointments if you are not prompted by someone else or by a reminder such as a calendar or diary?”). We used a slightly modified version of this inventory which required participants to rate on a 6-point scale the frequency of experiencing each failure. The rating scale points were marked: n/a = not applicable, 1 never, 2 = rarely, 3 = sometimes, 4 = often, 5 very often. Participants were instructed to mark not applicable if the item does not apply to them because they never perform the task described. The scale contains two subscales, one focusing on prospective memory and the other on retrospective memory. Crawford, Smith, Maylor, Della Salla and Logie (2003) found that both subscales yield high reliability coefficients (prospective memory = .84, retrospective memory = .80). In the present sample, test-retest reliabilities were .87 and .78 for the prospective memory and retrospective memory subscales respectively. We computed participants’ prospective memory and retrospective memory scores by averaging their ratings on the prospective memory and the retrospective memory items, respectively. We then took the average of participants’ session one and session two subscale scores. Possible scores on each subscale range from 1 to 5, with higher scores indicating a greater frequency of memory failures. Cognitive Failures Questionnaire (CFQ). The CFQ is a standardized self-report inventory developed by Broadbent, Cooper, FitzGerald and Parkes (1982). The inventory is used for assessing the frequency of experiencing everyday cognitive slips or errors, in other words, the frequency with which individuals make mistakes on simple tasks. The CFQ contains 25 questions concerning various action slips or errors (e.g., “Do you start doing one thing at home and get distracted into doing something else unintentionally?”). For each item, participants indicate the frequency of experiencing the failure on a 5-point scale, with scale points marked: 1 = never, 2 = very rarely, 3 = occasionally, 4 = quite often, 5 = very often. The questionnaire contains four subscales: Distractibility, memory, blunders and memory for names. Each of these subscales is reliable, with reliability coefficients ranging from .76 to .86 (Wallace, Kass & Stanny, 2002). In the present sample, test-retest reliabilities ranged from .80 to .84. We computed participants’ scores on the distractibility subscale by averaging the ratings they gave on the relevant items. We then took the average of participants’ session one and session two scores on the distractibility subscale. Possible scores on the distractibility subscale range from I to 5, with higher scores indicating a greater degree of distractibility. 105 Padua Inventory. The Padua Inventory is a standardized self-report inventory developed by Sanavio (1988). The inventory contains 60 statements describing common obsessive-compulsive behaviours (e.g., “I return home to check doors, windows, drawers, etc., to make sure they are properly shut”, “I wash my hands more often and longer than necessary”). For each item participants use a 5-point scale to indicate the degree of disturbance the behaviour creates, with scale points marked: 0 = none at all, 1= a little, 2 quite a lot, 3 = a lot, 4 very much. The inventory contains four subscales each of which measures different obsessive-compulsive concerns: Checking, contamination, impaired control of mental activities and worries about losing control over motor behaviours (Sanavio, 1988). The questionnaire is a reliable measure of obsessive-compulsive behaviour, as evidenced by high test-retest reliability (.80) and alpha coefficients (.94) (Sanavio, 1988). In the present sample, test-retest reliabilities were .91 for the checking subscale, .90 for the contamination subscale, .87 for the impaired control of mental activities subscale and .84 for the urges and worries about losing control over motor behaviours subscale. We indexed the severity of participants’ checking compulsions by summing the ratings they gave on the items contained on the checking subscale of the inventory. We then took the average of participants’ session one and session two checking subscale scores. Possible scores range from 0 to 32, with higher scores indicating more severe checking compulsions. Using these scores we also created a binary checking status variable. Participants with scores over 10 were classified as high checkers and participants with scores between 0 and 10 were classified as low checkers. For the remainder of this report we will use the term severity of checking compulsions when referring to the continuous index of checking and the term checking status when referring to the binary index of checking. As an index of the severity of washing compulsions we computed participants’ contamination scores by summing the ratings they gave on the items contained in the contamination subscale of the inventory. We then took the average of participants’ session one and session two contamination subscale scores. Possible scores range from 0 to 44, with higher scores indicating more severe washing corn pulsions. Beck Depression Inventory (BDI). The BDI is a self-report inventory for measuring depression (Beck, Ward, Mendelson, Mock, & Erbaugh, 1961; Beck, 1987). The inventory contains 21 groups of four statements related to symptoms of depression (e.g., “0 = I don’t feel disappointed in myself, 1 = I am disappointed in myself, 2 = I am disgusted with myself, 3 = I hate myself”). Participants are instructed to circle the number corresponding to the one statement in each group which best describes how they have been feeling during the past week. The inventory is reliable with split-half reliabilities ranging from .78 to .93 and test-retest reliabilities ranging from .48 for psychiatric patients to .74 for undergraduate students (Beck, 1987). In the present sample, the test-retest reliability coefficient was .85. We computed 106 participants’ depression scores by averaging all of the items they circled in the inventory. We then took the average of participants’ session one and session two depression scores. Possible scores range from 0 to 3, with higher scores reflecting more severe depression. State Trait Anxiety Inventory (STAI). The STAI is an inventory for measuring state and trait anxiety (Spielberger, Gorsuch, Lushene, Vagg, & Jacobs, 1983). The inventory contains two parts, one to assess state anxiety or “feelings of apprehension, tension, nervousness and worry” (Spielberger et al., 1983, p.2), and another to assess trait anxiety or “relatively stable individual differences in anxiety-proneness” (Spielberger et al., 1983, p.1). The state anxiety section contains 20 statements that assess how individuals feel at the moment of completing the inventory. The trait anxiety section contains 20 statements that assess how individuals generally feel. For each item participants use a 4-point scale to indicate how they feel. On the state anxiety section the points are marked: I = not at all, 2 = somewhat, 3 = moderately so, 4 = very much so. On the trait anxiety section the points are marked: 1 = almost never, 2 = sometimes, 3 = often, 4 almost always. Both sections of the inventory yield high alpha coefficients for college students (state anxiety = .92, trait anxiety = .91) (Spielberger et al., 1983). Test-retest reliability in college students is high on the trait anxiety section (.73 to .86) (Spielberger et al., 1983). As expected, test-retest reliability in college students is much lower on the state anxiety section (.16 to .54) (Spielberger et al., 1983). In the present sample, test- retest reliabilities were .93 for the trait anxiety section and .66 for the state anxiety section. We computed participants’ scores on each subscale by averaging the ratings they gave on the items contained within each subscale. We then took the average of participants’ session one and session two subscale scores. Possible scores for each section range from 1 to 4, with higher scores reflecting more anxiety. Personality Assessment Inventories. Participants also completed three personality inventories: the Multidimensional Perfectionism Scale (MPS) (Hewitt & Flett, 2004), the Frost Multidimensional Perfectionism Scale (F-MPS) (Frost, Marten, Lahart & Rosenblate, 1990), the NEC Five Factor Inventory (Costa & McCrae, 1992). In order to keep the present paper focused on the relationship between compulsive checking and prospective memory, the personality test results are not reported here. Task Relevance Ratings. Participants were also asked to make ratings concerning the personal relevance of each of the assigned episodic prospective memory tasks. The primary purpose of these ratings was to provide a manipulation check, to find out whether participants perceived the personally relevant task as more important to themselves and the personally irrelevant task as more important to the experimenter. Participants were required to make three ratings of each of the assigned tasks: one concerned the overall importance of completing the 107 task according to instructions, another concerned how important it was for them that they completed the task according to instructions and a third concerned how important it was for the experimenter that they completed the task according to instructions. Participants were given a 5-point scale for making these ratings with scale points marked: 1 = extremely unimportant, 2 = not very important, 3 = somewhat important, 4 = very important, 5 = extremely important. Participants were also asked to rate how responsible they felt for completing each of the assigned tasks according to instructions. They were given a 5-point scale for making these ratings with scale points marked: 1 = extremely unresponsible, 2 = not very responsible, 3 = somewhat responsible, 4 = very responsible, 5 = extremely responsible. Procedure Participants were tested one at a time in a quiet room. The study consisted of two one hour testing sessions. In session one, we first obtained written informed consent and collected basic demographic information. Participants were randomly assigned one of the four versions of the episodic prospective memory tasks. When participants indicated that they understood the instructions for the episodic prospective memory task, they were assigned the habitual prospective memory task. Participants then completed the set of questionnaires in the following order: TCPMQ, PMQ, PRMQ, CFQ, Padua Inventory, BDI, STAI, MPS, F-MPS, NEC Five Factor Inventory. Finally, participants completed the Task Relevance Ratings. In session two, participants were assigned the complementary version of the episodic prospective memory task (see Table 4.1). After participants indicated that they understood the task instructions they were given the same set of questionnaires in the same order as in session one. They then completed the Task Relevance Ratings. Finally, they received a full debriefing. Results Data Preparation All data were checked for recording and transcription errors. Data accuracy was greater than 99%. We examined all of the questionnaire data, the habitual prospective memory task data, the continuous (i.e., delay) data for the episodic prospective memory tasks and the clock checking data for the time-cued prospective memory tasks in each session for univariate outliers, defined as scores falling more than three standard deviations away from the mean. We discovered 22 outliers in the data from the first session (i.e., < .005%) and 25 outliers in the data from the second session (i.e., <.005%), and they were replaced with the nearest non outlying value, specifically, a score either -3 or +3 standard deviations away from the corresponding mean. 108 Habitual Prospective Memory Task Performance Our primary objective was to examine whether checkers demonstrate deficits on a behavioural measure of habitual prospective memory. To address this objective we required participants to write their subject ID code on the top right hand corner of every page of each of the questionnaire booklets. Participants were given the task instructions only in the first session and they were told that it would be up to them to remember to perform the task when they returned for the second session. For each session, the critical dependent variable was the percentage of pages in the questionnaire booklet with a missing subject ID code. The main findings from the habitual prospective memory task are depicted in Figure 4.1. A 2 x 2 mixed ANOVA of these data, with Session (session one, session two) as a within subject factor and Checking Status (low checkers, high checkers) as a between subjects factor, showed a significant main effect due to Checking Status, F (1, 185) = 4.08, MSE = 298.22, p < .05. No other effects were significant. Figure 4.1. Mean Performance (+SE) on the Habitual Prospective Memory Task in Each Session for Low Checkers (n = 88) and High Checkers (n = 99). 15 Cl) 10 0) C 0 Episodic Prospective Memory Task Performance Our secondary objective was to gain insight into the dissociation we previously reported between checkers’ performance on an event- and time-cued episodic prospective memory task (Cuttler & Graf, 2007). To address this objective we required participants to give the experimenter a different reminder in each of two testing sessions. Each participant completed a time-cued task, an event-cued task, a personally relevant task and a personally irrelevant task o Low Checkers • High Checkers Session One Session Two 109 but the manner in which these tasks were combined and their order of administration varied across participants (see Table 4.1). In a preliminary analysis we examined whether participants perceived the personally relevant task to be more relevant to themselves than to the experimenter and the personally irrelevant task to be more relevant to the experimenter than to themselves. The results showed that participants rated the personally relevant task to be more important to themselves than to the experimenter, t (190) = 3.00, p < .01. However, the participants rated the irrelevant task to be equally important to themselves and the experimenter, t (189) = .17, p> .1. We also examined all task relevance rating items to determine whether participants gave different ratings for the relevant and irrelevant tasks and whether low and high checkers gave different ratings, and the data revealed no significant differences. This outcome indicates that our personal relevance manipulation was not entirely successful and reduces our power to find significant effects due to this manipulation. It may be very difficult to come up with an effective manipulation of personal relevance under controlled experimental conditions. Participants may perceive all tasks as important and relevant to themselves because they view their performance in the lab as a reflection of their abilities. In a second preliminary analysis we examined episodic prospective memory task failure rates across the sessions. The results showed that Session produced no main effect and did not interact with Checking Status (low checkers, high checkers). For this reason we did not include Session as a variable in any further analyses. Of primary interest was the influence of Cue Type (time, event) and Personal Relevance (relevant, irrelevant) on episodic prospective memory task performance. We examined the influence of these variables by means of two separate 2 x 2 mixed ANOVAs, the first with Cue Type as the within subject factor and Checking Status as the between subject factor, and the second with Personal Relevance as the within subject factor and Checking Status as the between subject factor. For each analysis, the dependent variable was failure rates on the tasks. The results of both analyses, as depicted in Figures 4.2 and 4.3, revealed a significant main effect due to Checking Status, F (1, 184) = 4.71, MSE = .06, p < .05. No other main effects or interactions were significant. To examine whether the dissociation we previously reported between checkers’ performance on an event- and time-cued episodic prospective memory task (Cuttler & Graf, 2007) reflected our method of measuring time-cued prospective memory (i.e., measuring the delay in providing the reminder rather than failure rates), we reexamined performance on the event- and time-cued episodic prospective memory tasks using the continuous scoring methods described in the Instruments section. These scoring methods reflect the delay in providing the 110 Figure 4.3. Prospective Failure Rates (+SE) on the Personally Relevant and Personally Irrelevant Episodic Memory Tasks for Low Checkers (n = 88) and High Checkers (n = 98). reminder (i.e., the timeliness of giving the reminder), and for this reason, those participants who failed to give a reminder could not be included in this analysis. We analyzed the reminder delay data with a 2 x 2 mixed ANOVA with Cue Type (time, event) as a within subject factor and Checking Status (low checkers, high checkers) as a between subjects factor. The results, depicted in Figure 4.4, revealed no significant main or interaction effects. 20 Figure 4.2. Failure Rates (+SE) on the Time- and Event-Cued Episodic Prospective Memory Tasks for Low Checkers (n = 88) and High Checkers (n = 98). 15 D Low Checkers • High Checkers 10 5 0 lime-Cued Eent-Cued 20 15 1 0 5 ci Low Checkers High Checkers 0 Relevant Irrelevant 111 Figure 4.4. Mean Delay in Providing Reminders (+SE) for the Time- and Event-Cued Episodic Prospective Memory Tasks for Low Checkers (n = 77) and High Checkers (n = 76). 8 Figure 4.5. Mean Number (+SE) of Clock Checks during the 35 Minute Interval of the Time- Cued Prospective Memory Tasks, Broken Down into 5-Minute Intervals for Low Checkers (n = 90) and High Checkers (n = 100). 3 o Low Checkers • High Checkers 1- 0.5 - 0- For the time-cued prospective memory tasks participants had to give the experimenter a reminder as close as possible to the 35 minute mark. They had access to a clock which they could consult at any time by pressing the spacebar on a computer keyboard. To assess whether low and high checkers employed different clock monitoring strategies we conducted a 2 x 7 ANOVA with Checking Status (low checkers, high checkers) as a between subjects factor and 5-minute segments of the retention interval as a within subject factor. The results in Figure 4.5 revealed a significant main effect due to the 5-minute segments of the retention interval, F (6, 1128) = 91.27, MSE = .50, p < .001, but no other effects were significant. lime-Cued Eent-Cued U) C.) a) Ci C.) 0 C.) .4- 0 a) .0 E z z • Low Checkers - -- - High Checkers 2 1 0 1-5 mins 5-10 mins 10-15 mins 15-20 mins 20-25 mins 25-30 mins 30-35 mins Time Interval 112 Self-Reported Prospective Memory Performance Do checkers experience more prospective memory failures in their everyday life? Do they have confidence in their prospective memory or do they recognize their prospective memory is impaired? To answer these questions, we examined the ratings participants gave on the various subscales of the TCPMQ, PMQ and PRMQ. The main findings from the TCPMQ are depicted in Figure 4.6. We explored participants’ ratings using a 2 x 3 mixed ANOVA with Subscale (time-cued prospective memory failures, punctuality, time-cued prospective memory aiding strategies) as the within subject factor and Checking Status (low checkers, high checkers) as the between subjects factor. The results showed significant main effects due to Subscale, F (2, 382) = 377.70, MSE = .18, p < .001, and Checking Status, F (1, 191) = 7.07, MSE .41, p < .01 as well as an interaction between these factors, F (2, 382) = 5.23, MSE = .18, p <.01. Follow-up independent samples t-tests indicated that the high checkers gave significantly higher ratings on the time-cued prospective memory failures subscale, t (191) = -2.45, p < .05, and on the time-cued prospective memory aiding strategies subscale, t (191) = -2.86, p < .01. However, on the punctuality subscale, the high and low checkers’ ratings did not differ significantly, t (191) = -.08, p> .1. Figure 4.6. Mean Ratings (+SE) on the Three Subscales of the Time-Cued Prospective Memory Questionnaire (TCPMQ) for Low Checkers (n = 90) and High Checkers (n = 103). D Low Checkers ‘ • High Checkers 2- C’) CU a) LLO > a) a) L (fl4 ‘4- o >c: a)0- z. — . -. Time-Cued Prospecth Time-Cued Prospectie Memory Memory Aiding Strategies The findings from the PMQ are depicted in Figure 4.7. Participants’ ratings on the PMQ were subjected to a 2 x 4 mixed ANOVA with Subscale (episodic, habitual, internally cued, memory aiding strategies) as a within subject factor and Checking Status (low checkers, high checkers) as a between subjects factor. The results showed significant main effects due to Subscale, F (3, 573) = 376.45, MSE = .24, p < .001, and Checking Status, F (1, 191) = 18.86, MSE = .80, p < .001 as well as an interaction between these factors, F (3, 573) = 4.35, MSE = Punctuality 113 .24, p < .01. Follow-up independent samples t-tests showed that high checkers gave significantly higher ratings on all four subscales, t (191) = -1.91, p < .05 (one tailed), t (191) - 3.44, p <.01, t (191) = -3.22, p <.01, t (191) = -4.18, p <.001, respectively, for episodic, habitual, internally cued and memory aiding strategies. Figure 4.7. Mean Ratings (+SE) on the Four Subscales of the Prospective Memory Questionnaire (PMQ) for Low Checkers (n = 90) and High Checkers (n = 103). El Low Checkers 3 • High Checkers I ___I _ Episodic Prospectie Habitual Prospectn Internally Cued Memory Aiding Memory Memory Prospectie Memory Strategies The findings from the PRMQ are depicted in Figure 4.8. We assessed participants’ ratings on the PRMQ using 2 x 2 mixed ANOVA with Subscale (prospective memory, retrospective memory) as a within subject factor and Checking Status (low checkers, high checkers) as a between subjects factor. The results showed significant main effects due to Subscale, F (1, 191) 77.52, MSE = .11, p <.001, and Checking Status, F (1, 191) = 11.19, MSE = .70, p < .01. The interaction between these factors was not significant. Figure 4.8. Mean Ratings (+SE) on the Two Subscales of the Prospective and Retrospective Memory Questionnaire (PRMQ) for Low Checkers (n = 90) and High Checkers (n = 103). c Low ChecIi 3-1 • High Checkers I I ____ ____ Prospectk Memory Retrospectie Memory 114 The Influence of Depression, State Anxiety, Trait Anxiety and Distractibility Do depression, anxiety and distractibility account for the relationships between checking compulsions and problems with prospective memory? Checking compulsions are known to overlap with washing compulsions11 (Rachman & Hodgson, 1980). Therefore in order to address our question about possible confounding variables in its purest form we first ran a regression analysis on the severity of checking compulsions to remove any influences due to washing compulsions, and we saved and used the residuals for all subsequent analyses on the severity of checking compulsions. Next, we computed the zero-order correlations — listed in Table 4.2 — among the severity of checking compulsions, depression, state anxiety, trait anxiety, distractibility and our various measures of prospective memory. Consistent with previous research, the correlations among the severity of checking compulsions, depression, state anxiety, trait anxiety and distractibility were significant (Cuttler & Graf, 2008; Frost et al., 1986; Hasler, et al., 2005; Spielberger et al., 1983). Of primary importance to the present study, the severity of checking compulsions was related to increased levels of depression, trait anxiety, state anxiety and distractibility. Habitual Prospective Memory Task Performance. As shown in Table 4.2, failures on the habitual prospective memory task were significantly correlated with the severity of checking compulsions (r = .18), depression (r =19) and trait anxiety (r =.13). To examine whether depression and/or trait anxiety account for the relationship between habitual prospective memory task failures and the severity of checking compulsions we computed partial correlations, controlling for depression and trait anxiety. The relationship between habitual prospective memory task failures and the severity of checking compulsions remained significant after controlling for depression, r (184) .13, p < .05 (one tailed), and trait anxiety, r (184) = .16, p < .05. The pattern of findings indicates that there is a direct relationship between habitual prospective memory task failures and the severity of checking compulsions that is independent of depression, anxiety and distractibility. Episodic Prospective Memory Task Performance. The correlation between overall failures on the episodic prospective memory tasks and the severity of checking compulsions was not significant, r (184) = .06, p> .1. However, when we entered checking status as a binary variable12 (low checkers vs. high checkers), the correlation with episodic prospective memory The correlation between severity of checking compulsions and washing compulsions was .50 in the present sample. 12 Consistent with the other partial correlation analyses, the influence of washing compulsions was first removed by regressing checking on washing and saving the residuals. 115 T ab le 4. 2. Ze ro -O rd er Co rre la tio ns A m on g Ch ec ki ng ,D ep re ss io n, St at e A nx ie ty ,T ra it A nx ie ty , D ist ra ct ib ili ty an d th e V ar io us M ea su re s of Pr os pe ct iv e M em or y. V ar ia bl e Ch ec ki ng D ep re ss io n St at e A nx ie ty Tr ai tA nx ie ty D ist ra ct ib ili ty D ep re ss io n . 27 ** - - - - St at e A nx ie ty . 28 ** . 67 ** - - - Tr ai tA nx ie ty . 32 ** 79 ** 78 ** - - D ist ra ct ib ili ty . 32 ** 42 ** . 42 ** 52 ** - H ab itu al Pr os pe ct iv e M em or y Ta sk Pe rf or m an ce . 18 * . 19 *1 . 08 . 13 *1 . 09 Ep iso di c Pr os pe ct iv e M em or y Ta sk Pe rf or m an ce . 05 ( l 5* )t . 09 . 13 *1 . 12 . 05 TC PM Q — Ti m e- Cu ed Pr os pe ct iv e M em or y . 20 * . 25 ** 29 ** . 20 * 55 ** TC PM Q— Pu nc tua lity . 01 . 13 *1 . 12 . 08 . 18 * TC PM Q— Ti me -C ue d M em or y Ai di ng St ra te gi es . 20 * . 17 * . 25 ** . 15 * . 25 ** PM Q — Ep iso di c Pr os pe ct iv e M em or y . 17 * 26 ** . 29 ** 24 ** 59 ** PM Q — H ab itu al Pr os pe ct iv e M em or y . 18 * 24 ** 32 ** 26 ** PM Q — In te rn al ly Cu ed Pr os pe ct iv e M em or y . 25 ** . 36 * 43 ** 43 ** PM Q — M em or y Ai di ng St ra te gi es . 28 ** . 22 * . 21 * . 21 * PR M Q — Pr os pe ct iv e M em or y . 23 * 34 ** 35 ** 37 ** 74 ** PR M Q — R et ro sp ec tiv e M em or y . 27 ** . 28 ** 34 ** . 32 ** . 68 ** * 1 in di ca te s p . 05 (o ne ta ile d), * in di ca te s p . 05 ,* * in di ca te s p . 00 1 tin di ca te s th e co rr el at io n w ith th e bi na ry Ch ec ki ng St at us (lo w ch ec ke rs ,h ig h ch ec ke rs )v ar ia bl e - - 0) task failures was significant, r (183) = .15, p < .05. As shown in Table 4.2, of the four potential confounding variables only state anxiety was significantly correlated with episodic prospective memory task failures. We computed a partial correlation to examine whether the correlation between failures on the episodic prospective memory tasks and checking status was due to state anxiety. The partial correlation between checking status and task failures remained significant, r (183) = .16, p < .05. The results suggest that there is a direct link between episodic prospective memory task failures and checking that is independent of depression, state anxiety, trait anxiety and distractibility. Self-Reported Prospective Memory Performance. As shown in Table 4.2, the severity of checking compulsions, depression, state anxiety, trait anxiety and distractibility were each significantly correlated with higher ratings on the various prospective memory questionnaire subscales, with an exception; only depression and distractibility were significantly correlated with the punctuality subscale of the TCPMQ. To examine whether the relationships between the severity of checking compulsions and ratings on the prospective memory questionnaire subscales13 are secondary to depression, anxiety and/or distractibility we conducted a series of partial correlations which are displayed in Table 4.3. The second column of Table 4.3 shows the correlations between the severity of checking compulsions and ratings on each of the questionnaire subscales after controlling for depression. The results show that the correlation between the severity of checking compulsions and ratings on the episodic prospective memory subscale of the PMQ was no longer significant after controlling for depression. The results suggest that depression accounts for the relationship between self-reported problems with episodic prospective memory and the severity of checking compulsions. Depression does not account for the relationship between the severity of checking compulsions and ratings on any other questionnaire subscale. The third column of Table 4.3 shows the correlations between the severity of checking compulsions and ratings on each of the questionnaire subscales after controlling for both state and trait anxiety14.When anxiety was controlled for the relationships between the severity of checking compulsions and ratings of episodic and habitual prospective memory subscales of the PMQ were no longer significant. The results suggest that the relationship between the severity of checking compulsions and ratings of episodic and habitual prospective memory are 13 Since checking compulsions were not significantly related to self-reported problems with punctuality we did not include the punctuality subscale of the TCPMQ in the partial correlation analyses. 14 Examining state and trait anxiety separately produced no differences in the results of these analyses. 117 T ab le 4. 3. Pa rti al Co rre la tio ns B et w ee n Se ve rit y of Ch ec ki ng Co m pu lsi on s an d R at in gs o n th e TC PM Q, PM Q an d PR M Q Su bs ca le s A fte rC on tro lli ng fo rD ep re ss io n, St at e an d Tr ai tA nx ie ty an d D ist ra ct ib ili ty Ch ec ki ng Ch ec ki ng Ch ec ki ng Qu es tio nn air ea n d Su bs ca le (C on tro llin gf or (C on tro llin gf or St at e (C on tro llin gf or D ep re ss io n) A nx ie ty an d Tr ai t D ist ra ct ib ili ty ) A nx ie ty ) TC PM Q — Ti m e- Cu ed Pr os pe ct iv e M em or y . 14 *1 . 14 *1 03 TC PM Q — Ti m e- Cu ed M em or y Ai di ng St ra te gi es . 17 * . 16 * 14 *1 PM Q — Ep iso di c Pr os pe ct iv e M em or y . 11 . 09 - . 03 PM Q — H ab itu al Pr os pe ct iv e M em or y . 13 *1 . 10 . 02 PM Q — In te rn al ly Cu ed Pr os pe ct iv e M em or y . 17 * . 12 *1 . 01 PM Q — M em or y Ai di ng St ra te gi es . 24 ** . 23 ** . 20 * PR M Q — Pr os pe ct iv e M em or y . 15 * . 12 *1 - . 02 PR M Q — R et ro sp ec tiv e M em or y . 22 * . 18 * . 07 * 1 in di ca te s p . 05 (o ne ta ile d), * in di ca te s p . 05 ,* * in di ca te s p . 00 1 - - co secondary to anxiety. Anxiety does not account for the correlations between the severity of checking compulsions and ratings on the remaining questionnaire subscales. Finally, the fourth column of Table 4.3 shows the correlations between the severity of checking compulsions and ratings on each of the questionnaire subscales, after controlling for distractibility. When distractibility was controlled for the correlations between the severity of checking compulsions and ratings on all of the questionnaire subscales tapping memory were no longer significant. Only the partial correlations between the severity of checking compulsions and the memory aiding strategies subscales of the TCPMQ and PMQ remained significant. The results of these analyses indicate that the relationships between the severity of checking compulsions and ratings on the various subscales assessing memory are driven by distractibility. Discussion In two recent papers we demonstrated a link between checking compulsions and deficits in prospective memory (Cuttler & Graf, 2007, 2008). For the present study, we examined this link further, first, by exploring whether checkers are impaired on a behavioral — as opposed to a self-report — measure of habitual prospective memory, and second, by investigating the origins of a dissociation we previously reported between checkers’ performance on event- and time- cued episodic prospective memory tasks (Cuttler & Graf, 2007). The results revealed a clear deficit in the habitual prospective memory task performance of checkers. They also showed that the previously reported dissociation between event- and time-cued prospective memory tasks was an artifact of the different methods typically employed to score performance on these tasks. The finding that checkers’ performance on a behavioral measure of habitual prospective memory is impaired is novel and complements the self-report data from previous research (Cuttler & Graf, 2007, 2008) as well as the self-report data from the present study. One possible inference that might be drawn from the similarity between the behavioral and self-report data is that either method may be used to assess the performance of checkers, that the measures index the same underlying construct. If so, it might be argued that the self-report method is preferable because questionnaires tend to be standardized and are easier to administer. However, in view of the data to be discussed below, we believe these inferences are not warranted, that the behavioral and self-report measures are sensitive to subtly different influences. More importantly, our new behavioral data provide the first direct evidence of checkers’ impaired habitual prospective memory task performance. They show that checkers have trouble performing the types of routine tasks that they tend to check (e.g., locking doors), thereby supporting the theory that checking compulsions may develop as a compensatory reaction to a deficit in prospective memory. 119 Our second major finding concerns the dissociation we reported previously (Cuttler & Graf, 2007). The results from the present study indicate that the dissociation was an artifact of the methods used for scoring performance. Typically event-cued prospective memory task performance is assessed via failure rates while time-cued prospective memory task performance is assessed via delays in task performance, and we followed these conventions in our previous study (Cuttler & Graf, 2007). For the present study we assessed failure rates and delays in performance of both the event- and time-cued episodic prospective memory tasks and the results showed that compared to low checkers, high checkers have higher failure rates on both types of tasks and that the two groups show no differences in delays in performing either type of task. The two groups even demonstrated a similar pattern of clock checking in anticipation of the need to perform the time-cued task. These results underscore the importance of using commensurate methods to assess event- and time-cued episodic prospective memory task performance. Of primary relevance, they suggest that checkers do not have problems with punctuality or with monitoring time. Rather than being late in performing tasks checkers are simply more likely to completely fail to perform a task in the future. Additional evidence that the dissociation we previously reported is an artifact of the different scoring methods comes from the self-report data. To our knowledge the TCPMQ is the only self-report instrument available for assessing various aspects of time-cued prospective memory task performance. We developed the inventory for the present study, to complement the currently available inventories (i.e., the PMQ and PRMQ) which primarily assess various aspects of event-cued prospective memory task performance. The results from all three inventories — the TCPMQ, PMQ and PRMQ — are consistent with one another indicating that compared to low checkers, high checkers more frequently fail to perform a variety of time- and event-cued prospective memory tasks. The results from the punctuality subscale of the TCPMQ indicate that high checkers do not have problems with punctuality. Thus, the pattern of findings from the self-report and behavioural measures of time- and event-cued prospective memory converge, each showing that high checkers are more likely to fail time- and event-cued tasks but when they do remember to perform tasks they perform them on time. Do self-report and behavioural measures of prospective memory index the same underlying construct? The self-report and behavioural data discussed thus far are entirely consistent with each other, suggesting that the methods are comparable and that each indexes the same underlying construct. However, the data from our follow up partial correlation analyses indicate that the behavioral and self-report measures are sensitive to subtly different influences, more specifically, that the self-report measures pick up more than just problems with memory. 120 The results from the partial correlation analyses on the behavioural measures of prospective memory indicate that there are direct links between checking compulsions and habitual and episodic prospective memory task performance that are independent of checkers’ increased levels of depression, anxiety and distractibility. However, the results from the partial correlation analyses on the self-report data conflict with these findings as well as with findings from a previous study (Cuttler & Graf, 2008). They indicate that depression accounts for the relationship between the severity of checking compulsions and self-reports of episodic prospective memory performance, that anxiety accounts for the relationship between the severity of checking compulsions and self-reports of habitual and episodic prospective memory performance and that distractibility accounts for the relationship between the severity of checking compulsions and self-reports on all of the various subscales assessing memory performance in everyday life. The conflicting results of the behavioural and self-report measures suggest that self- reports may reflect more than just actual abilities; that problems with depression, anxiety and distractibility may to some extent bias participants’ self-reports of prospective memory performance in everyday life. Indeed, depression, anxiety and distractibility each showed higher correlations with the self-report measures than with the behavioural measures of prospective memory, indicating that each influences self-reports to a greater extant than actual performance. Depression may bias self-reports in part because of the negative self-esteem and self-criticism associated with depression. Depressed individuals may also remember more of their failures because of mood congruency effects (Eich, Macaulay & Lam, 1997). Anxiety is resource demanding (Eysenck, 1979) and may compromise the resources that participants require to remember their performance on previous prospective memory tasks. Alternatively anxiety may direct individuals’ attention to past failures. Finally, distractibility may contribute to a sense of incompletion because distractible individuals may jump from one task to another rather than focusing on one task until it is complete. This sense of incompletion may contribute to higher ratings of prospective memory failures in everyday life. We have previously used results from self-report measures of prospective memory as an index of participants’ confidence in their prospective memory (Cuttler & Graf, 2007, 2008). The disparate findings between the self-report and behavioural measures in the present study support that interpretation. Individuals’ ratings of their confidence in prospective memory are likely a function of their actual memory abilities and other factors that influence confidence (e.g., depression). Thus, the findings indicate that depression, anxiety and distractibility account for the relationships between the severity of checking compulsions and various aspects of 121 participants’ confidence in prospective memory but that there is a direct link between checking compulsions and actual prospective memory task performance. Conclusions The results of this study provide additional support for the theory that a deficit in prospective memory contributes to the development and maintenance of checking compulsions. We suggest that checkers’ increased experience with prospective memory failures leads them to distrust their ability to perform certain tasks and that this distrust results in intrusive doubts that specific tasks were not completed and ultimately to the compulsion to check. In other words if individuals frequently forget to perform tasks they may develop intrusive doubts about failures and a strategy of checking to ensure that important tasks don’t go completely undone. The theory that checking compulsion may develop as a compensatory reaction to a deficit in prospective memory has important treatment implications. The stigma associated with mental illness prevents many individuals from seeking treatment. Our research reframes compulsive checking as a reaction to a neuropsychological impairment. This perspective removes some of the stigma associated with compulsive checking and in doing so may increase the likelihood that individuals will seek treatment. In addition our findings may assist in the development of more effective treatment interventions. Thus far they suggest that augmenting current treatment practices with prospective memory enhancing and confidence boosting techniques may reduce the frequency of checking compulsions and therefore improve checkers’ quality of life. We chose to use undergraduate students, with a wide range of sub-clinical checking compulsions, as participants for the present study for a number of reasons. First, we chose to focus on sub-clinical checkers because a number of factors may confound the comparison of clinical checkers and healthy controls, including psychiatric status, medication status and comorbid diagnoses. Comparisons of high and low sub-clinical checkers are less confounded by these factors (Gibbs, 1996). Second, we focused on sub-clinical checkers because they are far more accessible than their clinical counterparts. Sub-clinical checking compulsions are 10 times more prevalent than clinical checking compulsions, with approximately 15% of the population demonstrating sub-clinical checking compulsions (Sher, et al., 1983; Stein, et al., 1997). Finally, we chose to focus on sub-clinical checking compulsions because it is generally agreed that sub-clinical and clinical checking compulsions are not fundamentally different, that checking compulsions occur on a continuum which maps the frequency of experiencing the compulsion to check, the degree to which the compulsions interfere with daily functioning and the level of distress that is caused by them (Gibbs, 1996; Rasmussen & Eisen, 2002). 122 Our sample of high checkers reported fairly severe checking compulsions. The high checkers’ mean Padua Inventory checking subscale score was 18.47 which is comparable to those reported in previous research with clinical checkers (Hermans et al., 2003; TalUs, et al., 1999). Thus, we assume that the evidence we have found for the theory using this group of sub clinical checkers will extend to individuals with clinical checking compulsions. Nevertheless, this assumption requires empirical validation, It is possible that the mechanisms that underlie sub clinical and clinical checking compulsions differ. A deficit in prospective memory may contribute to the development and maintenance of the more common and less severe form of sub-clinical checking compulsions but separate factors may contribute the development and maintenance of the less common, more severe and more dysfunctional form of clinical checking compulsions. 123 References American Psychiatric Association (2000). Diagnostic and Statistical Manual of Mental Disorders (4th Edition). Washington, DC: American Psychiatric Association. Beck, A. T. (1987). Beck depression inventory (BDI). In K. Corcoran, & J. Fisher (Eds.), Measures for clinical practice: A sourcebook (pp. 107-110). New York: The Free Press. Beck, A.T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961)An inventory for measuring depression. Archives of General Psychiatry 4, 561-571. Broadbent, D. E., Cooper, P. F., FitzGerald, P., & Parkes, K. R. (1982). The cognitive failures questionnaire and its correlates. British Journal of Clinical Psychology, 21, 1-16. Craik, F. I. M. (1986). A functional account of age differences in memory. In F. Klix & H. Hagendorf (Eds.), Human memory and cognitive capabilities: Mechanisms and performances (pp. 409-422). New York: Elsevier Science. Crawford, J. R., Smith, G., Maylor, E. A., Della Sala, S., Logie, R. H. (2003). The prospective and retrospective memory questionnaire (PRMQ): Normative data and latent structure in a large non-clinical sample. Memory, 11(3), 261-275. Costa, P. T., & McCrae, R. R. (1992). Revised NEC Personality Inventory (NEC PI-R) and NEC Five-Factor Inventory (NEC-FFI): Professional manual. Cdessa, Florida: Psychological Assessment Resources. Cuttler, C., & Graf, P. (in submission). Checking-in on the memory deficit and meta-memory deficit theories of checking compulsions. Cuttler, C., & Graf, P. (2007) Sub-clinical compulsive checkers’ prospective memory is impaired. Journal of Anxiety Disorders, 21(3), 338-352. Cuttler, C., & Graf, P. (2008). Sub-clinical checking compulsions are related to impaired prospective memory independently of depression, anxiety and distractibility. Journal of Anxiety Disorders, 22(4), 642-654. Eich, E., Macaulay, E., & Lam, R. W. (1997). Mania, depression, and mood dependent memory. Cognition & Emotion, 11(5-6), 607-618. Einstein, G. C., & McDaniel, M. A. (1990). Normal aging and prospective memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 16(4), 717-726. Einstein, G. C., & McDaniel, M. A. (1996). Retrieval processes in prospective memory: Theoretical approaches and some new empirical findings. In M. Brandimonte, G. C. 124 Einstein, & M. A. McDaniel (Eds.) Prospective memory: Theory and applications (pp. 115-141). New Jersey: Lawrence Erlbaum Associates. Eysenck, M. W. (1979). Anxiety, learning and memory: A reconceptualization. Journal of Research in Personality, 13(4), 363-385. Freud, S. (1909). Notes upon a case of obsessional neurosis (the rat man’). In A. Richards (Ed.), The Penguin Freud Library: Case Histroies II. Harmondsworth: Penguin. Frost, R. 0., Marten, P., Lahart, C., & Rosenblate, R. (1990). The dimensions of perfectionism. Cognitive Therapy and Research, 14(5), 449-468. Frost, R. 0., Sher, K. J., & Geen, T. (1986). Psychopathology and personality characteristics of nonclinical compulisive checkers, Behaviour Research and Therapy, 24(2), 133-143. Gibbs, N. A. (1996). Nonclinical populations in research on obsessive-compulsive disorder: A critical review. Clinical Psychology Review, 16(8), 729-773. Graf, P. (2005). Prospective memory retrieval revisited. In N. Ohta, C. M. MacLeod, & B. UttI (Eds.), Dynamic cognitive processes (pp. 305-332). Tokyo: Springer. Graf, P. & Grodin, S. (2006). Time perception and time-based prospective memory. In Glicksohn, J. & Myslobodsky, M. S. (Eds.) Timing the future: The case for a time-based prospective memory (pp 1-24). Singapore: World Scientific Publishing. Graf, P., & Uttl, B. (2001). Prospective memory: A new focus for research. Consciousness and Cognition, 10, 437-450. Hannon, R., Adams, P., Harrington, S., Fries-Dias, C, Gibson, M. T. (1995). Effects of brain injury and age on prospective memory self-rating and performance. Rehabilitation Psychology, 40(4), 289-298. Harris, L. M., & Menzies, R. G. (1999). Mood and prospective memory. Memory, 7(1), 117-127. Hasler, G., LaSalle-Ricci, V. H., Ronquillo, J. G., Crawley, S. A., Cochran, L. W., Kazuba, D., Greenberg, B. D., & Murphy, D. L. (2005). Obsessive-compulsive disorder symptom dimensions show specific relationships to psychiatric comorbidity. Psychiatry Research, 135, 121-132. Henderson, G., & Pollard, A. (1988). Three types of obsessive compulsive disorder in a community sample. Journal of Clinical Psychology, 44(5), 747-752. Hermans, D., Martens, K., De Cort, K., Pieters, G., Eelen, P. (2003). Reality monitoring and metacognitive beliefs related to cognitive confidence in obsessive-compulsive disorder. Behaviour Research and Therapy, 41, 383-401. 125 Hewitt, P. L., & Flett, G. L. (2004). The Multidimensional Perfectionism Scale: Technical Manual. Toronto: Multihealth Systems. Janet, P. (1903). Les Obsessions et Ia Psychaesthenie. Paris: Alcan. Kliegel, M., & Jagar, T. (2006). The influence of negative emotions on prospective memory : A review and new data. International Journal of Computational Cognition, 4(1), 1-17. Kolada, J. L., Bland, R. C., & Newman, S. C. (1994). Epidemiology of psychiatric disorders in Edmonton. Obsessive-compulsive disorder. Acta Psychiatrica Scandinavica Supplementum, 376, 24-35. Kvavilashvili, L., & Ellis, J. (1996). Varieties of intentions: Some distinctions and classifications. In M. Brandimonte, G. 0. Einstein, & M. A. McDaniel (Eds.) Prospective memory: Theory and applications (pp. 23-51). New Jersey: Lawrence Erlbaum Associates. Mataix-Colis, D., Rosario-Campos, M. C., & Leckman, J. F. (2005). A multidimensional model of obsessive-compulsive disorder. American Journal of Psychiatry, 162(2), 228-238. Meacham, J. A., & Dumitru, J. (1976). Prospective remembering and external-retrieval cues. JSAS Catalog of Selected Documents in Psychology, 6 no 65, 1-12. Muller, J., & Roberts, J. E. (2005). Memory and attention in obsessive-compulsive disorder: A review. Journal of Anxiety Disorders, 19, 1-28. Rachman, S., & Hodgson, R. J. (1980). Obsessions and compulsions. New Jersey: Prentice Hall. Rachman, S., & Shafran, R. (1998). Cognitive and behavioral features of obsessive-compulsive disorder. In R. P. Swinson, M. M. Antony, S. Rachman, & M. A. Richter (Eds.) Obsessive-compulsive disorder: Theory, research, and treatment (51-78). London: The Guilford Press. Ramussen, S. A., & Eisen, J. A. (1988). Clinical and epidemiological findings of significance to neuropharmacologic rituals in OCD. Psychopharmacological Bulletin, 24, 466-470. Rude, S. S., Hertel, P. T., Jarrold, W., Covich, J., & Hedlund, 5. (1999). Depression-related impairments in prospective memory. Cognition and Emotion, 12(3), 267-276. Sanavio, E. (1988). Obsessions and compulsions: The padua inventory. Behaviour Research and Therapy, 26(2), 169-1 77. Sellen, A. J., Louie, G., Harris, J. E., Wilkins, A. J. (1997). What brings intentions to mind? An in situ study of prospective memory. Memory, 5(4), 483-507. 126 Sher, K., Frost, R. 0., & Otto, R. (1983). Cognifive deficits in compulsive checkers: An exploratory study. Behaviour Research and Therapy, 2 1(4), 357-363. Smith, G. V., Della Salla, S, Logie, R. H, & Maylor, E. A. M. (2000). Prospective and retrospective memory in normal ageing and dementia: A questionnaire study. Memory, 8,311-321. Spielberger, C. D., Gorsuch, R. L., Lushene, R., Vagg, P. R., & Jacobs, G. A. (1983). Manual for the state-trait anxiety inventory (form Y). California: Consulting Psychologists Press. Stein, M. B., Forde, D. R., Anderson, G., & Walker, J. R. (1997). Obsessive-compulsive disorder in the community: An epidemiologic survey with clinical reappraisal. American Journal of Psychiatry, 154(8), 1120-1126. Talus, F., Pratt, P., Jamani, N. (1999). Obsessive compulsive disorder, checking and non-verbal memory: A neuropsychological investigation. Behaviour Research and Therapy, 37, 161-1 66. Utti, B. (2008). Transparent meta-analysis of prospective memory and aging. PLoS ONE 3(2): el 568. doi: 10.1 371/journal.pone.0001 568 Wallace, J. C., Kass, S. J., & Stanny, C. J. (2002). The cognitive failures questionnaire revisited: Dimensions and correlates. Journal of General Psychology, 129(3), 238-256. Woods, C. M., Vevea, J. L., Chambless, D. L., & Bayen, U. (2002). Are compulsive checkers impaired in memory? A meta-analytic review. Clinical Psychology: Science and Practice, 9(4), 353-366. 127 CHAPTER 5 GENERAL DISCUSSION The main goal of this dissertation was to examine the link between checking compulsions and memory, with a primary focus on prospective memory. The impetus for examining the link between checking compulsions and prospective memory came from the observation that the types of tasks that checkers tend to become concerned with are predominantly prospective memory tasks. Following this observation we developed the prospective memory deficit theory; we reasoned that a deficit in prospective memory may contribute to the development and maintenance of checking compulsions. Concretely, if individuals frequently forget to perform prospective memory tasks they may develop intrusive doubts about whether they performed important tasks and when the perceived consequences of a failure are serious these doubts may lead to checking behaviour. As a first step toward our goal, we presented a review of the literature on memory in obsessive-compulsive disorder (OCD) (Chapter 1). While previous authors have reviewed much of this literature (Cox, 1997; Muller & Roberts, 2005; Otto, 1992; Savage, 1998; Tallis, I 995a; 1995b; 1997; Woods, Vevea, Chambless, & Bayan, 2002), ours is the first to examine whether memory deficits and meta-memory (i.e., subjective memory) deficits are unique to checkers. The review was guided by two primary objectives: 1) to examine whether there is evidence that retrospective memory deficits and/or meta-memory deficits are unique to individuals with checking compulsions and 2) to examine the evidence for the memory deficit and meta-memory deficit theories of checking compulsions in the domain of prospective memory15.The results relevant to the first objective indicated that memory and meta-memory deficits are not unique to checkers. Specifically, they showed that OCD patients with checking compulsions show no greater deficits in memory or meta-memory than OCD patients without these compulsions. These findings indicate that, rather than contributing to the compulsion to check, the retrospective memory and meta-memory deficits that checkers tend to exhibit are likely a consequence of some factor or combination of factors associated with OCD (e.g., medication, brain dysfunction, depression, anxiety). Thus, the first section of the review paper contributes to the body of research on memory deficits and meta-memory deficits in checkers, indicating that deficits in retrospective memory do not hold the power to explain the compulsion to check. The first study (Chapter 2) was intended to be exploratory; it was a first attempt to examine whether there is a link between deficits in prospective memory and checking 15 Since the results pertaining to the second objective were primarily derived from the studies presented in Chapters 2 and 3, we will discuss these results after the summaries of those chapters. 128 compulsions. The study was guided by two main objectives: 1) to examine whether checking compulsions are related to objective impairments in prospective memory and 2) to examine whether checking compulsions are related to subjective (i.e., self-reported) impairments in prospective memory. The results were encouraging, providing the first piece of evidence that there is a link between deficits in prospective memory and checking compulsions. With respect to subjective memory, the results showed that relative to low checkers, high checkers report experiencing more of a variety of prospective memory failures and using more prospective memory aiding strategies in everyday life. The results of the objective measures of prospective memory were mixed. High checkers were found to perform worse than low checkers on an event-cued episodic prospective memory task but to perform similarly to low checkers on a time-cued episodic prospective memory task. This latter finding extends previous research on prospective memory showing that event- and time-cued episodic prospective memory task performance is dissociable in various populations (Bastin & Meulemans, 2002; Einstein & McDaniel, 1995; Kliegel & Jagar, 2006; Nigro, Senese, Natullo & Sergi, 2002). More importantly, the demonstration of a link between deficits in prospective memory and checking compulsions provides some initial support for the prospective memory deficit theory, indicating that a deficit in prospective memory may contribute to the development and maintenance of checking compulsions. As the study presented in Chapter 2 was the first to demonstrate a link between deficits in prospective memory and checking compulsions, the second study (Chapter 3) was undertaken in an attempt to replicate the findings while controlling for a number of competing explanations. The study had two primary objectives: 1)to examine whether there is a direct relationship between deficits in prospective memory and checking compulsions or whether the relationship is secondary to checkers’ comorbid psychopathology and 2) to examine whether the link between deficits in prospective memory and checking compulsions is due to distractibility. The replication was successful; the results showed that checking compulsions are associated with objective and subjective deficits in prospective memory. Moreover, they showed that the link between deficits in prospective memory and checking compulsions is independent of the elevations in depression and anxiety that are associated with checking compulsions. With respect to the second objective, the results showed that checking compulsions are related to elevated levels of distractibility, that distractibility is associated with subjective but not objective deficits in prospective memory and that the increased levels of distractibility associated with checking compulsions contribute to, but do not completely account for, the relationship between subjective prospective memory deficits and checking compulsions. The results pertaining to distractibility are novel. They contribute to the characterization of checkers, showing that they have problems with distractibility, as well as to the field of prospective memory, showing that 129 distractibility is highly related to everyday life problems with prospective memory. Of primary relevance, the results provide further support for the prospective memory deficit theory, indicating that the link between deficits in prospective memory and checking compulsions is direct and is not secondary to depression, anxiety or distractibility. The prospective memory deficit theory was developed specifically to account for checking compulsions and there is no reason to believe that deficits in prospective memory contribute to the development and maintenance of any other type of obsessive-compulsive behaviour. In an attempt to examine whether prospective memory deficits are unique to checkers, the data from the first two studies (Chapters 2 and 3) were combined and the objective and subjective prospective memory of those categorized as checkers, washers and controls was compared. The results, which are described in the second section of the review paper (Chapter 1), revealed evidence for the specificity of prospective memory deficits to checkers. They showed that checkers have greater objective prospective memory deficits than controls and washers, whose performance does not differ. In addition, the results from the subjective measures indicated that checkers report experiencing more prospective memory failures than controls while washers report experiencing prospective memory failures at a similar frequency to controls. The results of direct comparisons of the subjective prospective memory of checkers and washers were less straightforward, showing that checkers only report experiencing significantly more habitual prospective memory failures than washers. These results lend increasing support to the prospective memory deficit theory. They suggest that prospective memory deficits are uniquely linked with the compulsion to check and therefore may hold the power to explain this compulsion. The final study (Chapter 4) was guided by two primary objectives: 1) to examine whether checkers are impaired on an objective (i.e., behavioural) — as opposed to subjective (i.e., self- report) — measure of habitual prospective memory and 2) to investigate the origins of the dissociation we reported between checkers’ performance on event- and time-cued episodic prospective memory tasks in the first study (Chapter 2). The impetus for examining habitual prospective memory came from the observation that the types of tasks that checkers tend to become concerned with are habitual in nature (e.g., individuals need to remember to lock the door and turn off the iron on a routine basis). The results of the study revealed a clear link between deficits in habitual prospective memory task performance and checking compulsions. With respect to the second objective we discovered that the root of the dissociation reported in the first study (Chapter 2) was the differential methods of scoring event- and time-cued episodic prospective memory. We found that checkers show higher failure rates on both event- and time cued episodic prospective memory tasks but they show no differences in the delay of 130 performing either type of task. The latter results represent an important contribution to the field of prospective memory, underscoring the importance of using commensurate methods of scoring event- and time-cued prospective memory particularly when direct comparisons between participants’ performance on the two tasks are to be made. Moreover, the results provide additional support to the prospective memory deficit theory, showing that checking compulsions are linked to objective impairments on a variety of prospective memory tasks: habitual prospective memory tasks, event-cued episodic prospective memory tasks and time- cued episodic prospective memory tasks. Aside from the two primary objectives described above, the final study (Chapter 4) was undertaken in an attempt to replicate the results of the second study (Chapter 3). That is, the final study also focused on controlling for a number of potential confounding variables. According to our theory a deficit in prospective memory contributes directly to the compulsion to check. However, we recognize that there are a number of competing explanations for the link between deficits in prospective memory and checking compulsions. For instance, checking compulsions tend to be associated with elevations in depression, anxiety and distractibility (Frost, Sher, & Geen, 1986; Hasler, et al., 2005), each of which could impair prospective memory (Harris, & Menzies, 1999; Kliegel & Jagar, 2006; Rude, Hertel, Jarrold, Covich, & Hedlund, 1999) and therefore account for the link between deficits in prospective memory and checking compulsions. To examine this possibility we examined the link between deficits in prospective memory and checking compulsions after controlling for these potential confounding factors. The results of the objective (i.e., behavioural) measures used in the third study (Chapter 4) were consistent with the results of the second study (Chapter 3), showing that the links between checking compulsions and failures on habitual and episodic prospective memory tasks are independent of depression, anxiety and distractibility. However, the results of the subjective (i.e., self-report) measures were less consistent with the results of the second study (Chapter 3). Specifically, the results of the final study (Chapter 4) indicated that depression accounts for the relationship between the severity of checking compulsions and self-reports of episodic prospective memory performance, that anxiety accounts for the relationship between the severity of checking compulsions and self-reports of habitual and episodic prospective memory performance and that distractibility accounts for the relationship between the severity of checking compulsions and self-reports on the various Prospective Memory Questionnaire (PMQ) and Prospective and Retrospective Memory Questionnaire (PRMQ) subscales which assess memory performance in everyday life. 131 One difference between the studies that may account for the discovery of alternative explanations for the link between subjective deficits in prospective memory and checking compulsions in the final study (Chapter 4) but not in the second study (Chapter 3) pertains to the methods of analysis. First, while the two techniques are nearly identical we used hierarchical regression in the second study (Chapter 3) and partial correlation in the final study (Chapter 4). Our decision to use partial correlation was guided by the advice of an expert in statistics who noted that the method is more commonly used and as such would reach a wider audience. A second, more critical, difference in the analyses is that prior to conducting the partial correlation analyses in the final study (Chapter 4) we regressed participants’ checking scores onto their washing scores and used the residuals as an index of the severity of checking compulsions. We did so because checking and washing compulsions are correlated and we wanted to remove the influence of washing compulsions on checking compulsions in order to examine checking in its purest form, in other words to ensure that any effects were not secondary to washing compulsions. To examine whether it was these changes in the methods of analysis that resulted in the discovery of alternative explanations for the link between deficits in subjective prospective memory and checking compulsions in the final study (Chapter 4), but not in the second study (Chapter 3), we reanalyzed the data from the second study. Specifically, we regressed participants’ checking scores onto their washing scores and used the residuals as an index of checking compulsions, examined the zero-order correlations between these residualized checking scores and ratings on the various questionnaire subscales and conducted the same partial correlation analyses used in the final study (Chapter 4). The zero-order correlations with the residualized checking variable are presented in Table 5.1. A comparison of these results with the results of the final study (Table 4.2) reveals smaller correlations between checking and the PMQ and PRMQ subscales in the final study (Chapter 4). It is unclear why the magnitude of the correlations varied across the two samples. Nevertheless, the correlations between checking and subjective prospective memory may have simply been too small in the final study (Chapter 4) to survive the partial correlation analyses. The results of the partial correlation analyses using the residualized checking variable are depicted in Table 5.2. A comparison of this table with Table 3.1 shows that the change in the method of analysis produced four differences in the results. The results shown in Table 3.1 indicate that the relationship between subjective deficits in memory and checking compulsions is direct and independent from depression, anxiety and distractibility while the results presented in Table 5.2 indicate that anxiety and distractibility account for four of the discovered relationships. Specifically, the results presented in Table 5.2 indicate that anxiety accounts for 132 Table 5.1. Zero-Order Correlations Between the Residualized Checking Variable and Each of Depression, State Anxiety, Trait Anxiety, Distractibility and the PMQ and PRMQ Subscales in the Sample of Participants in the Second Study (Chapter 3). Variable Checking Depression .17*1 State Anxiety .12 Trait Anxiety .26* Distractibility PMQ — Episodic Prospective Memory .26* PMQ — Habitual Prospective Memory PMQ — Internally Cued Prospective Memory PMQ — Memory Aiding Strategies PRMQ — Prospective Memory PRMQ — Retrospective Memory .21* *1 indicates p .05 (one tailed), * indicates p .05, ** indicates p .001 Table 5.2. Partial Correlations Between the Residualized Checking Variable and Ratings on the PMQ and PRMQ Subscales After Controlling for Depression, State and Trait Anxiety and Distractibility in the Sample of Participants in the Second Study (Chapter 3). Checking Checking Checking Questionnaire and Subscale (Controlling (Controlling (Controlling for for State and for Depression) Trait Anxiety) Distractibility) PMQ — Episodic Prospective Memory .23* .22* .12 PMQ — Habitual Prospective Memory .26* .27* .17*1 PMQ — Internally Cued Prospective Memory •33** .30** 23* PMQ — Memory Aiding Strategies •37** .36** 33* PRMQ — Prospective Memory .24* .23* .14 PRMQ — Retrospective Memory .16*1 .11 .05 *1 indicates p .05 (one tailed), * indicates p .05, ** indicates p .001 the relationship between checking compulsions and self-reported problems with retrospective memory. It also shows that distractibility accounts for the relationship between checking compulsions and self-reported problems with episodic prospective memory as well as with the prospective memory and retrospective memory subscales of the PRMQ. It appears that 133 removing the influence of washing compulsions on checking compulsions primarily influenced the results pertaining to retrospective memory and it produced few differences in the results pertaining to prospective memory. Thus, the difference in the methods of analysis across the two studies does not seem to account for the discrepant results between them. By reanalyzing the data from the second study (Chapter 3) in the same manner as the final study (Chapter 4) we have made it possible to make direct comparisons of the results of the two studies. These comparisons suggest that the alternative explanations reported in the final study (Chapter 4) may have been an artifact of low power. The magnitude of the zero-order correlations between subjective prospective memory and checking were reduced in the final study and they may have simply been too small to survive the partial correlation analyses. Indeed, at least with respect to depression and anxiety it was the smallest correlations which failed to survive the partial correlation analyses. With the larger zero-order correlations present in the second study (Chapter 3) the vast majority of the partial correlations between subjective prospective memory and checking compulsions remained significant. Limitations There are a number of aspects of the research described throughout this dissertation which might raise concerns. First, the sizes of many of the effects relevant to the prospective memory deficit theory were small. The correlations between checking compulsions and objective prospective memory ranged from .15 to .18, indicating that objective impairments in prospective memory only account for 2.25% to 3.24% of the variability in checking compulsions. The correlations between checking compulsions and subjective prospective memory tended to be larger, ranging from .17 to .48, indicating that subjective impairments in prospective memory account for 2.89% to 23.04% of the variability in checking compulsions. A number of factors likely contributed to reductions in the magnitude of many of the effect sizes, particularly in the magnitude of the correlations between checking compulsions and objective prospective memory. Many of the objective measures of prospective memory were binary and dichotomization reduces true effect sizes by at least 30% when the success and failure rates are equivalent (UttI, 2008). Indeed, this likely accounts for why the correlations between checking compulsions and the binary measures of objective prospective memory tended to be smaller than the correlations between checking compulsions and the continuous measures of subjective prospective memory. The binary method of scoring prospective memory task performance also likely contributed to the presence of near ceiling effects on several of the objective measures of prospective memory. For instance, in the final study (Chapter 4), over 90% of participants 134 succeeded on the time-cued episodic prospective memory tasks and nearly 95% of participants succeeded on the event-cued episodic prospective memory tasks. Ceiling effects are a common, although often ignored, problem in prospective memory research which severely reduce the power to find significant effects. Effect sizes are reduced by 70% or more when scores are dichotomous and success rate is 90% (UttI, 2008). Thus, the binary method of scoring objective prospective memory and the presence of ceiling effects on several of these tasks likely dramatically reduced the size of the effects we reported with objective prospective memory. Our choice to use undergraduate students as participants also likely contributed to the low effect sizes reported throughout this dissertation. First, the use of highly functioning undergraduate students would have contributed to the presence of ceiling effects. Second, our choice to use sub-clinical rather than clinical samples may have reduced the range of severity of checking compulsions and therefore our power to find correlations with checking compulsions (McNemar, 1969). Taken together it seems likely that the correlations we reported underestimate the true size of the relationship between deficits in prospective memory and checking compulsions. Future research using more demanding prospective memory tasks and a patient sample may provide a more accurate index of the magnitude of the correlations between deficits in prospective memory and checking compulsions. Our use of undergraduate sub-clinical checkers as participants, rather than clinical checkers, represents an additional issue which may raise concerns. However, our choice to focus on sub-clinical checkers was guided by a number of factors including practical considerations. Sub-clinical checking compulsions are far more prevalent than clinical checking compulsions, with estimates of approximately 15% of the population demonstrating sub-clinical checking compulsions (Sher, Frost, & Otto, 1983; Stein, Forde, Anderson & Walker, 1997). Thus, our use of sub-clinical checkers permitted us to obtain large samples of participants in a reasonable period of time. Our choice to focus on sub-clinical checkers was also informed by evidence that sub clinical and clinical checking compulsions are not fundamentally different. Checking compulsions appear to occur on a continuum and it is generally believed that what distinguishes sub-clinical from clinical compulsions are the frequency with which they are experienced, the degree to which they interfere with daily functioning and the distress that they cause (Rasmussen & Eisen, 2002). Sub-clinical and clinical checkers show similar symptom profiles, personality and psychological characteristics, coping strategies, cognitive dysfunction and comorbid psychopathology with cognitive deficits and comorbid psychopathology proportional to the degree of obsessive-compulsive symptoms (Gibbs, 1996). 135 As indicated above, sub-clinical checkers demonstrate levels of comorbid psychopathology and cognitive deficits intermediate to those of clinical checkers and healthy controls (Gibbs, 1996), thus concerns with the confounding influence of these variables are attenuated by the use of sub-clinical checkers. Concerns with the influence of a number of additional potential confounds including psychiatric status, medication status and potential brain dysfunction are also assuaged by the use of sub-clinical rather than clinical checkers. Our samples of sub-clinical checkers reported fairly severe checking compulsions. The high checkers’ mean Padua Inventory checking subscale scores ranged from 17.97 to 18.47 which is comparable to those reported in previous research with clinical checkers (Hermans, Martens, De Cort, Pieters, Eelen, 2003; Tallis, Pratt, & Jamani, 1999), indicating that many of the sub-clinical checkers in our studies may meet diagnostic criteria for obsessive-compulsive disorder (OCD) but have yet to seek treatment. In light of all of these considerations, we assume that the prospective memory deficits we have found in our samples of sub-clinical checkers will extend to clinical samples. Nevertheless, this assumption requires empirical validation. Future Directions The results of the studies described in this dissertation open a number of avenues for future research. First, future research should focus on extending the findings to a clinical population. While we assume that the findings will extend to a clinical population, it is possible that the mechanisms that underlie sub-clinical and clinical checking compulsions differ. A deficit in prospective memory may contribute to the development and maintenance of the more common and less severe form of sub-clinical checking compulsions but separate factors may contribute to the less common, more severe and more dysfunctional form of clinical checking compulsions. Due to the numerous potential confounds inherent in comparisons of clinical patients with healthy controls (e.g., psychiatric status, medication, brain dysfunction, comorbid psychopathology), it is important that researchers carefully control for these variables. We recommend that they use OCD patients without checking compulsions as a control group. Future research should also focus on examining how memory deficits and meta-memory deficits interact with other factors believed to contribute to OCD. For instance, prevailing cognitive-behavioural models have emphasized the role of dysfunctional beliefs in the etiology and maintenance of OCD (Clark, 2004; Frost & Steketee, 2002; Obsessive Compulsive Cognitions Working Group, 2005; Salkovskis, 1996). OCD patients have been shown to be perfectionists, to be intolerant of uncertainty, to have an inflated sense of responsibility and to overestimate the likelihood of threat (Obsessive Compulsive Cognitions Working Group, 2005). Not all OCD patients hold these dysfunctional beliefs (Taylor, 2006) and certainly not everyone 136 who has a deficit in prospective memory develops checking compulsions. It seems unlikely that someone with a prospective memory deficit would develop checking compulsions if they did not worry about their frequent prospective memory failures, feel responsible for them or perceive their consequences to be serious. Rather, it seems likely that only those individuals who have a prospective memory deficit and who hold dysfunctional beliefs will develop checking compulsions. Concretely, if someone has frequent prospective memory failures which they feel are unacceptable due to perlectionistic standards and they believe that failures could result in catastrophic outcomes for which they would be responsible (e.g., “I will be responsible for the fire that will result if I left my iron on”) then checking compulsions seem highly likely to ensue. Similarly, future research should focus on using a longitudinal design in order to determine whether individuals with deficits in prospective memory are more susceptible to developing checking compulsions as opposed to other forms of OCD (e.g., washing compulsions, hoarding compulsions). Rather than increasing the probability of developing OCD, it seems likely that a deficit in prospective memory contributes to shaping the expression of the disorder to checking compulsions. As OCD most commonly develops in late adolescence and early twenties (Talus 1995a), researchers could focus on recruiting a large sample of participants in their early adolescence. Yearly assessments could focus on participants’ risk factors for developing QCD (e.g., family history of OCD, dysfunctional beliefs), their objective and subjective prospective memory and their obsessive-compulsive behaviours in order to determine whether and how these factors interact to produce checking compulsions. For instance, individuals who are at risk for developing OCD and who exhibit deficits in prospective memory seem the most likely to develop checking compulsions. In contrast, those who are at risk for developing OCD but who do not demonstrate deficits in prospective memory seem more likely to exhibit other forms of OCD, such as washing compulsions. Contributions and Implications The major contribution of this dissertation is the re-contextualization of the memory deficit and meta-memory deficit theories of compulsive checking in the domain of prospective memory. The results described in the dissertation show that there is little evidence for either of the theories in the domain of retrospective memory. More importantly, the results demonstrate promising evidence for the memory and meta-memory deficit theories in the domain of prospective memory. Specifically, they show that checkers have problems performing the types of tasks for which they tend to become concerned about and feel compelled to check. Moreover, they show that checkers’ deficits in prospective memory are specific; washers do not demonstrate deficits in objective or subjective prospective memory. Finally, they show that the link between checking compulsions and prospective memory is direct and is not secondary to 137 depression, anxiety or distractibility. It is hoped that the insights gained from this research will guide investigators of the memory deficit and meta-memory deficit theories to focus on prospective memory. The research described in this dissertation also has practical implications. The stigma associated with mental illness prevents many individuals from seeking treatment. This reluctance is exacerbated in OCD because, as compared to other mental illnesses (e.g., schizophrenia), OCD tends to be associated with high levels of insight. Patients’ belief that their compulsions are irrational can lead to feelings of embarrassment and shame which can contribute to their reluctance to discuss their problems with others and to seek professional help. Indeed, a World Health Organization study suggests that over 57% of individuals with COD are untreated (Kohn, Saxena, Levav, & Saraceno, 2003). Our discovery of a direct and specific link between deficits in prospective memory and checking compulsions is somewhat discordant with current perceptions of OCD as being entirely irrational (Rachman & Shafran, 1998; Rachman 2002). If an individual’s prospective memory is unreliable then checking may be a reasonable response to ensure that important tasks are completed. This perspective may help to remove some of the stigma associated with compulsive checking and thereby increase the likelihood that individuals will seek treatment. For those individuals who do seek professional help, the most common and effective treatments are pharmacological and behavioral. Pharmacological interventions involve serotonin transport inhibitors (Greist, Jefferson, Kobak, Katzelnick, & Serlin, 1995; Pigott & Seay, 1998), and the most common behavioral intervention, cognitive behavioral therapy, tends to focus on exposure and response prevention (see Foa, Frankliin & Kozak, 1998; Rowa, Antony & Swinson, 2007 for reviews). However, up to 50% of COD patients do not respond to these treatments (Bolton, Raven, Madronal-Luque, & Marks, 2000; Hafner, Gilchrist, Bowling, & Kalucy, 1981) and full remission of symptoms is rare even in those who do respond (Tallis, 1995). The results of the research described in this dissertation may eventually lead to the development of more effective treatment interventions. Thus far they suggest that augmenting current treatment practices with prospective memory enhancing and confidence boosting techniques may reduce the frequency of checking compulsions and therefore improve checkers’ quality of life. 138 References Andrews, G., Steward, G., Allen, R., & Henderson, A. S. (1990). The genetics of six neurotic disorders: A twin study. Journal of Affective Disorders, 19, 23-29. Bastin, C, & Meulemans, T. (2002). Are time-based and event-based prospective memory affected by normal aging in the same way? Current Psychology Letters: Behavior, Brain & Cognition, 7, 105-121. Bolton, D., Raven, P., Madronal-Luque, P., & Marks, I. M. (2000). Neurological and neuropsychological signs in obsessive-compulsive disorder: Interaction with behavioural treatment. Behaviour Research and Therapy, 38, 695-708. Cary, G., & Gottesman, I. I. (1981). Twin and family studies of anxiety, phobic and obsessive disorders. In D. G. Klein & J. Rabkin, (Eds), Anxiety: New research and changing concepts. New York: Raven. Clark, D. A. (2004). Cognitive-behavioral therapy for OCD. New York: Guilford Press. Ccx, C. S. (1997). Neuropsychological abnormalities in obsessive-compulsive disorder and their assessments. International Review of Psychiatry, 9, 45-59. Einstein, G. 0., & McDaniel, M. A. (1995). Aging and prospective memory: Examining the influences of self-initiated retrieval processes. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21(4), 996-1007. Foa, E. B., Franklin, M. E., & Kozak, M. J. (1998). Psychosocial treatments for obsessive- compulsive disorder: Literature review. In R. P. Swinson, M. M. Antony, S. Rachman, & M. A. Richter (Eds.), Obsessive-compulsive disorder: Theory, research and treatment. New York: The Guilford Press. Frost, R. 0., & Steketee, G. (2002). Cognitive approaches to obsessions and compulsions: Theory, assessment and treatment. Oxford: Elsevier Gibbs, N. A. (1996). Nonclinical populations in research on obsessive-compulsive disorder: A critical review. Clinical Psychology Review, 16(8), 729-773. Greist, J. H., Jefferson, J. W., Kobak, K. A., Katzelnick, D. J., & Serlin, R. C. (1995). Efficacy and tolerability of serotonin transport inhibitors in obsessive-compulsive disorder: A meta-analysis. Archives of General Psychiatry, 52(1), 53-60. Hafner, R. J., Gilchrist, P., Bowling, J., & Kalucy, R. (1981). The treatment of obsesssional neurosis in a family setting. Austratlian and New Zealand Journal of Psychiatry, 15, 145- 151. 139 Hermans, D., Martens, K., De Cort, K., Pieters, G., Eelen, P. (2003). Reality monitoring and metacognitive beliefs related to cognitive confidence in obsessive-compulsive disorder. Behaviour Research and Therapy, 41, 383-401. Kessler, R. C., McGonagle, K. A., Zhao, S., Nelson, C. B., Hughes, M., Eshleman, S., Wittchen, H., & Kendler, K. S. (1994). Lifetime and 12-month prevalence of DSM-Ill-R psychiatric disorders in the United States. Archives of General Psychiatry, 51, 8-19. Kliegel, M., & Jagar, T. (2006). The influence of negative emotions on prospective memory : A review and new data. International Journal of Computational Cognition, 4(1), 1-17. Kohn, R., Saxena, S., Levav, I., & Saraceno, B. (2003). The treatment gap in mental health care. Bulletin of the World Health Organization, 82(11), 858-866. Lenane, M. C., Swedo, S. E., Leonard, H., Pauls, D. L., Sceery, W., & Rapoport, J. L. (1990). Psychiatric disorders in first degree relatives of children and adolescents with obsessive- compulsive disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 29, 407-412. McKay, D., Abramowitz, J. S., Calamari, J. E., Kyrios, M., Sookman, D., Taylor, S., & Wilhelm, 5. (2004). A critical evaluation of obsessive-compulsive disorder subtypes: Symptoms versus mechanisms. Clinical Psychology Review, 24, 283-313. McNemar, Q. (1969). Psychological statistics (4th ed.) New York : Wiley. Muller, J., & Roberts, J. E. (2005). Memory and attention in obsessive-compulsive disorder: A review. Journal of Anxiety Disorders, 19, 1-28. Nigro, G., Senese, V. P., Natullo, 0., Sergi, I. (2002). Preliminary remarks on the type of task and delay in children’s prospective memory. Perceptual and Motor Skills, 95(2), 515- 519. Obsessive Compulsive Cognitions Working Group (2005). Psychometric validation of the Obsessive Beliefs Questionnaire and the Interpretations of Intrusions Inventory: Part 2. Behaviour Research and Therapy, 43(11), 1527-1542. Otto, M. W. (1992). Normal and abnormal information processing: A neuropsychological perspective on obsessive compulsive disorder. Psychiatric Clinics of North America, 15(4), 825-848. Pigott, T. A., & Seay, S. (1998). Biological treatments for obsessive-compulsive disorder. In R. P. Swinson, M. M. Antony, S. Rachman, & M. A. Richter (Eds.), Obsessive-compulsive disorder: Theory, research and treatment. New York: The Guilford Press. 140 Rachman, S., & Shafran, R. (1998). Cognitive and behavioral features of obsessive-compulsive disorder. In R. P. Swinson, M. M. Antony, S. Rachman, & M. A. Richter (Eds.) Obsessive-compulsive disorder: Theory, research, and treatment (51-78). London: The Guilford Press. Rasmussen, S. A., & Eisen, J. L. (2002). The course of clinical features of obsesssive compulsive disorder. In K. L. Davis, D. Charney, J. T., Coyle, & C. Nemeroff (Eds.), Neuropsychoparmacology: The fifth generation of progress. Philadelphia: American College of Neuropsychopharmacology. Riddle, M. A., Scahill, L., King, R., Hardin, M. T. Twobin, K. E., Ort, S. I., Leckman, J. F., & Cohen, D. J. (1990)Obsessive compulsive disorder in children and adolescents: Phenomenology and family history. Journal of the American Academy of Child and Adolescent Psychiatry, 29, 766-772. Robins, L. N., & Regier, D. A. (Eds.). (1991). Psychiatric disorders in America: The epidemiological catchment area study. New York: The Free Press. Rowa, K., Antony, M. M., & Swinson, R. P. (2007). Exposure and response prevention. In M. M. Antony, C. Purdon, & L. J. Summerfeldt (Eds.), Psychological treatment of obsessive- compulsive disorder: Fundamentals and beyond. Washington: American Psychological Association. Salkovskis, P. M. (1996). Cognitive-behavioral approaches to the understanding of obsessional problems. In R. M. Rapee (Ed.), Current controversies in the anxiety disorders (pp. 103- 134). New York: Guilford Press. Savage, C. R. (1998). Neuropsychology of obsessive-compulsive disorder: Research findings and treatment implications. In M. A. Jenike, L. Baer, & W. E. Minichiello (Eds.), Obsessive-Compulsive Disorders: Practical Management (3rd ed.). St. Louis: Mosby. Shapiro, S., Skinner, E. A., Kessler, L. G., Vonkorff, M., German, P. S., Tischler, G. L., Leaf, P. J., Benham, L., Cottler, L. & Regier, D. A. (1984). Utilization of health and mental health services. Archives of General Psychiatry, 41, 971-978. Sher, K., Frost, R. 0., & Otto, R. (1983). Cognitive deficits in compulsive checkers: An exploratory study. Behaviour Research and Therapy, 2 1(4), 357-363. Stein, M. B., Forde, D. R., Anderson, G., & Walker, J. R. (1997). Obsessive-compulsive disorder in the community: An epidemiologic survey with clinical reappraisal. American Journal of Psychiatry, 154(8), 1120-1126. 141 Talus, F. (1995a). Obsessive compulsive disorder: A cognitive and neuropsychological perspective. Chichester: John Wiley & Sons. Tallis, F. (1995b). The characteristics of obsessional thinking: Difficulty demonstrating the obvious? Clinical Psychology and Psychotherapy, 2(1), 24-39. Tallis, F. (1997). The neuropsychology of obsessive-compulsive disorder: A review and consideration of clinical implications. British Journal of Clinical Psychology, 36, 3-20. Tallis, F., Pratt, P., Jamani, N. (1999). Obsessive compulsive disorder, checking and non-verbal memory: A neuropsychological investigation. Behaviour Research and Therapy, 37, 161-1 66. Taylor, S. et al. (2006). Do dysfunctional beliefs play a role in all types of obsessive-compulsive disorder? Journal of Anxiety Disorders, 20, 85-97. Torgerson, S. (1983). Genetic factors in anxiety disorder. Archives of General Psychiatry, 40, 1085-1 089. Uttll, B. (2008). Transparent meta-analysis of prospective memory and aging. PLoS ONE, 3(2): el 568. doi:1 0.1 371/journal.pone.0001 568. Woods, C. M., Vevea, J. L., Chambless, D. L., & Bayan, U. J. (2002). Are compulsive checkers impaired in memory? A meta-analytic review. Clinical Psychology: Science and Practice, 9(4), 353-366. 142 IJBC The University of British Columbia Office of Research Services Behavioural Research Ethics Board Suite 102, 6l9oAgronomy Road, Vancouver, B.C. V6T 1Z3 CERTIFICATE OF APPROVAL - MINIMAL RISK AMENDMENT PRINCIPAL INVESTIGATOR: DEPARTMENT: UBC BREB NUMBER: eter Graf UBC/Arts/Psychology, Department of H03-80566 NSTITUTION(S) WHERE RESEARCH WILL BE CARRIED OUT: Institution I Site JBC Point Grey Site)ther locations where the research will be. conducted: nla DO-INVESTIGATOR(S): Evgeniya Berger Jess Gao )aniel Slu Darrie Cuttler Michael Dodd Lindsay Nagamatsu SPONSORING AGENCIES: ‘datural Science Engineering Research Council - “Individual Differences in Memory” - “Prospective Memory Across the AdultJfespan” - “Individual Differences in Memory: Time Flies” PROJECT TITLE: Individual Differences in Memory: Just Checking Expiry Date - Approval of an amendment does not change the expiry date on the current UBC BREB approval of this study. An application for renewal is required on or before: August 24, 2007 MENDMENT(S): IAMENDMENT APPROVAL DATE: IFebruary 12, 2007 Document Name I Version I Date onsent Forms: onsent Form 3 January 12, 2007 Ehe amendment(s) and the document(s) listed above have been reviewed and the procedures were found to be acceptable n ethical grounds for research involving human subjects. Approval is issued on behalf of the Behavioural Research Ethics Board and signed electronically by one of the following: Dr. Peter Suedfeld, Chair Dr. Jim Rupert, Associate Chair Dr. Arminee Kazanjian, Associate Chair Dr. M. Judith Lynam, Associate Chair 143

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0066985/manifest

Comment

Related Items