"Education, Faculty of"@en . "Educational and Counselling Psychology, and Special Education (ECPS), Department of"@en . "DSpace"@en . "UBCV"@en . "Strangway, Carrie Lynn"@en . "2009-12-16T21:01:00Z"@en . "2005"@en . "Master of Arts - MA"@en . "University of British Columbia"@en . "This study examined the sensitivity of a computerized neuropsychological screening battery\r\n(ImPACT) to the cognitive effects of ADHD in a sample of 68 young adults with ADHD and\r\n68 healthy students matched for age, education, gender, and history of head injury. Students\r\nwith ADHD self-reported more cognitive difficulties on the Post-Concussion Scale of ImPACT\r\n(p < .005, d = .68, medium-large effect size), and performed more poorly on the Memory\r\nComposite (p < .005, d = .50, medium effect size). The two groups did not differ significantly\r\non the Processing Speed Composite or the Impulse Control Composite. There was a\r\nnonsignificant trend for the individuals with ADHD to display slower reaction times (p < .076,\r\nd = .33, small effect size). This is the second study using fmPACT in ADHD research. The\r\nbrevity and sensitivity of fmPACT to the cognitive effects of ADHD warrants further research\r\nwith this population."@en . "https://circle.library.ubc.ca/rest/handle/2429/16827?expand=metadata"@en . "C O M P U T E R I Z E D S C R E E N I N G IN A D O L E S C E N T S A N D Y O U N G A D U L T S W I T H A D H D by Carrie Lynn Strangway B . A . , University of Western Ontario, 1996 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L M E N T OF T H E R E Q U I R E M E N T S F O R T H E D E G R E E OF M A S T E R OF A R T S in T H E F A C U L T Y OF G R A D U A T E S T U D I E S (School Psychology) T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A August 2005 \u00C2\u00A9 Carrie L y n n Strangway, 2005 Abstract This study examined the sensitivity of a computerized neuropsychological screening battery ( ImPACT) to the cognitive effects of A D H D in a sample of 68 young adults with A D H D and 68 healthy students matched for age, education, gender, and history of head injury. Students with A D H D self-reported more cognitive difficulties on the Post-Concussion Scale of I m P A C T (p < .005, d = .68, medium-large effect size), and performed more poorly on the Memory Composite (p < .005, d = .50, medium effect size). The two groups did not differ significantly on the Processing Speed Composite or the Impulse Control Composite. There was a nonsignificant trend for the individuals with A D H D to display slower reaction times (p < .076, d = .33, small effect size). This is the second study using f m P A C T in A D H D research. The brevity and sensitivity of f m P A C T to the cognitive effects of A D H D warrants further research with this population. Table of Contents Page Abstract i i Table o f Contents i i i List of Tables , iv List of Figures v Acknowledgments v i Introduction 1 Literature Review . 3 Overview of Attention Deficit Hyperactivity Disorder 3 Neuropsychological Functioning in adults and adolescents with A D H D 7 Summary of Results on Specific Neuropsychological Tests . 20 Tests of Executive Functions 20 Language Skills Tests 26 Learning and Memory Tasks..... 27 Tests of Intelligence 29 Summary 30 Rationale for the Current Study 31 Immediate Post-Concussion Assessment and Cognitive Testing ( ImPACT) 33 Hypotheses 39 Methodology 42 Participants 42 Procedure 43 Measures , 43 Module 1 Word Discrimination 46 Module II Symbol Memory & Module III Color Cl ick 47 Module IV Symbol Match 48 Module V Color-Word Match 49 Module V I Sequential Digit Tracking/Trigram Memory ; 50 Module VII Visual Attention Span 50 I m P A C T composite scores 51 Analyses \u00E2\u0080\u00A2 53 Results 53 Discussion 58 Conclusions 63 Limitations 65 Future Directions 67 References 69 i i i List of Tables Page Table 1. Summary of Studies Reporting on the Neuropsychological Performance of A D H D Adults 9 Table 2. Number of Studies Reporting Significant differences between Adults with A D H D , n o n - A D H D controls, and patients with psychiatric disorders, on commonly administered neuropsychological tests..... 19 Table 3. Post-Concussion Scale 45 Table 4. Descriptive statistics for the I m P A C T composite scores 54 Table 5. Pearson's Correlation coefficients among the composite scores 55 Table 6. Descriptive statistics, significance tests, and effect sizes (Cohen's d) 56 Table 7. Percentages and cumulative percentages o f subjects with unusual scores 57 List of Figures Page Figure 1. Overlapping distributions based on effect sizes (using the IQ metric) 63 Acknowledgments This thesis was conducted as partial fulfillment of the primary author's Master of Art 's degree in Educational and Counselling Psychology. I thank Dr. Shelley Hymel and Dr.Grant Iverson for their supervision with the content and completion of this study. I also thank Dr. Nicholas Bogod, Dr.Tracey Brickel l , Dr. Rael Lange, and Dr. Todd Woodward, for their guidance in writing the manuscript. I would also like to thank Jennifer Bernardo for her friendship and assistance with preparing the final manuscript. I thank my Mother, Linda Nixon, for believing in me and helping me to believe in myself. Finally, I thank my Father, Dr. J.F. Strangway, for both the love he has given me, and the sacrifices he has made to provide me with the support to achieve this goal. I aspire to carry out my life with as much dignity and selflessness as he has shown me. v i Introduction Attention deficit hyperactivity disorder ( A D H D ) is recognized as a serious neurobehavioral disorder that arises in early childhood and is prevalent throughout the lifespan (American Psychiatric Association, 2000; Barkley, 1997; Culbertson & Kru l l , 1996; Epstein, Johnson, Varia, & Conners, 2001). Neuropsychological problems in children with A D H D have been well documented (e.g., Halperin et al., 1990; Konrad, Gauggel, Manz, & Scholl, 2000; Kupietz, 1990; Loge, Staton, & Beatty, 1990; Seidel & Joschko, 1990; van der Meere & Sergeant, 1988). In the past, the disorder was thought to subside in adolescence (Culbertson & K r u l l , 1996; Johnson et al., 2001), and as a result the neuropsychological deficits associated with A D H D in adults are less well documented. Adults with A D H D have been shown to perform significantly more poorly on a variety of cognitive tests measuring attention, memory, reaction time, processing speed, set shifting (i.e., the ability to shift from one cognitive task to another), inhibition, and problem solving (Epstein et al., 2001; Murphy, Barkley, & Bush, 2001; Murphy, 2002b; Seidman, Biederman, Weber, Hatch, & Faraone, 1998; Walker, Shores, Trollor, Lee, & Sachdev, 2000; Woods, Lovejoy, Stutts, Ba l l , & Fals-Stewart, 2002). However, these studies have been criticized for their lack of consistency in differentiating adults and adolescents with A D H D from controls (Corbett & Stanczak, 1999; Kovner et al., 1998; Murphy, 2002a; Schmitz et al., 2002; Seidman, Biederman, Faraone, Weber, & Ouellette, f 997). A s a result, the existing literature examining cognitive functioning in young adults with A D H D is not conclusive, and further research with more refined measures of cognitive functioning is needed. Woods, Lovejoy, and Ba l l (2002) suggest that the use of new neuropsychological assessment tools incorporating multiple cognitive constructs that assess a broad array of 1 attentional and executive functions is needed. Many of these tests measure the frontal-. subcortical circuit dysfunction proposed to underlie A D H D (Faraone & Biederman, 1998). Implementing these measures in A D H D research may advance the understanding of the neurocognitive deficits associated with A D H D in adults. The present thesis builds upon previous research on adults with A D H D by utilizing a new assessment tool called I m P A C T (Immediate Post-Concussion Assessment and Cognitive Testing; Lovel l et al., 2003; Maroon et al., 2000) in a young adult A D H D sample. I m P A C T is a computerized, self-administered, 25-minute neuropsychological test battery composed of seven individual test modules that are used to measure five aspects of cognitive functioning: attention, memory, reaction time, processing speed, and impulse control. The purpose of this study is to examine the sensitivity of this computerized neuropsychological screening battery for distinguishing the cognitive effects of A D H D in young adults from the performance of non-A D H D controls. Specifically, the present study compares young adults with an A D H D diagnosis (as determined by self-report) to n o n - A D H D controls matched for education, gender, and history of head injury, to determine whether or not there are cognitive differences between the two groups in terms of their memory, reaction time, processing speed, and impulse control. I m P A C T also includes a symptom questionnaire. O f additional interest is whether or not there are differences in self-reported cognitive difficulties, such as perceived concentration and memory functioning, between the A D H D sample and the matched n o n - A D H D controls. If I m P A C T proves sensitive to cognitive problems associated with A D H D in young adults, it might be a useful measure in research and clinical practice due to its ease and rapidity of administration. It might also be useful as a primary outcome variable in the evaluation of the efficacy of pharmacological interventions on cognitive symptoms associated with A D H D . 2 Literature Review Overview of Attention Deficit Hyperactivity Disorder Attention Deficit Hyperactivity Disorder ( A D H D ) is the one of the most common emotional, cognitive, and behavioral disorders diagnosed and treated in youth. A D H D affects approximately 3-5% o f school age children with a male: female gender ratio of 3:1 (American Psychiatric Association, 2000; Barkley, 1997; Culbertson & K r u l l , 1996; Schachar, Mota, Logan, Tannock, & K l i m , 2000; Wilens, Biederman, & Spencer, 2002). The disorder emerged in the early 1900's medical literature, with researchers describing aggressive and defiant children as having poor volitional inhibition and defective moral regulation of behavior (Barkley, 1997). Under the diagnostic term hyperkinetic reaction of r childhood, it first appeared in the second edition of the Diagnostic and Statistical Manual of Mental Disorders ( D S M ; American Psychiatric Association, 1968), referring to excessive activity as the primary symptom (Culbertson & K r u l l , 1996). In the early 1970's, Douglas (1972; as cited in Schachar et al., 2000) provided a model that proposed attention deficit, rather than excessive activity, was the main feature of the disorder. This model guided future research and led to the term attention deficit disorder, which appeared in the DSM-1II (American Psychiatric Association, 1980) reflecting an attention deficit, rather than excessive activity, as the main feature of the disorder. B y the 1980's there was a movement towards distinguishing a hyperactivity component, and the term attention deficit hyperactivity disorder emerged as a separate disorder from A D D . This was adopted in the DSM-I I I -R (American Psychiatric Association, 1987). In both the D S M - I V and D S M - I V - T R (American Psychiatric Association, 1994, 2000) A D H D is divided 3 into three primary subtypes: predominantly inattentive type, predominantly hyperactive-impulsive type, and combined type. A D H D is characterized by age-inappropriate levels of inattention, with or without impulsivity, and overactivity that occurs across settings and causes functional impairment. A D H D begins in childhood, and 50-70% of childhood cases persist into adulthood (Murphy & Schachar, 2000; Wilens et al., 2002). Although the observable symptoms can change in quality and quantity over time, most individuals with A D H D continue to experience some symptoms as adults (Mercugliano, 1999). Adolescents and adults with A D H D are at a higher risk for academic problems, poor peer and family relations, anxiety and depression, aggression, conduct problems, delinquency, early substance abuse, driving accidents, and unemployment (Barkley, 1997; Gallagher & Blader, 2001; Schmitz et al., 2002). A s a result, A D H D is considered a major clinical and public health problem, due to its associated morbidity and disability in individuals across all ages (Wilens et al., 2002). > The majority of researchers and clinicians agree that the primary disturbance of A D H D results via poor control over executive functions, with at least some of these executive functions linked to the frontal and sub-cortical regions of the brain (Faraone & Biederman, 1998; Gallagher & Blader, 2001; Royall et al., 2002). However, evidence to support this view has been somewhat mixed, and it is not yet clear which aspects of cognitive functioning reliably distinguish people with A D H D from those without. According to Barkley (1997), the primary behavioral characteristics of A D H D are age-inappropriate levels of inattention, impulsivity, and hyperactivity. These problems are considered to reflect difficulty with the management and executive control of behavior. Deficits in attention are not considered the result of an inability to attend, but rather a problem in the executive tasks of organising 4 attention (i.e., sustaining attention on stimuli and shifting attention). Similarly, impulsivity is not defined as a failure to control one's actions. Rather, it is a difficulty with deciding when actions should be taken, and controlling the impact and order of those actions. Lastly, hyperactivity is not seen as a result of overactivity, but as a disturbance in the executive task of controlling the appropriate situational level of arousal and activity (Gallagher & Blader, 2001). Using neuropsychological testing in children with A D H D , researchers have identified reduced performance in attention, inhibition, and executive functions compared to n o n - A D H D controls (Benezra & Douglas, 1988; Douglas & Parry, 1983; Pineda et al., 1998; Schachar et al., 2000; Seidel & Joschko, 1990). Despite these findings, the role of neuropsychological testing in children remains controversial. This is due to the inconsistent findings across neuropsychological measures, and concerns regarding inadequate positive and negative predictive power of the cognitive measures. These may be partially attributable to methodological limitations of the studies (Kempton et al., 1999). In addition, childhood A D H D might be characterized by more overt behavioral characteristics (i.e., hyperactivity), rather than cognitive deficits in attention and executive functioning. The behavioral symptoms of A D H D subside as children mature, and signs of excessive gross motor activity are less common in early adolescence (American Psychiatric Association, 2000). Researchers believe that adults are more likely to exhibit cognitive deficits in attention and inhibition than the behavioral deficits that characterize children with A D H D (Barkley, 1997; Woods, Lovejoy, & Ba l l , 2002). However, the potential cognitive differences between children with A D H D and adults with A D H D have not yet been established, due to the limited research in this area. Preliminary research suggests that young adults and adults with A D H D have deficits compared to n o n - A D H D controls on measures of attention, memory, reaction time, processing 5 speed, inhibition, problem solving, and set shifting (Epstein et al., 2001; Murphy et al., 2001; Murphy, 2002b; Seidman et al., 1998; Walker et al., 2000; Woods, Lovejoy, Stutts et al., 2002). However, similar to the children's literature, these results are not consistent across tests and studies. Moreover, controversy still exists regarding the validity of A D H D as a disorder of adulthood due to the lack of a reliable set of diagnostic criteria, and the few neuropsychological tests that can clearly differentiate the putative cognitive deficits associated with the condition (Johnson et al., 2001). According to D S M - I V - T R (American Psychiatric Association, 2000), an individual ( must have had symptoms of A D H D that date back to childhood. For an adult that was not diagnosed with A D H D as a child it is not always possible to obtain this information from multiple raters or sources. A s a result, the clinician or researcher must often rely on an individual's self-report of their symptoms. Further, the age-of-onset criterion of the disorder y requires that symptoms be apparent prior to the age of seven. A n adult's ability to validly self-report past symptoms necessary to retrospectively diagnose the condition has been constantly debated in the literature (Applegate et al., 1997; Barkley & Biederman, 1997; Levin, 1998; Mota & Schachar, 2001). Murphy and Schachar (2000) explain that researchers and clinicians are often forced to rely on an individual's account of current and childhood symptom because it is often impractical or impossible to obtain this information from a former teacher, parent, or current employer. To advance research on young adults with A D H D it is necessary to determine i f they can accurately self-report past and current symptoms. Moreover, according to Woods, Lovejoy, and Ba l l (2002), new neuropsychological assessment techniques that include a comprehensive 6 battery approach (e.g., assess multiple constructs at one time) are needed to better characterize the cognitive deficits in adults with A D H D . Neuropsychological Functioning in Adolescents and Adults with ADHD O f primary interest to the present study is the role of neuropsychological tests in the identification of adolescents and adults with A D H D . Surprisingly few studies to date have examined whether neuropsychological tests of particular cognitive abilities can consistently differentiate adults with A D H D from those without the disorder. A. review o f studies designed specifically to examine the neurocognitive functioning of adolescents and adults with A D H D is provided in Table 1. The purpose of the review is to provide the reader with an overview of the current state of this literature in terms of the varied measures, diagnostic criteria, and sample selections. Due to the limited literature available in this area, all studies pertaining to this area were included. Studies were not excluded based on possible methodological weaknesses, including small sample sizes, or the use of tests with limited psychometric properties. These findings are reviewed according to the neuropsychological tests (rather than the study) and the cognitive areas measured. Most of these studies focus on attention and executive functions because much of the current research is based on the conceptualization of A D H D as a frontal-subcortical disorder (Barkley, 1997; Faraone & Biederman, 1998; Gallagher & Blader, 2001; Johnson et al., 2001; Mercugliano, 1999). The emerging neuroimaging literature suggests the presence of abnormalities in frontal networks as shown by positron emission tomography brain imaging in adults with A D H D (Wilens et al., 2002). However, empirical support for the precise neural pathways associated with A D H D remains elusive. Hence, the nature of the proposed connection between frontal lobe function 7 and actual cognitive task performance in A D H D is not yet supported by consistent data, as stated earlier in the review. Given the recency of this area of inquiry, it is not surprising that many researchers are considering many different indices of cognitive performance that (they hope or hypothesize) are loosely linked to the broad construct called \"executive functioning.' i 8 Table 1 Summary of Studies Reporting on the Neuropsychological Performance of ADHD Adults Study Participants (Sex, age range) Diagnostic criteria (Medication use, subtypes of A D H D indicated; use of subtypes in analysis) Design and grouping procedures (Using IQ as a covariate) Measures Major Findings Murphy 18 A D H D DSM-IV semi- IQ not significant SSRT; Adults with A D H D performed significantly more poorly (2002a) 18 Controls structured interview between groups. GSRT than controls on tasks of inhibitory control. However the (Males, aged (Medication: not results were not significant between the two groups on a 27-58 years) indicated, subtypes not reaction time test. indicated) Murphy (2002b) 18 A D H D DSM-IV semi- IQ not significant B V R T ; Adults with A D H D performed significantly more poorly 18 Controls structured interview between groups. T O H ; than the control group on tests of executive control (i.e., (Males, aged ' (Medication: not T M T - T O H , and T M T - B). 27-58 years) indicated, subtypes not A & B indicated) Johnson et al. 56 A D H D DSM-IV Semi- Age was used as a WMS-R; Adults with A D H D showed deficits relative to controls (2001) 38 N C structured interview covariate, IQ not C O W A ; on tasks of memory, selective attention, visuomotor (71% males, (Medication: subjects significant between Stroop; tracking, and reaction time. Using IQ as a covariate aged 20-63 were washed out, groups. (IQ was WCST; showed no significant differences between groups. years) subtypes identified but used as a covariate T M T - A & not used in analysis) in a secondary T M T - B ; analysis). RTT 9 Design and Study Participants Diagnostic Criteria grouping criteria Walker et al. 30 A D H D DSM-IV criteria Age was used as a (2000) 30 Controls (Medication: subjects covariate, IQ not 30 Psychiatric were not on any, significant between (Sex mixed, subtypes not identified) groups. aged 17-50 years) Seidman et al. 64 A D H D DSM-III criteria, and Age was used as a (1998) 73 Controls self-report of childhood covariate, IQ not (Sex mixed, symptoms (Medication: significant between agedl9-59 subjects were not on groups. years) any, subtypes not identified) Measures Major findings C O W A ; CPT; Stroop; T M T ; WAIS-R subtests WAIS-FD; C V L T ; Stroop. WCST; CPT; ROCF Compared to healthy controls, adults with A D H D scored significantly more poorly on the dependent measures. However, no significant differences on any task were identified between adults with A D H D and those with a psychiatric disorder. Significant differences were found between the adults with A D H D and control subjects on reaction time and verbal learning tasks. The groups did not differ on any traditional measures of executive functioning. Epstein et al. (2001) Epstein et al. (1998) 25 A D H D 15 Controls 15 Psychiatric (Sex mixed, aged 18-65 years) c 60 A D H D 72 Controls (Sex mixed, mean age 25 & 35 years) Computer interview-self report of symptoms (Medication: subjects were not on any, subtypes were not identified) Semi-structured interview (Medication: not reported, subtypes identified and used in analysis) Gender, age, and CPT; education were not PVOT; different between SST any groups. IQ was not measured. Age was used as a CPT covariate, IQ was not examined. Adults with A D H D demonstrated significantly poorer performance on measures of response inhibition (e.g., reaction time) as compared to normal controls and individuals with anxiety disorders. Adults with A D H D performed significantly more poorly on all CPT indices (i.e., omission, commission, and reaction time scores) compared to normal controls. No significant differences between the A D H D subtypes were found. Diagnostic classification results for the CPT were moderate. 10 Design and grouping Study Participants Diagnostic Criteria ' procedures Measures Major Findings Murphy et al. (2001) 105 A D H D 65 Controls Sex mixed, aged 17-28 years) Structured interview, (Medication ceased 24 hours prior to testing, subtypes identified and used in analysis) Groups did not differ on age or sex. IQ scores were significant on the KBIT. (IQ and gender were used as covariates in a secondary analysis) CPT; After controlling for IQ significant between group Stroop; differences were found in areas of attention, nonverbal WAIS-III working memory, interference control, and verbal Digit Span fluency. Women with A D H D scored significantly higher & Digit than men on one measure (digit symbol subtest). No Symbol; significant differences between the A D H D subtypes were C O W A found. Corbett & Stanczak(1999) 27 A D H D 10 Controls (Sex mixed, aged 18-72 years) Semi-structured interview, (Medication: subjects were asked to refrain on day of testing, subtypes not identified) No difference between groups on gender and age. IQ was not examined. Stroop; Significant differences between the adults with A D H D T O A D and normal controls were found on the dependent measures. The Goldman-Fristoe-Woodcock Test of Auditory Discrimination (TOAD) showed high specificity and predictive power in discriminating adults with A D H D from controls. This test appears to be a measure of distractibility and behavioral disinhibition. Woods, Lovejoy, Starts et al. (2002) Kovner et al. (1998) 26 A D H D DSM-IV criteria from No group C O W A ; 26 N C normative database, differences C V L T ; (Sex mixed, (Medication: Subjects between groups on Stroop; aged 21-55 were not on any, gender, age, or T M T ; years) subtypes identified) education. (IQ was WAIS-R used in analysis) FD 19 A D H D Structured interview No group WAIS-R; 10 Psychiatric (Medication: Subjects differences were Benton; (Sex mixed, were not on any, found on age, Boston aged 18-57 subtypes were education, or Naming years) identified but not used intelligence. (IQ Test; in the analysis) was not examined) CPT; SST; W M R T Significant group differences between adults with A D H D and normal controls were found using a discrepancy analysis between intelligence and executive function. The diagnostic accuracy was moderate for the individual tests. Adults with A D H D scored significantly lower than a psychiatric group on a measure of simple attention (WAIS-R Digit Span subtest) and the reaction time component of the Shifting Sets Test (SST). Group classification rates between the psychiatric group, and the A D H D group, were adequate. 11 Study Participants Diagnostic criteria Design and grouping procedures Nigg et al. (2002) Fischer et al. (1990) 22 A D H D 21 Controls (Sex mixed, mean age was 23 and 21 years for the two groups) 100 A D H D 60 Controls (Males, aged 12-20 years) Previous diagnoses by psychiatrist and self-report of symptoms. (Medication: Subjects were not on any, subtypes identified but not used in analysis) Structured clinical interview. (Medication discontinued prior to testing, subtypes not identified) The groups were not significantly different on age, gender, or IQ. Age was used as a covariate. The groups were not significantly different on IQ. Schmitz et al. 30 ADHD Structured clinical The groups were (2002) 60 Controls interview (Medication: not significantly (Sex mixed, some subjects were different on any aged 12-16 taking medication, demographic years) subtypes were variables. (No identified and used in effect of sex or IQ analysis) was found in any measure). Seidman et al. 118 ADHD Structured clinical The groups were (1997) 99 Controls interview significantly (Males, aged (Medication: 80% of different on age 9-22 years) A D H D group on and IQ. (IQ was medication. No purposefully not significant differences controlled for). between medicated and non-medicated groups, subtypes not identified) 12 Measures Major Findings Anticasc- Young adults with A D H D had significantly more de task; difficulty with effortful motor inhibition on a computer Negative task than the control group Priming Task Stroop Young adults with A D H D showed no significant Test; differences from healthy controls on any of the dependent WCST; measures. CPT; ROCF WCST; The authors examined effects of the three subtypes of Stroop; A D H D . Adolescents with predominantly inattentive and Digit combined subtypes performed more poorly on tasks of Span. attention and psychomotor speed. Adolescents with predominantly hyperactive-impulsive type did not differ from the control group on any of the measures. These findings moderately support the diagnostic distinction among the A D H D subtypes proposed in the DSM-IV. Young adults with A D H D performed significantly more poorly on tasks of visual memory, problem solving, and set shifting. No significant differences were found between the two groups on tasks of attention and reaction time (CPT), or on a verbal list learning task (CVLT). WCST; ROCF; Stroop test; CPT; C V L T Design and grouping Study Participants Diagnostic Criteria procedures Measures Major Findings Stearns et al., 70 ADHD Structured interview, Sample did not WAIS-III; In a group of-adults with A D H D no significant (2004) (Sex mixed, (21.1% taking differ on age, sex, WMS-III relationship was found between working memory and mean age= 25 medication, subtypes or education. IQ (Working self-reported symptoms. Moreover, no significant effects years) not identified) scores were not Memory were found for gender or those taking stimulant significantly Indices); medications. different between Brown sample (i.e., sex, or A D D those on Scales medication). Note: ADHD= Attention Deficit Hyperactivity Disorder group. SSRT = Stop Signal Reaction Time Test; GSRT = Go Signal Reaction Time; B V R T ; Benton Test of Visual Recognition; TOH= Tower of Hanoi; Stroop = Stroop Test Color Word Task; WCST= Wisconsin Card Sorting Test; T M T = Trail Making Test; RTT = Reaction Time Test; COWA= Controlled Oral Word Association; WMS-R = Wechsler Memory Scale-Revised; CPT = Continuous Performance Test; WAIS-R = Weschler Adult Intelligence Scale; WRAT=Wide Range Achievement Test; C V L T = California Verbal Learning Test; ROCF = Rey-Osterrieth Complex Figure; PVOT=Posner Visual Orientating Task; T O A D = Goldman-Fristoe-Woodcock Test of Auditory Discrimination; W M R T = Warrington Recognition Memory Test. 13 A general profile of poor performance on tests of frontal/executive function is evident in adults with A D H D (Epstein et al., 2001; Johnson et al., 2001; Murphy, 2002a, 2002b; Woods, Lovejoy, & Ba l l , 2002). However, as shown in the reviewed studies (Table 1) adults with A D H D do not perform consistently more poorly than healthy controls across studies, nor do they consistently perform more poorly on specific measures (Fischer et al., 1990; Kovner et al., 1998; Seidman et al., 1998). Alexander and Stuss (2000) argue that the lack of discriminant validity (i.e., differentiating A D H D from healthy controls; differentiating clinical subtypes of A D H D ; differentiating A D H D from other disorders) on specific measures may be due to the fact that individual neuropsychological tests (e.g., C P T , W C S T ) assess too many dimensions of executive functioning. A s a result, they are not always helpful in discriminating neurocognitive problems associated with A D H D , where primary involvement is thought to reside with frontal subcortical systems. Future research needs to incorporate new assessment techniques to help characterize the cognitive deficits of adults with A D H D (Alexander & Stuss, 2000; Barkley, 1997; Johnson et al., 2001; Schmitz et al., 2002; Seidman et al., 1998). Moreover, implementing a battery approach utilizing multiple neuropsychological tests has been shown to improve the discriminant validity in this population (e.g., Woods, Lovejoy, Stutts et al., 2002). Numerous methodological limitations are inherent in the A D H D literature. Primarily, the lack of a reliable and valid method for diagnosing A D H D in adults is a major problem. In the reviewed studies, researchers often relied on only one diagnostic indicator (e.g., self-report in a clinical interview) with no cross-validation of symptoms from multiple settings (Epstein et al., 1998; Epstein et al., 2001; Murphy, 2002a, 2002b; Nigg et al., 2002). 14 Murphy and Schachar (2000) explored the use of self-ratings in the assessment of symptoms of A D H D in adults (ranging from 20 to 50 years of age) in two studies. The first study examined the validity of childhood recollections of A D H D behavior by having the individual with A D H D and a parent or spouse fill out similar rating scales. The second study followed a similar methodology but examined an individual's ability to self-report current symptoms of A D H D . Correlations between the self-reports of over 2100 individuals with A D H D , and their partners or spouses, were calculated. Results from both studies revealed significant similarity between ratings, and the authors concluded that adults with A D H D can provide reliable self-reports of past and present A D H D symptoms (Murphy & Schachar, 2000). The use of small sample sizes was a weakness in many of thestudies reviewed in Table 1. Most studies had no more than 20 participants in the A D H D or control group (e.g., Corbett & Stanczak, 1999; Kovner et al., 1998; Murphy, 2002a, 2002b; Schmitz et al., 2002) resulting in low statistical power and increasing the probability o f type II errors. Moreover, many researchers did not provide effect sizes or other indices of diagnostic efficiency (e.g., Fischer et al., 1990; Schmitz et al., 2002; Seidman et al., 1998), leaving specific interpretation of the findings uncertain, and decreasing the clinical utility of the results (Woods, Lovejoy, & B a l l , 2002). The age ranges o f the samples were often large, with participants ranging in age from 18-72 years of age (Corbett & Stanczak, 1999; Kovner et al., 1998; Murphy, 2002a, 2002b; Seidman et al., 1998; Walker et al., 2000; Woods, Lovejoy, Stutts et al., 2002). Thus, the potential effects of developmental changes that occur across the lifespan, and potential age-related declines in executive functions, were not controlled for. 15 Numerous studies have examined the effects of age-related declines on healthy adults using measures o f executive and other cognitive functions. The prefrontal cortex (i.e., the primary region examined by tests of executive function) has been found to be the most vulnerable area to aging compared to other areas of the brain, and age-related declines are thought to begin in early adulthood (Salthouse, 2003). However, most of the A D H D research to date is cross sectional, and longitudinal data is needed to examine the precise relationship between aging and its effects on cognitive tests in the A D H D population. Many of the reviewed studies had stringent inclusion criteria and controlled for factors known to affect cognitive performance (i.e., comorbid disorders such as depression or anxiety, and medication use). However, some studies did not (e.g., Epstein et al., 1998). Although individuals with A D H D often reported higher levels o f depression than controls (Seidman et al., 1998; Walker et al., 2000), it is often difficult to distinguish individuals with A D H D from those with a psychiatric disorder (e.g., anxiety disorder) on neuropsychological tests. From the studies reviewed, only three examined psychiatric groups compared to individuals with A D H D (Epstein et al., 2001; Kovner et al., 1998; Walker et al., 2000), and only one study (Kovner et al., 1998) reported differences between individuals with A D H D and a psychiatric sample. Differentiation of A D H D from commonly comorbid disorders (i.e., depression, anxiety, substance abuse) is cited as a weakness of the literature (Woods, Lovejoy, & B a l l , -2002). Future research needs to examine the ability of tests to discriminate A D H D from other comorbid conditions. The use of stimulant medications is generally found to increase the cognitive performance of individuals with A D H D (Schmitz et al., 2002). In the study for this thesis, the use of medication could not be controlled for. However, two studies reviewed examined this 16 issue (Seidman et al., 1997; Steams et al., 2004) and found no significant differences on cognitive performance between medicated and unmedicated participants with A D H D , although differences were found in Schmitz et al. (2002). In the children's A D H D literature, executive function deficits have been examined more thoroughly in males than in females (Carte, N igg , & Hinshaw, 1996; Nigg , Hinshaw, Carte, & Treuting, 1998). Some researchers suggest that girls with A D H D display greater cognitive impairment, while boys with A D H D display more obvious behavioral impairments (Gaub & Carlson, 1997). However, Halperin et al. (1990) found no cognitive differences between boys and girls with A D H D . In the adult A D H D literature, potential sex differences have received little attention, and some studies include only male participants (Fischer et al., 1990; Murphy, 2002a, 2002b; Seidman et al., 1997). Other studies included both sexes (Kovner et al., 1998; Murphy et al., 2001; Woods, Lovejoy, Stutts et al., 2002), but only analyzed whether the ratio of males to females in the A D H D group differed from the sex ratio in the control group. However, these studies did not examine whether or not there were performance differences on tests according to sex. However, Murphy et al., (2001) explored the potential cognitive differences between males and females with A D H D , and reported that women demonstrated higher scores than men on two measures: attention and working memory. In contrast, other studies (Schmitz et al., 2002; Seidman et al., 1998) found no significant sex differences in young adults with A D H D on multiple measures of cognitive functioning. A long-standing debate in the literature is whether or not to separate adults with A D H D by D S M - I V - T R subtypes. Some researchers maintain that executive function deficits do not appear to be a function of A D H D subtype (Woods, Lovejoy, & Ba l l , 2002). Due to small sample sizes, the majority of reviewed studies did not partition their A D H D sample by subtype 17 (Epstein et al., 2001; Kovner et al., 1998; N i g g et al., 2002; Walker et al., 2000). Only three studies have identified subtypes of A D H D and included them in their analyses (Epstein et al., 1998; Murphy et al., 2001; Schmitz et al., 2002), and only one study o f the three found cognitive differences across the three D S M - I V - T R subtypes (Schmitz et al., 2002). Specifically, Schmitz et al. (2002) found that adolescents with predominantly inattentive and combined type performed more poorly on tasks o f attention and psychomotor speed. Adolescents with the predominantly hyperactive-impulsive subtype did not differ from the control group on any measures. These findings appear to support the diagnostic distinction among the A D H D subtypes proposed in the D S M - I V that individuals with the hyperactive-impulsive subtype experience attentional deficits to a lesser degree than the other subtypes, and show the greatest decline in A D H D symptoms with age (Epstein et al., 1998; Schmitz et al., 2002). However, these results are very preliminary and run counter to the bulk of the literature to date. The present study did not partition the A D H D sample by A D H D subtype because this information was not available. Accounting for the different results across studies is difficult due to methodological differences (i.e., sample sizes, age ranges, and inclusion criteria). Further, the disparity may relate to the differences in the measures and the constructs examined. However, these findings mirror results in the pediatric literature that, thus far, have not consistently identified different cognitive profiles among the A D H D subtypes (Barkley, Grodzinsky, & DuPaul, 1992; Carlson, Lahey, & Neeper, 1986; Trommer, Hoeppner, Lorber, & Armstrong, 1988). Future research with larger numbers of participants and more stringent inclusion criteria is needed to determine whether or not there are cognitive differences on tests of executive functions among the A D H D subtypes. The total number of tests across studies that have reported significant and non-18 significant group differences between adults with A D H D , n o n - A D H D controls, and psychiatric groups is reported in Table 2. Table 2 Number of Studies Reporting Significant differences between Adults with ADHD, non-ADHD controls, and patients with psychiatric disorders, on commonly administered neuropsychological tests \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 . Measure N o n - A D H D Psychiatric Controls Group Attention/Executive Function Tests Yes N o Yes N o Continuous Performance Test (CPT) 7 3 0 2 Wisconsin Card Sorting Test (WCST) 1 4 0 1 Stroop Test 6 3 0 1 Trail Making Test- A ( T M T - A ) 2 2 0 1 Trail Making Test- B ( T M T - B ) 3 1 0 1 Language Ski l l Tests Word Fluency: C O W A T 3 1 0 - 1 Learning/Memory Tests California-Verbal Learning Test ( C V L T ) 2 2 0 0 Rey Complex Figure Test (RCFT) 1 2 0 0 W M S - R Logical Memory . 2 0 0 1 W M S - R Visual Reproduction 1 0 0 1 Intelligence Tests W A I S - R Ful l Scale IQ* . 2 5 0 0 WISC-III & W A I S - R Digit Span 3 0 0 0 WISC-III & W A I S FD/Digi t Symbol 3 1 0 1 Note: Yes = significant differences between adults with A D H D and comparison group/psychiatric group; No = No significant differences between adults with A D H D and comparison/psychiatric group; A D H D =Attention Deficit Hyperactivity Disorder group. WMS-R = Wechsler Memory Scale-Revised; WAIS-R = Weschler Adult Intelligence Scale-Revised; WISC-III = Wechsler Intelligence Scale for Children- Third edition; C O W A T = Controlled Oral Word Association Test. 19 Summary of Results for Specific Neuropsychological Tests A s can be seen in Table 2, the results from the studies are highly variable. However, most studies (although not all) have demonstrated significant differences between young adults with A D H D and healthy controls. The individual tests and subsequent results presented in Table 2 are discussed in detail below. Tests of Attention/Executive Functions. Barkley (1997) defines executive functions as a variety of higher-order cognitive skills that assist with the self-regulation of behavior. However, others have proposed different definitions, and a precise standardized operational definition has yet to be agreed upon in the literature. The term executive functions seems to incorporate planning or any goal-directed action, including persistence toward achieving a goal, inhibition (i.e., one's ability to resist a response or behavior), problem solving and strategy development, including monitoring, flexibility, and self-awareness across time (Barkley, 2000). These activities are underpinned by many lower order cognitive operations, with working memory being one of the most important processes (Spreen & Strauss, 1998). It is important to note that impairment of executive functions such as planning, flexibility, and judgment can be present with any major change in intellectual functioning (Lezak, 1995). Following this approach, many researchers' working definition of executive functions includes the following components: working memory, response inhibition, planning, cognitive flexibility, and concept formation (Royall et al., 2002). The cognitive tasks that measure these executive functions are typically components of large neuropsychological batteries, made up of many different tests, which include tests of frontal or executive functioning, and tests of other cognitive abilities (e.g., language and memory). 20 The Continuous Performance Test (CPT) is the most commonly used test for assessing attention deficits in A D H D (Woods, Lovejoy, & Ba l l , 2002). The C P T is a computerized test of attention, impulsivity, and vigilance. It involves discriminating between visually presented target and non-target stimuli (Spreen & Strauss, 1998). The task requires an individual to rapidly press a computer space bar or click a mouse button to letters presented in sequence in the center of the screen. In the standard condition, the examinee is required to press the space bar (mouse click) for every letter presented except the letter \" X . \" The test takes 14 minutes to complete. The omission errors (number of targets the person did not respond to), commission errors (number of times the person responded to a non-target \" X \" ) , incidental reaction time (mean response time), and variability of reaction time (consistency of response time) are calculated (McGee, Clark, & Symons, 2000). L o w to moderate correlations are reported between the C P T and other measures of attention. However, the precise cognitive processes assessed by the C P T are unclear; The general consensus (Halperin, Sharma, Greenblatt, & Schwartz, 1991; as cited in Spreen & Strauss, 1998) is that omission errors reflect difficulties with sustained attention, while commission errors reflect problems with impulsivity, attention, and memory. The C P T normative data is most representative of males (75.4% in sample) between 6 and 30 years o f age (only one-fifth of the sample was over eighteen years of age). The usefulness of the C P T as a measure that can distinguish children with A D H D from controls has been demonstrated consistently in the literature (Barkley, DuPaul, & McMurray , 1990; Halperin et al., 1990; Halperin et al., 1991). However, the measure has not been demonstrated to discriminate children with A D H D from other clinical groups (Barkley et al., 1990). 21 A more mixed pattern of results using the C P T is found in the adult A D H D literature. Kovner et al. (1998) found no differences on C P T performance between adults with A D H D and a psychiatric control group. Further, the C P T did not discriminate between adults with A D H D and controls in terms of hit reaction time (Murphy et al., 2001). In contrast, two other studies comparing adults with A D H D to healthy controls found that the A D H D group made significantly more errors of commission and omission than the control group (Epstein et al., 1998; Walker et al., 2000). St i l l , the majority of studies have found significant differences between groups on the C P T (see Table 2). The Wisconsin Card Sorting Test (WCST) assesses conceptualization and measures an individual's ability to problem-solve and/or shift cognitive strategies (Johnson et al., 2001). The task requires the examinee to sort a set of cards based upon three different criteria (i.e., color, form, and number). N o instructions on the sorting criteria are given to the examinee. The examinee is instructed to match their cards, one at a time, with one of four \"key cards.\" They are then only given feedback on a correct or incorrect placement, and are required to use that feedback to guide their future card placements to optimize correct responses. After correctly matching a card according to a matching stimulus (i.e., color) for 10 consecutive trials, the matching feature changes without warning (e.g., color to form), and the examinee must again discern the correct sorting criteria by the examiner's feedback (Lezak, 1995). This occurs six times (i.e., color, form, number, color, form, number), or until all 128 cards are administered. Successful performance on the W C S T requires that an individual determine the correct response for each set, maintain it, and then shift set (e.g., color to form) according to feedback (Romine et al., 2004). The problem-solving component involves the examinee considering a variety of hypotheses and maintaining or rejecting them according to the feedback they receive. 22 Performance is scored by categories completed (the number of correct matches completed in each category), trials to complete first category (the number of cards it takes to complete the first matching task), number of failures to maintain set (the number of times the examinee makes an incorrect category response more than four times in a row), and percent preservative errors (which reflects the amount of preservative errors as a percentage of overall test performance). Normative data for the test are available for individuals 6 to 89 years of age (Spreen & Strauss, 1998). Age has the strongest demonstrated relationship to W C S T performance. Performance increases are reported from ages 5 to 19 years; performance stability is noted for individuals aged 20-50 years; and declines in some aspects of performance are reported in individuals aged 60 years and above (Heaton, Chelune, Talley, Kay , & Curtis, 1993; as cited in Spreen & Strauss, 1998). The test's sensitivity and specificity as a measure of executive functioning has been demonstrated in numerous studies (Lezak, 1995). Similar to the C P T , statistically significant differences on the W C S T between children with A D H D and controls frequently are reported (Barkley et al., 1992; Schmitz et al., 2002). However, there has been little success using the W C S T to discriminate children with A D H D from other clinical populations (Snow, 1998). With regard to the reviewed literature, only the studies that included adolescents and young adults with A D H D found significant differences compared to healthy controls (Schmitz et al., 2002; Seidman et al., 1997). Two other studies using older participants (aged 19-63) found no significant differences on the W C S T compared to healthy controls (Johnson et al., 2001; Seidman et al., 1998). Moreover, similar to the difficulty discriminating children with A D H D from other clinical populations, Fischer et al. (1990) found no{ significant differences on the W C S T between a group of adults with A D H D 23 and psychiatric controls. Taken together (see Table 2), the majority of studies fail to differentiate between adults with A D H D and controls using the W C S T , especially in adulthood. The Stroop Test is a classic test of reading fluency, visual attention, mental flexibility, and inhibitory control requiring participants to read lists of words and colors (Johnson et al., 2001). The first part of the task requires the examinee to read color names printed in black ink (e.g., red, green). The second part of the test requires the individual to name the colors that colored X s , are printed in. In the last part of the task the individual is presented with color names that are now printed in different colored ink (e.g., the word \"red\" printed in blue ink), and they are required to disregard the word, and name the ink color. This requires the individual to inhibit the over-learned, automatic response of reading the stimulus word in order to respond to the more novel task of naming the color of the ink. For each part of the test, both the time to complete the test and the number of errors are recorded and scored. Having to inhibit the over-learned (prepotent) word reading response results in significantly slower performance than word reading or color naming alone. This decrement in performance has been labeled the Stroop \"interference effect\" (Spreen & Strauss, 1998). The construct validity of the Stroop test has been examined in adolescents with and without A D H D . MacLeod and Prior (1996) found significant correlations between Stroop interference and the Paced Auditory Serial Addition Task (a task requiring speeded processing, mental arithmetic, and the ability to divide attention), but not between the Stroop and a test purported to measure intelligence (Slossan Test of Intelligence). However, normative data for the Stroop test suggests that both age and intellectual levels are strong predictors of performance. Individuals aged 25-35 years of age have shown higher levels of performance on ,the first part of the task (reading the color names in black ink), and lower levels of performance 24 on the interference part of the task (naming the ink color). However, older participants, aged 70-80 years of age are relatively slow on the first part of the task, but relatively faster on the interference part of the task (Klein, Ponds, Houx, & Jolles, 1997, as cited in Spreen & Strauss, 1998). The Stroop Test has been frequently used to differentiate between children with A D H D and healthy controls (Barkley et al., 1992; Loge et al., 1990; Pineda et al., 1998) and has been effective in discriminating A D H D and non A D H D individuals in the majority of studies conducted (see Table 2). However, similar to the previous reviewed tests of executive functioning (e.g., CPT , W C S T ) , only the studies that included adolescents and young adults with A D H D found significant differences compared to healthy controls (Murphy et al., 2001; Schmitz et al., 2002; Seidman et al., 1997). Two other studies using older participants, up to sixty-three years of age, found no significant differences on the Stroop test compared to healthy controls (Johnson et al., 2001; Seidman et al., 1998). The Trail Making Test (TMT) is a test of speed o f visual scanning/attention, visuomotor speed, sequencing, and mental flexibility. The T M T has two parts. Part A requires the participant to draw lines in numerical sequence (1-25) connecting 25 circled numbers placed randomly on a page. Part B also requires the examinee to draw lines in a similar fashion, but with the addition of having to alternate between connecting numbers (i.e., 1-13) and letters (i.e., A - L ) (i.e., connecting 1-A-2-B-3-C and so on). Part B is a more complex task than Part A because it requires divided attention, set shifting, and maintaining two different streams of information in working memory. Moreover, besides switching between numbers and letters in Part B , the actual distances between the circles are larger (Spreen & Strauss, 1998). Both Parts A and B are timed, and feedback is given to the examinee i f an incorrect move is made (e.g., 25 connecting out of sequence) and the participant is required to continue with the correct connection. The score reflects the time to completion in seconds (Johnson et al., 2001). Normative data for the Trail Making test demonstrates that both parts of the test (A & B) are sensitive to age, education, and intelligence (Salthouse & Fristoe, 1995). However, Part B is more sensitive to age-related declines and differences in intelligence, which likely reflects the differing cognitive demands between the two tasks. The reported correlation between Part A and B is only .49 (less than 25% of the variance in performance), suggesting they are underpinned by substantially different cognitive functions (Heilbrormer, Henry, Buck, Adams, & Fogle, 1991; as cited in Spreen & Strauss, 1998). Similar to the children's A D H D literature, studies comparing the performance of adults with A D H D to clinical or n o n - A D H D control samples show inconsistent results (Barkley et al., 1992). Specifically, two studies in the review demonstrated that adults with A D H D performed significantly more poorly than healthy controls on both Part A and Part B of the T M T (Murphy, 2002a; Woods, Lovejoy, Stutts et al., 2002). In contrast, Walker et al. (2000) found no difference on Parts A or B when they compared adults with A D H D to either healthy controls or a psychiatric group. Finally, Johnson et al. (2001) administered the T M T to adults with A D H D and n o n - A D H D controls and found that the A D H D group performed significantly more poorly on Part B , but not on Part A . The authors suggest the reason for this discrepancy was that Part B is a more complex task with far greater cognitive demands, and is therefore more sensitive to differences between adults with A D H D and healthy controls. Language Skill Tests. Measures of verbal fluency require an individual to generate words associated with a certain letter (phonemic verbal fluency - e.g., g= golf, gift, great) or a certain category (semantic verbal fluency - e.g., animals = dog, horse, cat) in a fixed amount of 26 time (typically 60 seconds). Verbal fluency tests have shown mixed results in distinguishing children with A D H D from n o n - A D H D controls (e.g., Barkley et al., 1992). The results with adults have been mixed, too (see Table 2). The Controlled Oral Word Association Test ( C O W A T ) was the only test used to measure verbal fluency in the adult A D H D literature reviewed. O f the studies reviewed, three demonstrated significant differences in C O W A T performance between adults with A D H D and n o n - A D H D controls (Murphy et al., 2001; Walker et al., 2000; Woods, Lovejoy, Stutts et al., 2002), and one study found no differences (Johnson et al., 2001). Significant differences in performance between adults with A D H D and n o n - A D H D controls on the C O W A T are likely due to the identified deficits in the functioning of frontal systems in A D H D that are tapped by the C O W A T ' s demands on sustained attention to stimuli, organization, and retrieval of verbal information (Woods, Lovejoy, & Ba l l , 2002). Similar to the poor discriminatory power o f other tests o f executive function in differentiating A D H D performance from the performance of psychiatric groups (e.g., W C S T , Stroop test), Walker et al. (2000) reported that performance on the C O W A T did not distinguish adults with A D H D from psychiatric controls. Thus, as suggested in Table 2, results of studies examining differences between A D H D and non A D H D individuals using the C O W A T have yielded rather equivocal results, with some studies showing significant difference and some not. Learning and Memory Tasks. Verbal list learning difficulties are commonly identified in adults with A D H D . A s a result, list learning tests are the most widely used measures in studies of learning and memory in the A D H D population (Seidman et al., 1997; Seidman et al., 1998; Woods, Lovejoy, Stutts et al., 2002). The California Verbal Learning Tests ( C V L T , C V L T - I I , C V L T - C ) are the most popular memory tests used with this population. The C V L T is 27 constructed to provide an assessment of the strategies and processes involved in learning and remembering verbal material. The 16 items included in the C V L T (e.g., items from a shopping list) are presented five times to the examinee. A distractor list is then administered once, after which short delay free and cued recall, long-delay free and cued recall, and recognition memory trials are administered (Woods, Lovejoy, Stutts et al., 2002). Both Seidman et al. (1998) and Woods, Lovejoy, Stutts et al. (2002) reported that adults with A D H D performed significantly more poorly than n o n - A D H D controls on aspects of the C V L T , including the total words learned after five learning trials and in their use of semantic clustering. However, although they learned fewer words initially, they retained the same percentage of the total words they learned after a long delay. One possible interpretation for the above findings is that adults with A D H D lack the ability to discern and/or use the inherent semantic structure o f the groups of words in the list (e.g., tools, articles of clothing) to aid in the organization of material to be learned (Woods, Lovejoy, & Ba l l , 2002). A s a result, adults with A D H D appear to have deficits in the encoding of verbal information, but not with the storage or retrieval of the material. This profile o f impairment is evident in the children's literature as well . The poor use of efficient semantic clustering learning strategies is proposed . to reflect the frontal-subcortical impairment believed to underpin the deficits of individuals with A D H D (Woods, Lovejoy, & Ba l l , 2002). This pattern of performance does not seem to be reflected in tests of visual learning and memory (Kovner et al., 1998; Murphy, 2002a). Only one o f the reviewed studies, using the Visual Reproduction subtest of the W M S - R , found that adults with A D H D performed more poorly than a control group (Johnson et al., 2001). In contrast, other studies using different tests of visual memory (e.g., Benton Test, Rey-Osterrieth Complex Figure Test) did not find any 28 differences between adults with A D H D and n o n - A D H D controls (Kovner et al., 1998; Murphy, 2002a; Seidman et al., 1998). This may reflect the fact that visual tests, unlike the verbal list learning paradigm, do not appear to have an obvious advantageous learning strategy. Tests of Intelligence. A s illustrated in Table 2, tests of intelligence (IQ) have not been successful in discriminating between adults with A D H D and n o n - A D H D controls (Seidman et al., 1998; Walker et al., 2000). However, most of the studies used only estimates of intelligence based on either oral reading tests (e.g., Shipley test) or short-forms of intelligence tests (e.g., Kovner et al., 1998; Walker et al., 2000), thus reducing the possibility of finding differences.. However, Murphy and colleagues (2001) found that, after controlling for IQ, differences found on tests of executive functioning between young adults with A D H D and n o n - A D H D controls did not retain their significance. Similarly, Woods et al. (2002) applied an alternative method of neuropsychological test interpretation by using an intra-individual discrepancy analysis. They examined the differences between adults with A D H D and n o n - A D H D controls in terms of their intellectual functioning and performance on a battery of six tests measuring executive functioning. Significant discrepancies between the adults with A D H D and n o n - A D H D controls were found between their IQ estimates and multiple tests of executive functioning. The Freedom from Distractibility Index (i.e., Arithmetic and Digit Span subtests) from the W A I S - R and the WAIS-III has been used successfully to distinguish between adults with A D H D and n o n - A D H D controls (e.g., Kovner et al., 1998; Woods, Lovejoy, Stutts et al., 2002). Moreover, the Digit Symbol Subtest from the W A I S - R and the WAIS-III has also shown success in identifying differences in adult A D H D research (e.g., Murphy et al., 2001; Walker et al., 2000). However these findings are not consistent (e.g., Seidman et al., 1998). 29 The use of intellectual measures for identifying individuals with A D H D is a contentious issue (Woods, Lovejoy, & Ba l l , 2002). Many researchers question the utility of studies reporting significant differences on measures of intelligence due to the fact that many measures of IQ are composed o f tasks that are sensitive to attentional deficits and executive functions. A s a result, these findings could potentially be explained by impairments in attention, and or frontal/executive functions, rather than differences in IQ (Woods, Lovejoy, & B a l l , 2002). Summary In summary, research has produced variable results in terms of the utility of neuropsychological measures for differentiating individuals with A D H D from n o n - A D H D controls. A detailed review of the literature indicates that the most success to date has been found in discriminating children with A D H D from healthy controls. Studies of adolescents and young adults are somewhat weaker, with the most variable findings reported in the adult A D H D literature. Even greater difficulty has been encountered when neuropsychological tests are used to discriminate persons with A D H D from other clinical populations such as psychiatric patients. This is likely a result of the overlapping deficits in attention and executive dysfunction, as is seen, for example, in schizophrenia, and the fact that the poorly defined construct of executive functioning reflects a number of higher-order cognitive abilities which are predicated on more basic cognitive functions such as attention and memory abilities. Further, the fact that the neuropsychological literature continues to reflect poor general agreement as to what abilities constitute the executive functions compounds the difficulty faced by researchers. Existing tests of executive functioning are often confounded to some degree by the need to assess the executive functions in association with tasks that utilize other non-executive cognitive abilities (for example, set shifting on the Trails B task is assessed via visual 30 scanning and graphomotor ability). Further, many of the tests used by researchers lack good psychometric data (Lezak, 1995). This is perhaps one of the most serious weaknesses in the rapidly growing field of clinical neuropsychology, and is also attributed to the lack of consensus on the definition of executive functions. The literature also reflects a number of additional significant methodological limitations. Specifically, poor research design, small sample sizes (no more than 20 participants), poor control for medications, poor gender control, poor age control, lack of IQ control, and very limited investigation of the impact of the subtypes of A D H D on neuropsychological performance. One trend that clearly emerges from the overall literature is that neuropsychological batteries composed o f tests of a number of different cognitive abilities, with executive functioning components, appear to be more successful at discriminating the cognitive effects of A D H D from the performance of n o n - A D H D controls and of other clinical populations. However, battery approaches tend to be time consuming, expensive and burdensome, and as a result the cost-benefit ratio for large-scale screening and diagnosis of A D H D tends to be poor. Research in this area is ongoing, but there appears to be a clear need for a rapid, objective, and cost effective screening approach that can reliably identify the cognitive profile associated with A D H D from n o n - A D H D controls and other clinical populations. Rationale for the Current Study Due to the multiple cognitive deficits associated with A D H D , there is a need for tests that cover a broad array of attentional and executive functions. A s many authors have identified, neuropsychological assessment in the adult A D H D population w i l l be most useful when multiple cognitive constructs are assessed (Alexander & Stuss, 2000; Barkley, 1997; 31 Johnson et al., 2001; Schmitz et al., 2002; Seidman et al., 1998). Most neuropsychological tests require specialized training to administer and score, and can be very time consuming to administer. Computerized testing has a number of advantages over traditional pencil and paper tests, including greater reliability due to decreased variability in administration, and more precise response recording. Moreover, the administration of standardized examiner-administered neuropsychological tests requires a substantial amount of training. Disadvantages of computerized testing include the absence of behavioral observations (i.e., qualitative information) during the test process, and the poorly understood and investigated influence of using a computer interface and administration method on performance characteristics. I m P A C T , the test used for this study, is a 20-25 minute computerized battery that is sensitive to subtle cognitive problems, such as those associated with concussions in sports (Lovell et al., 2003;. Lovel l , Collins, Iverson, Johnston, & Bradley, 2004). If I m P A C T could also be demonstrated to be sensitive to the problems associated with A D H D , it might be useful in clinical practice and research, and especially in clinical trials involving medications. Given its brevity, minimal practice effects, and multiple alternate forms, it also has the potential to be very useful in longitudinal A D H D research. This thesis w i l l contribute to the literature by examining whether I m P A C T is sensitive to cognitive problems in young adults with A D H D . This study does not overcome all of the noted methodological limitations associated with previous research. However, it does have an adequate sample size for the statistical analyses, samples across one age group, and compares young adults with A D H D with n o n - A D H D controls matched on several relevant demographic variables (i.e., age, gender, education). A s such, it takes significant steps towards overcoming some of the methodological limitations that exist in the extant A D H D literature. 32 Immediate Post-Concussion Assessment and Cognitive Testing (ImPACT) I m P A C T (Maroon et al., 2000) was developed to address the need for rapid screening of the large number of athletes pre-season and after experiencing concussions. This battery was designed to address the limitations associated with traditional neuropsychological testing in sports (e.g., administration time, expense, practice effects). Prior to 1998 there were few neuropsychological test batteries developed specifically for use with athletes. I m P A C T is composed of a demographic questionnaire, injury evaluation form, symptom inventory, and a neuropsychological test battery (Collins et al., 2002; Lovel l et al., 2003). The neuropsychological test battery consists of seven individual test modules (word discrimination, symbol memory, color click, symbol matching, color word match, sequential digit tracking, and visual attention span) that measure aspects of cognitive functioning including attention, memory, reaction time, processing speed, and impulse control. Composite scores for the test modules are computed by standardized formulas derived from the results of seven cognitive tasks (Collins et al., 2003). The seven modules can be administered as a complete test battery, or can be administered individually (Maroon et al., 2000). Various indices of performance are derived from these seven tasks, and can be combined to yield four composite scores, reflecting the individual's reaction time, memory, processing speed, and impulse control. The tasks involved in each module, and the indices included in each composite, are described in greater detail in the methods section. The I m P A C T battery includes a Post-Concussion Scale that is frequently used in both amateur and professional sports (Collins et al., 2003; Collins et al., 2002; Iverson, Gaetz, Lovel l , & Collins, 2004a, 2004b). The Post-Concussion Scale asks each participant to report on 22 symptoms (e.g., headaches, dizziness, problems with sleep, irritability, sadness, feeling 33 slowed down, difficulties concentrating, poor memory, visual problems) using a 7-point Likert scale (i.e., 0-6). The present study examined whether young adults with A D H D endorse more symptoms with greater severity in total than a group of matched n o n - A D H D controls. Symptoms included on the questionnaire that are known to be associated with A D H D (e.g., difficulty concentrating) are of particular interest to the present study. ' I m P A C T was designed as a rapid screening tool to permit the evaluation o f a large numbers of athletes in a limited time. A s such, the I m P A C T test battery is brief (approximately 20 to 25 minutes for baseline evaluations) and does not evaluate all cognitive functions (e.g., it does not include tests of intelligence, achievement, or language). I m P A C T was initially constructed to evaluate the areas of cognitive functioning most likely to be affected after cerebral concussions. When an individual experiences a concussion, cognitive functioning is disrupted. Immediately following the concussion, individuals are found to have difficulties in the areas of orientation, attention, executive functioning, information processing, mental set shifting, concentration, and memory (Delaney, Lacroix, Gagne, & Antoniou, 2001; Erlanger, Kutner, Barth, & Barnes, 1999). Although I m P A C T was not specifically designed to screen for cognitive functioning in A D H D , it evaluates areas of cognitive functioning with tests that have been demonstrated to discriminate between individuals with A D H D and n o n - A D H D controls (e.g., verbal list learning, response inhibition, sustained attention; Johnson et al., 2001; Walker et al., 2000; Woods, Lovejoy, Stutts et al., 2002). I m P A C T is automatically computer scored. The test stimuli are randomized from one testing session to another. This allows for the test battery to be used repeatedly over short intervals, while controlling for practice effects (Iverson, Lovel l , Collins, & Norwig , 2002; 34 Maroon et al., 2000). Most examiners can administer the battery after a few hours of instruction and review of materials, and little supervision of the test-taker is required (Maroon et al., 2000). I m P A C T has been used in several studies of concussion in amateur athletes, and has been shown to be sensitive to the immediate effects of concussion, and to reliably identify rapid improvement in functioning (Collins et al., 2003; Collins et al., 2002; Iverson et al., 2004a, 2004b; Love l l et al., 2003; Lovel l et al., 2004). Several aspects of the reliability (e.g., test-retest reliability) and validity of I m P A C T have been investigated (Iverson, Lovel l , & Collins, 2002; Iverson, Love l l , Collins et al., 2002; Iverson, Lovel l , Podell, & Collins, 2003). Iverson, Love l l , Podell, and Collins (2003) summarized the reliability and validity data for version 1.0 of I m P A C T . The reliability studies have addressed test-retest reliability and the determination of reliable change. It is not possible to assess internal consistency on the individual subtests or the composite scores because individual subtest responses cannot be downloaded from the program, and the composite scores are composed of a small number of subtest scores (thus, they are not amenable to reliability analyses). The test-retest reliability and estimates of reliable change have been presented for version 1 and version 2 of I m P A C T (Collins et al., 2003; Iverson, Lovel l , & Collins, 2002, 2003). Reliable Change methodology uses statistical formulae to identify whether change in an individual's performance on a measure with repeated testing is the result of a \"true\" change in their performance, or remains within the confidence interval associated with the instrument's measurement error. In contrast, test-retest reliability provides an index of the consistency with which a measure evaluates a given function on repeat testing (i.e., how well results from testing at time 1 relate to testing at time 2; Hageman & Arrindell , 1993; Heaton et al., 2001; Jacobson & Truax, 1991; Temkin, Heaton, Grant, & Dikmen, 1999). 35 Test-retest reliability and reliable change estimates were derived from 49 amateur athletes tested over three occasions. The second administration of the test was given an average of 14 days (Range = 7-21 days) after baseline testing. The correlation coefficients from Time 1 to Time 2 ranged from .54 - .76 for the composite scores. The third administration was given approximately 4.5 days (Range = 2-7 days) after the second testing, and the correlation coefficients from Time 2 to Time 3 ranged from .48 - .68 for the composite scores. Iverson, Lovel l , and Collins (2005) conducted a study on the construct validity of I m P A C T . They compared I m P A C T (version 2.0) to a traditional neuropsychological measure, the Symbol Digit Modalities Test ( S D M T ) . The S D M T is a test of visual scanning, visuomotor ability, attention, and speed of processing. It has similar task demands as the Trail Making Test Part A , and the Digit Symbol (Coding) Test (Spreen & Strauss, 1998). The authors hypothesized that the Processing Speed and Reaction Time Composites of I m P A C T would correlate most highly with the Symbol Digit Modalities Test. Results from the analyses suggested that the S D M T , Reaction Time Composite, and Processing Speed Composite from I m P A C T were measuring similar constructs, demonstrating some preliminary convergent validity (Iverson et al., 2005). Because the Processing Speed and Reaction Time Composites from Version 1.0 of I m P A C T are identical to that of Version 2.0, the results of this research are relevant to the current study. The ongoing validation of a new test is a lengthy and time-consuming process (Lezak, 1995). Future validity research on I m P A C T needs to examine its convergent and discriminant validity with other tests (Iverson et al., 2005). The validity of I m P A C T as a battery that measures sports-related concussion has been examined (e.g., Iverson, Lovel l , & Collins, 2002). Amateur athletes (N= 120) who had completed pre-season testing were re-evaluated within three days of having a concussion. 36 Divergent validity was studied through an intercorrelation matrix of the composite scores at preseason and post injury. A t preseason the only statistically significant correlation was between the Reaction Time and Processing Speed (r =.35). A t post injury, there were significant, but small, correlations between Memory and Reaction Time (r =-.27), Memory and Processing Speed (r = .35), and Reaction Time and Processing Speed (r = .32). These results suggest that the composite scores do not share a great amount of variance, and are therefore capturing predominately different aspects of cognitive functioning. To date, the psychometric data available for I m P A C T is quite limited. Much additional research is needed. The battery appears to have adequate test-retest reliability, solid estimates of reliable change, and it is sensitive to the acute effects of concussions in high school and university students. In young people with A D H D , there is a substantial overlap in terms of the identified areas of compromised cognitive functioning evaluated by I m P A C T ; thus, there might be potential utility of the I m P A C T battery in the A D H D population. I m P A C T measures several areas of cognitive functioning that adults with A D H D appear to show deficits (e.g., attention, memory, reaction time, and processing speed; Epstein et al., 2001; Kovner et al., 1998; Murphy et al., 2001; Murphy, 2002b; Seidman et al., 1998; Walker et al., 2000; Woods, Lovejoy, Stutts et al., 2002). The sensitivity of this computerized battery to the subtle effects of concussion suggests that the battery may also be useful for identifying cognitive problems associated with A D H D . One study has been conducted with adolescents with A D H D . Iverson and Strangway (2004) examined I m P A C T version 2.0 performance in a sample of 38 adolescents with A D H D and 38 n o n - A D H D students matched for age, education, gender, and history of head injury. The average age of the students was 15.5 years (Range = 13-19) and their average education was 9.1 years (all were in grades 8-12). The 37 majority of the participants were boys (92%). A l l participants were derived from the I m P A C T normative sample. The students with suspected A D H D were not diagnosed through structured interviewing or testing; a psychologist did not evaluate them. Each individual in the A D H D group was identified from their self-reported responses on the demographic questionnaire, which asked them to identify (yes/no) whether or not they had ever been diagnosed with attention deficit hyperactivity disorder ( A D H D ) or attention deficit disorder ( A D D ) . Results from the study revealed significant differences between the control and A D H D groups in terms of their performance on the visual memory, processing speed, and impulse control composite scores. The groups did not differ significantly in terms of their verbal memory or reaction time composites. The results are similar to those found by Seidman and colleagues (1997) who used a comparable sample, and investigated the cognitive task performance of a group of young A D H D males (aged 9 to 22) and n o n - A D H D controls using standard neuropsychological tests. The individuals with A D H D performed significantly more poorly on a task of visual memory, and on tasks of concentration/executive functioning primarily involving components of impulse control (Stroop test), and problem solving, set shifting, and cognitive flexibility (WCST) . N o differences were found between the two groups on a verbal list learning task ( C V L T ) or a reaction time task (CPT). Schmitz et al. (2002) reported similar findings when they compared a group of adolescents with A D H D to n o n - A D H D controls on measures o f neuropsychological performance. These, and a number of additional studies, suggest that neuropsychological impairments identified in children with A D H D continue into adulthood (e.g., Halperin et al., 1990; Konrad et al., 2000; Kupietz, 1990; Loge et al., 1990; Seidel & Joschko, 1990; van der Meere & Sergeant, 1988). 38 Hypotheses The purpose of the study was to examine the potential utility o f I m P A C T for distinguishing A D H D and n o n - A D H D individuals in a sample of young adults. The participants were matched on education, gender, and history of head injury. The study investigated whether the A D H D and matched controls displayed cognitive differences in terms of their concentration, memory, reaction time, processing speed, and impulse control as measured by I m P A C T . O f additional interest was whether self-reported cognitive difficulties, as reported on I m P A C T ' s Post-Concussion Scale, distinguished between A D H D and non-A D H D participants. The specific hypotheses for this study are listed below: 1) Young adults with A D H D wi l l perform significantly more poorly on the memory composite than matched n o n - A D H D controls. Adults with A D H D appear to have deficits in the encoding and retrieval of verbal information, primarily related to executive aspects of efficient memory strategy and verbal organization skills (e.g., semantic versus phonemic chunking of information). Moreover, verbal memory deficits are one of the most common difficulties identified in adults with A D H D (e.g., Seidman et a l , 1997; Seidman et al., 1998; Woods, Lovejoy, Stutts et al., 2002). Some researchers have reported deficits on visual memory tests, too (Johnson et al., 2001). 2) The reaction time score for the young adults with A D H D w i l l not be significantly different than the normative comparison group. Reaction time tests have received little attention in the adult A D H D literature. However, Johnson et al. (2001) found that adults with A D H D performed more slowly than n o n - A D H D controls on a reaction time task (3RT) as the task became more complex. Only one other study (Murphy, 2002a) used a measure of simple reaction time (GSRT) and found that the test did not 39 discriminate adults with A D H D from controls. The most commonly used test to assess reaction time in the adult A D H D literature is the Continuous Performance Test (CPT). However, the C P T is not an explicit test of reaction time. Rather, reaction time is measured \"incidentally\" (i.e., the participant is not asked to solely respond to a stimulus as quickly as they can - because they are also required to monitor for \"X ' s\" to which they do not respond; Epstein et al., 1998; Epstein et al., 2001). For the purposes of the C P T , variability in reaction time over the duration of the task is used to identify inconsistent attentional patterns, or attention/arousal that diminishes over time. Further, the majority of the studies using the C P T to evaluate reaction time have not demonstrated significant differences between adults with A D H D and controls (Fischer et al., 1990; Murphy et al., 2001; Seidman et al., 1997). Iverson and Strangway (2004) reported no differences between young people with A D H D and controls on version 2.0 of I m P A C T . Because the literature does not appear to support reaction time differences between n o n - A D H D controls and individuals with A D H D , there is no empirical reason to expect that they would differ in terms of reaction time on I m P A C T . Hence it is expected that they w i l l follow the pattern of previous literature and perform similarly. 3) The processing speed composite w i l l be significantly slower for young adults with A D H D compared to n o n - A D H D controls. Tests of psychomotor speed have shown limited utility in discriminating between adults with A D H D and n o n - A D H D controls (Seidman et al., 1998; Walker et al., 2000). However, because results from Iverson and Strangway (2004) using I m P A C T version 2.0 demonstrated differences on the processing speed composite score between adolescents with A D H D and non-A D H D controls, there is some reason to believe that the I m P A C T processing speed composite w i l l discriminate between young adults with A D H D and n o n - A D H D controls. 40 4) Young adults with A D H D w i l l score significantly more poorly than the control group on the impulse control composite. Two traditional neuropsychological tests (CPT and Stroop Test) have been typically used to measure impulse/inhibitory control, and have been frequently used to differentiate between children with A D H D and n o n - A D H D controls (Barkley et al., 1992; Loge et al., 1990; Pineda et al., 1998). The C P T has generally been successful at distinguishing adults with A D H D from N o n - A D H D controls by measuring impulse control through errors of commission (i.e., responding to a target stimuli when withholding of response is required; Epstein et al., 1998; Murphy et al., 2001; Walker et al., 2000). The Stroop task measures inhibition by requiring the participant to suppress their prepotent (overlearned) reading response. The most numerous significant results differentiating participants with A D H D from controls with Stroop-type tests are reported in studies that include only adolescents and young adults with A D H D (Schmitz et al., 2002; Seidman et al., 1997). Similarly, using version 2.0 of I m P A C T , Iverson and Strangway (2004) found significant differences between adolescents with A D H D and n o n - A D H D controls. These ( \u00E2\u0080\u00A2 findings provide some empirical basis to believe that the I m P A C T Impulse Control composite score w i l l differentiate between the A D H D and control group. 5) Young adults with A D H D w i l l report significantly more symptoms on the Post-Concussion Scale than controls. Adults with A D H D self-report more psychiatric symptoms than N o n - A D H D controls (e.g., feeling down, feeling irritable, feeling depressed; Woods, Lovejoy, & Ba l l , 2002), and other symptoms tapped by the Post-Concussion Scale overlap with common symptoms o f A D H D (e.g., trouble concentrating and trouble with memory). A s such, the Post-Concussion 41 Scale is likely to reflect elevated scores in A D H D because it is expected that they w i l l endorse many of the symptoms with more frequency than N o n - A D H D controls. Methodology Participants From an initial database of 2, 389 subjects, 84 were identified as having self-reported A D H D . O f these subjects, 68 had complete data (e.g., 9 were missing data on education). The normative database for I m P A C T (N = 1, 746) was then used to select a matched group of 68 n o n - A D H D controls. Participants were matched precisely on education, gender, and number of previous concussions. Each group had 88% males and 12% females. The average number of completed years of education was 12.3 (SD = 2.0) for the A D H D group and 12.3 (SD = 2:0) for the control group. The average number of previous concussions was .68 (SD = 1.3) for the A D H D group and .62 (SD = 1.2) for the control group. The breakdown of self-reported educational problems in the A D H D group was as follows: repeated a grade = 7.4%, reading problem = 22.1%, spelling problem = 25.0%, math problem = 17.6%, and recipient of special education services = 16.2%. The control subjects, by selection criteria, did not have any self-reported educational problems. For the total sample, 39% percent were in high school and 61% were in university. The breakdown of participants by highest grade completed was as follows: Grade 9 = 13.2%, Grade 10 = 18.4%, Grade 11 = 13.2 %, Grade 12 = 22.1%, 1 s t year university = 20.6%, 2 n d year = 8.1%, 3 r d year = 13.2%, and 4 t h year = 4.4%. 42 Procedure A l l participants completed Version 1.0 of I m P A C T as part of a larger collection of normative data for I m P A C T . The testing was done in group settings (e.g., computer labs in schools). Each administration of I m P A C T took approximately 20-25 minutes. The students with self-reported A D H D were not diagnosed through structured interviewing or testing; and a psychologist did not evaluate them. This is a sample of convenience, derived from a normative database. The students were identified as having A D H D by their self-reports in the demographic questionnaire. Specifically, the students were asked whether or not they had ever had a diagnosis of attention deficit hyperactivity disorder ( A D H D ) or attention deficit disorder ( A D D ) . . Measures The following section provides a detailed description of I m P A C T . This program contains a demographic questionnaire, current symptoms questionnaire, and a neuropsychological screening battery. The first section of I m P A C T is the Subject Profile and Health Questionnaire. It requires the participant to input basic demographic and descriptive information including their name, date of birth, age, sex, grade level, and first language. It also requires the individual to report their height, weight, handedness, sport, and whether or not they have ever had a concussion. In addition, the questionnaire requires the test-taker to report whether or not they have received any speech therapy, attended special learning classes, repeated one or more years of school, or been diagnosed with A D H D . Section two of I m P A C T pertains to \"Current Symptoms and Conditions\", or what is referred to as the Post-Concussion Scale. The Post-Concussion Scale asks each participant to 43 report on 22 concussion-related symptoms on a 7-point Likert scale (i.e., 0-6) identifying the degree of difficulty, i f any, they are having with each symptom (e.g., problems with sleep, irritability, sadness, feeling slowed down, difficulties concentrating, poor memory, visual problems). The Post-Concussion Scale is reprinted in Table 3. 44 Table 3 Post-Concussion Scale Symptom Minor Moderate Severe Headache 1 2 3 4 5 6 Nausea 1 2 3 4 5 6 Vomiting 1 2 3 4 5 \u00E2\u0080\u00A2 6 Balance Problems 1 2 3 4 5 6 Dizziness 1 2 3 4 5 6 Fatigue 1 2 3- 4 5 6 Trouble Falling Asleep 1 2 3 4 5 6 Sleeping More Than Usual 1 2 3 4 5 6 Sleeping Less Than Usual 1 2 3 4 5 6 Drowsiness 1 2 3 4 5 6 Sensitivity to Light 1 2 3 4 5 6 Sensitivity to Noise 1 2 3 4 5 6 Irritability K 2 3 4 5 6 Sadness 1 2 3 4 5 6 Nervousness 1 2 3 4 5 6 Feeling More Emotional 1 2 3 4 5 6 Numbness or Tingling 1 2 3 4 5 6 Feeling Slowed Down 1 2 3 4 5 6 Feeling Mentally \"Foggy\" 1 2 3 4 5 6 Difficulty Concentrating 1 2 3 4 5 6 Difficulty Remembering 1 2 3 4 5 6 Visual Problems 1 2 3 4 5 6 Note: Participants checked a box i i \"they wer e\"not exp eriencing the sympl om.\" The sum of all responses on the Post-Concussion Scale was used to create a total post-concussion score (range = 22-132), with higher scores indicating a larger proportion of symptoms present during the test administration. For each item, scores ranged from 0-6 with 45 higher scores indicating more severe difficulties. The internal consistency of the entire scale as estimated with Cronbach's alpha was .92 (M =8.54, SD = 15.14) for the total sample, .93 (M= 11.47, SD = 16.09) for the A D H D group, and .90 ( M = 5.60, SD = 8.79) for the Control group. Section three of I m P A C T is composed of a battery of seven neuropsychological tests, referred to as modules. Each module contributes scores that produce four different composites (i.e., memory, processing speed, reaction time, and impulse control) that were used to assess specific aspects of cognitive functioning as described below. The breakdown of the scores that comprise each composite is provided in detail, after the descriptions of all modules. It is not possible to conduct internal consistency reliability analyses on the individual tests or the composite scores. This is because individual test responses cannot be downloaded from the program, and the composite scores are composed of a small number o f subtest scores (from the modules) and thus are not amenable to reliability analyses. Module I Word Discrimination. The first module evaluates attentional processes and verbal recognition memory by requiring the participant to discriminate between correct and incorrect words after two acquisition trials. Twelve target words are presented for 750 milliseconds each on the computer screen. The word list is presented twice in the same order at the same rate to facilitate learning. Immediately after the second presentation the participant is given a 24-word list that is composed of the twelve target words previously presented, and twelve non-target words. The target words are matched to the non-target words semantically (e.g., i f \"knife\" represents a target word, \"fork\" represents a non-target word). The participant responds to the words by mouse-clicking the \"yes and \"no\" buttons on the screen to specify whether the word presented was on the previously learned list. Subsequently, in a delayed condition that follows administration of all other test modules (approximately 20 minutes), this 46 task is re-administered using the same procedures as described above. There is no time limit for the immediate and delayed recognition portions during this module. For both the immediate and delayed assessment, the sum of correct and incorrect responses is computed. This module contributes to the memory composite (a total percent correct score is derived) with higher scores reflecting greater word learning and memory. The presentation of a word list in a visual format is similar to the Consortium to Establish a Registry for Alzheimer's Disease battery ( C E R A D ; Morris et al., 1989). This task is also conceptually similar to verbal list learning tasks, such as the California Verbal Learning Test ( C V L T ; Delis, Kramer, Kaplan, & Ober, 1987; Delis, Kramer, Kaplan, & Ober, 2000) and the Rey Auditory Verbal Learning Test ( R A V L T ; Rey, 1964). Module II Symbol Memory & Module III Color Click (distractor task). Symbol Memory measures visual working memory and visual processing speed. The Color Cl ick module serves as a distracter task, and is also a measure of focused attention, response inhibition, and reaction time. Prior to beginning the visual memory task (Symbol Memory), the participant is allowed to practice the distracter task (Color Click) . Color Cl ick is a choice reaction time test during which the participant is asked to click the left mouse button i f a red circle is presented and the right mouse button i f a blue square is presented. Once the participant has completed the practice task, the Symbol Memory task begins. For each of the trials of the memory task, a screen is displayed for 1.5 seconds that has a computer generated random assortment of X ' s and O's. Three of the X ' s or O's are illuminated in yellow on the screen. The participant is asked to remember the location of the illuminated objects. Immediately after the presentation of the three X ' s or O's the distracter task re-appears on the screen and distracter items (i.e., red circle or blue square) are presented for 30 seconds. Following the distracter task, the memory 47 screen ( X ' s and O's) re-appears and the participant is asked to click to identify the location of the previously illuminated 3 objects. The participant completes 4 trials involving presentation of the X ' s and O's, followed by the distracter task, followed by recall o f the location of the X ' s and O's. Scores are provided for the memory composite (correct identification of the X ' s and O's), reaction time composite (reaction time for the distracter task), impulse control composite (number of errors on the distracter task). The Symbol Memory component of this module is conceptually similar to the Spatial Location subtest from the Kaplan Baycrest Neurocognitive Assessment ( K B N A ; Leach, Kaplan, Rewilak, Richards, & Proulx, 2000). The task requires visual attention and visual-spatial working memory. The Color-Click task (distracter task) in this module is similar to the Connors' Continuous Performance Test (CPT; Conners, 2002), requiring speeded responding, impulse control, and sustained visual attention and vigilance. Module IV Symbol Match. The Symbol Match module evaluates visual processing speed, learning, and memory. Initially, the participant is presented with a screen that displays nine symbols (e.g., triangle, square, and arrow). Directly under each symbol is a number button from 1 to 9. Below this grid, a symbol is presented. The participant is required to click the matching number as quickly as possible, and to remember the symbol/number pairings. Correct performance is reinforced through the illumination of a correctly clicked number in green. Incorrect performance illuminates the number button in red. Following the completion of 27 trials, the symbols disappear from the top grid. The symbols again appear below the grid and the participant is asked to recall the correct symbol/number pairing by clicking the appropriate number button. This module provides an average processing speed score and a score for the memory condition. 48 The first part of this module resembles the Digit Symbol task from the Wechsler Adult Intelligence Scale, Third Edition (WAIS-III; Wechsler, 1997), and the Symbol Digit Modalities Test (Smith, 1972). Both of these tasks are underpinned by visual processing speed, visual scanning, and learning. The second part of the module resembles the incidental learning portion \u00E2\u0080\u00A2of the Digit Symbol task on the WAIS-III (Wechsler, 1997). Module V Color-Word Match. The Color-Word Match represents a choice reaction time task, and also measures impulse control and response inhibition. The first part of this test, a practice task, presents the participant with three squares of different colors (i.e, red, blue and green). The examinee is asked to click on either the red, blue, or green square as the word for that color appears on the screen. This process ensures that the participant can perform the basic task of matching a word to a color (e.g., match the word red, to the red square) ruling out colorblindness and grossly impaired reading ability. The actual test requires the examinee to click on the word (e.g., green) inside the box when it is a correct match between color and word (e.g., green word in a green ink). This is referred to as a congruent match. The examinee is required to inhibit or not respond when the word presented does not match the ink color. This is referred to as an incongruent match (e.g., the word green printed in red ink). A new stimulus (i.e., colored word in a box) is presented for two seconds with a one-second delay between the stimuli. In addition to providing a reaction time score, this task also contributes to the impulse control composite providing both omission (failing to click on a'congruent match) and commission error (clicking on an incongruent match) scores. This test measures impulse control and incidental reaction time. It relies on the examinee inhibiting an automatic word reading response in favor of a more novel response (i.e., identifying the ink color). It is very similar to the traditional Stroop test (Golden, 1978). 49 Module VI Sequential Digit Tracking/Trigram Memory. The Sequential Digit Tracking module measures working memory and visual-motor response speed. First, the participant is allowed to practice the distractor task, which consists of 25 numbered buttons ( 5 x 5 grid). The participant is instructed to click as quickly as possible on the numbered buttons in backward order starting with \"25.\" Once the participant has completed this initial practice task, he/she is presented with three consonant letters that are displayed on the screen and instructed to remember them. Immediately following display of the three letters, the numbered grid re-appears and the participant is instructed to click the numbered buttons in backward order as quickly as possible. After a period of 18 seconds, the numbered grid disappears and the participant is asked to recall the three letters by typing them from the keyboard. Both the number placement on the grid and letters displayed are randomized for each trial.. Five trials of this task are presented for each administration of the test. This module yields a memory score (total number of correctly identified letters) and a processing speed score (average number of correctly clicked numbers per trial from the distractor test). The three-letter task is similar to the Brown-Peterson short-term memory paradigm (Brown, 1958; Peterson & Peterson, 1959); it is also called the Auditory Consonant Trigrams Test (Mitrushina, Boone, Razani, & D'El ia , 2005). The speeded distractor task is conceptually similar to the Trail Making Test-A ( T M T - A ; Reitan & Wolfson, 1993), which is a visual motor task involving scanning and graphomotor speed. Module VII Visual Attention Span. The Visual Attention Span module evaluates visual attention span under two conditions: forward span and backward span. During the forward span task, the examinee is presented with a 3 x 3 grid of square buttons. The buttons are highlighted in random order. The examinee is required to remember the order and mouse click on the 50 correct sequence. Following a sample item, four trial sequences are presented. Each sequence involves more grid items to be repeated, with the final trial including nine squares. The backward span task is identical to the forward span condition, except that the participant is required to click on the presented sequence in backward order. The task begins with a sequence of two highlighted squares within the grid, and progresses until the participant reaches a maximum of eight squares to remember. Both the forward and backward component are discontinued once the participant fails two trials in a row at any level. Two scores from this module are calculated, which contribute to the memory and processing speed composites. This task is modeled on, and essentially identical in nature to, the WMS-I I I Spatial Span task. ImPACT Composite Scores. Performance across tasks on I m P A C T yielded four overall composite scores for each participant: Memory Composite, Reaction Time Composite, Processing Speed Composite, and Impulse Control Composite. The breakdown of the module scores that contribute to each composite is provided below: 1. The Memory Composite is comprised of the average of the following scores: (a) Word Discrimination total percent correct, (b) Symbol Match-Total correct hidden symbols, (c) Sequential Digit Tracking total percent of total letters correct, (d) Visual Attention Span- Total percent of numbers correct (forwards and backwards), and (e) Symbol Memory total percent of X ' s and O's correct. 2. The Reaction Time Composite is comprised of the average o f the following scores: (a) Symbol Memory X ' s and O's-Average correct R T (interference), (b) Symbol Match-Average correct RT/3 and, (c) Color Click-Average, correct R T . 51 3. The Processing Speed Composite is comprised of the average of following scores: (a) Symbol Memory-total correct (interference)/4, (b) Sequential Digit Tracking Three-letters-Average counted correctly*3, and (c) Visual Attention Span. 4. The Impulse Control Composite is comprised of the average of the following scores: (a) Symbol Memory-total incorrect- (interference), and (b) Color Match total commissions. Analyses The dependent variables were first examined for skewness and kurtosis, and transformations were performed on any variables that violated the assumptions of normality. Bivariate correlations (Pearson) among the composite variables of I m P A C T were calculated to establish the degree of association among the dependent variables. In order to evaluate whether the matched groups ( A D H D and non -ADHD) differed across the six dependent variables evaluated in this study, dependent t-tests were conducted for each of the variables (i.e., Post-Concussion Scale, Memory Composite, Reaction Time Composite, Impulse Control, and Processing Speed Composites). The dependent t-test is the most appropriate (i.e., robust) calculation to test the null hypotheses in a matched groups design. Statistically, the dependent t-test is almost identical to the independent t-test, except that it takes into account the degree of correlation between the two groups. Large correlations between the two groups on the dependent measures reduces the size of the error variance, making the t-test more powerful. Effect sizes for each comparison are reported using the original (untransformed) means and standard deviations for the A D H D and control group. In addition, analyses are conducted to determine whether self-reported academic problems or participation in special education was related to performance on I m P A C T (i.e., participants 52 taking special classes for reading). Lastly, for exploratory purposes, the three symptoms from the Post-Concussion Scale dealing with cognitive difficulties (feeling mentally foggy, poor concentration, and poor memory) were combined to examine whether or not the A D H D subjects endorse significantly greater cognitive difficulties than the control group. Results The descriptive statistics for the composite scores are presented in Table 4. Several variables violated assumptions of normality, and showed significant skewness and kurtosis including the Post-Concussion Scale, Memory Composite, and Impulse Control Composite. Variables were deemed to exhibit significant skewness and/or kurtosis i f the z-scores associated with these indices were outside the range of +/-3. Variables with a significant Kolmogorov-Smirnov statistic (p < .05) were considered to violate assumptions of normality. To correct for these violations of normality, these variables were transformed using the square root method as an alternative to the logarithmic transformation because some of the data points were 0, and therefore undefined in a logarithmic transformation. Instead of adding a constant of 1 to these variables, the more conservative square root method of transformation was applied. Square root transformation of the variables did not alter the significance of any of the relationships among the data on the dependent t-tests. A s a result, the means and standard deviations of the untransformed data were used for all analyses. This is preferable, because the square root transformation of the variables alters their natural distribution, general by artificially compressing high data points in a non-systematic way (e.g., a participant score might greatly exceed their matched control's score, resulting in the scores being altered in a 53 non-systematic way). Furthermore the t-test is relatively robust to violations of assumptions, especially when sample sizes are above twenty (Tabachnick & Fidell , 2001). Table 4 Descriptive statistics for the ImPACT composite scores Mean Standard Deviation Interquartile Range Skewness Kurtosis K S Symptoms 8.54 13.25 0.0-10.00 2.26 4.89 .00 Memory 86.24 10.84 79.53 - 95.49 -1.07 1.41 .00 Reaction Time .58 .07 .53- .62 .62 .37 .20* Processing Speed 34.23 7.02 29.71 -38.80 .09 .29 .20* Impulse Control 10.34 8.04 5.0-13.0 1.92 5.15 .00 Note: K S = Kolmogorov-Smirnov test of normality; * = Significant violations of normality. In assessing for univariate outliers in the data, the standardized values revealed that several cases were potential outliers (z > +/-3). These cases were further assessed by an examination of the histograms, stem and leaf plots, box plots, and the raw data itself. A l l potential outliers appeared to be connected to their respective distributions, and were therefore retained as legitimate values. Intercorrelations among the dependent variables are presented in Table 5. Correlations greater than 0.9 violate assumptions related to multicollinearity and singularity. However, the bivariate correlations among the dependent measures in the present sample were small to medium. Accordingly, each of the six dependent measures was considered separately in subsequent analyses. 54 Table 5 Pearson's correlation coefficients among the ImPACT composite scores Symptoms Memory Reaction time Impulse control Processing Speed Symptoms Memory Reaction time Impulse Control Processing Speed -.24** -.07 -.35** .15* -.16* -.08 -.15* .30** -.50** -.03 Correlat ion is significant at the 0.01 level (1 -tailed). **Correlation is significant at the 0.05 level (1-tailed). Dependent t-tests were conducted to evaluate differences between the adults with A D H D and the matched control group across all five dependent variables (Post-Concussion Scale, Memory Composite, Processing Speed Composite, Impulse Control Composite, Reaction Time Composite. For exploration purposes, independent t-tests were also run, but not reported \ Results of these analyses indicated that the individuals with A D H D and the control group differed significantly on the Post-Concussion Scale ( t (1, 33) = -2.46, p < .05), and the Memory Composite (t (1, 33) = 2.88, p < .05). The groups did not differ significantly on the Processing Speed Composite (t (1, 33) = .727, p > .05) or the Impulse Control Composite (t (1, 33) = -.866,/? > .05). Differences between A D H D and Control groups approached significance for the Reaction Time Composite (t (1, 33) = -.178,/? > .07). A s reported in Table 6, the effect sizes for the significant differences were medium. A n examination of the means for these analyses (see Table 6) indicated that young adults with A D H D report more symptoms (Post-Concussion Scale) and demonstrate a poorer ability across memory tasks. There was a 1 The same pattern of results were obtained when Independent t-tests were used rather than dependent t-tests. 55 nonsignificant trend for the individuals with A D H D to display slower reaction time than controls. Table 6 Descriptive statistics, significance tests, and effect sizes (Cohen's d) A D H D N o n - A D H D Controls M S D M SD E d Symptoms 11.47 16.09 5.60 8.87 .010 .68 Memory 83.59 10.73 88.88 10.34 .004 .50 Reaction Time .58 .06 -> .56 .06 .076 .33 Processing Speed 33.79 7.37 34.65 6.67 .479 .12 Impulse Control 10.86 8.05 9.80 8.05 .445 .13 Given that a significant number of individuals in the A D H D group reported academic difficulties, additional analyses were conducted to determine whether self-reports academic problems or participation in special education systematically affected performance on I m P A C T . Due to small sample sizes for specific educational problems, participants with A D H D were sorted into binary groups: those with one or more academic problems (i.e., reading, spelling, math, repeated grade, learning assistance, or special education) versus those with no self-reported academic problems. There were 31 participants with academic problems and 37 without problems. The two groups did not differ on total symptoms, The Memory Composite, Processing Speed Composite, or the Impulse Control Composite. A D H D subjects with academic problems had slower reaction times, however [t (66) = -.26, p <.012, d = .64.]. 56 For exploratory purposes, the three specific symptoms from the Post-Concussion Scale dealing with cognitive difficulties (feeling mentally foggy, poor concentration, and poor memory) were summed to create a single score (See Table 7). The A D H D subjects endorsed significantly greater cognitive problems ( M = 2.51, SD = 3.84) than the control subjects [ M = .88, SD = 2.0; t (67) = -2.98, p < ,005, d = .56]. Frequency distributions for the three scores that were significantly different between the A D H D group and the controls were examined and cutoff scores were selected. These cutoff scores represented the approximate 10 t h percentile for the control group. That is, 90% or more of the control group scored better than the cutoff. Specifically, the cutoff score for the Post-Concussion Scale was > 15 points, the cognitive symptom total score was > 3 points, and the Memory Composite was < 76.9% correct. These three cutoff scores were then examined, in combination, to determine i f they could reasonably separate the two groups. These results are presented in Table 7. Notice that 82% of the control subjects did not have a single unusual score, compared to 56% of the A D H D sample. Applying a decision rule of one or more unusual scores would result in a correct classification rate of 44.1% of the A D H D subjects and 82.4% of the controls. Applying a decision rule of two or more unusual scores would result in a correct classification rate of 23.5% for the A D H D group and 92.6% for the controls. Table 7 Percent of subjects with unusual scores A D H D Group , Control Group Number of Cumulative Cumulative Unusual Scores Percent Percent Percent Percent 0 55.9 55.9 82.4 82.4 1 20.6 76.5 10.3 92.6 2 16.2 92.6 5.9 98.5 3 7/4 100.0 , 1.5 100.0 Note: \"Unusual\" scores occur in less than 10% of the control group for each of the three variables. 57 Discussion Increasingly there is recognition that many symptoms of A D H D persist into adulthood (Epstein et al., 2001; Murphy, 2002a). However, because it is a diagnosis made based on childhood history, retrospective diagnosis in adulthood has proven challenging. N o \"gold-standard\" test exists that has been shown to reliably differentiate adults with A D H D from those without A D H D . Recently, research has examined the utility of neuropsychological tests for identifying differences between persons with and without A D H D (Johnson et al., 2001; Murphy, 2002b; N igg et al., 2002). The findings have tended to be mixed: The most consistent results have emerged from the childhood literature (e.g., Barkley et al., 1992; Snow, 1998); somewhat more variable results with adolescents (e.g., Schmitz et al., 2002; Seidman et al., 1997); and highly variable results with the adult A D H D population (e.g., Murphy et al., 2001; Seidman et al., 1998). Further complicating this literature is the fact that the construct \"executive functioning\" remains poorly operationalized in most of the neuropsychological literature, and executive functions such as planning, organizational skills, judgment, and inhibitory control are difficult to separate from other cognitive functions that underpin them. There is little consistency in the neuropsychological tests used, although most fall under the general category o f tests of executive functioning. Use of a battery of neuropsychological tests rather than individual measures increases the ability of neuropsychological tests to discriminate persons with A D H D from those without A D H D (e.g., Woods, Lovejoy, Stutts et al., 2002). However, administration of a test battery is both time consuming and expensive, and as a result generally not a viable or cost effective way of screening large numbers of individuals. Research continues to investigate 58 neuropsychological measures with a continued focus on ways to, rapidly differentiate persons with ADHD from those without it. ImPACT version 1.0 is a brief, computer administered and scored, neuropsychological battery initially constructed for rapid evaluation of post-concussive cognitive problems and symptoms in athletes. Recently, ImPACT Version 2.0 demonstrated some promise in a preliminary study examining its utility in differentiating adolescents with ADHD from non-ADHD controls (Iverson & Strangway, 2004). ImPACT is brief and easily administered. As a result, it represents a cost-effective solution for screening large numbers of adults for ADHD. The present study examined the utility of ImPACT Version 1.0 for differentiating young adults with ADHD from matched non-ADHD controls on four composite scores and the Post-Concussion Scale. There were significant differences between adults with ADHD and non-ADHD controls on two of the five variables considered, the Post-Concussion Scale and the Memory Composite, with a nonsignificant trend toward slower reaction times in the ADHD group. No significant differences were found on the Impulse Control Composite, or the Processing Speed Composite. Each of these is discussed below in light of relevant research. As predicted, young adults with ADHD were found to perform significantly more poorly on the ImPACT Memory Composite score than matched non-ADHD controls. The Memory Composite produced by ImPACT version 1.0 is composed of five subtest scores measuring different aspects of memory (e.g., verbal learning, visual learning, incidental learning, and working memory). Tasks of verbal learning and memory have been investigated in the adult ADHD literature, and have yielded the most consistent results in differentiating adults with ADHD from non-ADHD controls (e.g., Seidman et al., 1998; Woods, Lovejoy, Stutts et al., 2002). Adults with ADHD appear to have deficits in the encoding and retrieval of 59 verbal information, primarily related to executive aspects of efficient memory strategy and verbal organization skills (e.g., semantic versus phonemic chunking of information) (Woods, Lovejoy, & Ba l l , 2002). Research on differentiating adults with A D H D has not demonstrated consistent results using visual memory tasks (e.g., Murphy, 2002a). Because individual memory subtests that comprise the Memory Composite, were not readily available, it was not possible to determine which aspects of memory function contributed to the ability of the Memory Composite to reliably distinguish participants with A D H D from the matched controls. A s predicted, the reaction time score for the young adults with A D H D was not significantly different than the normative comparison group. This prediction was based on previous findings that tests of simple reaction time and incidental reaction time (e.g., extracted from the CPT) have not proven useful in differentiating adults with A D H D from controls (Murphy, 2002a). Despite the fact that the literature does not appear to support reaction time differences between n o n - A D H D controls and individuals with A D H D , a trend toward a significant effect was observed in the present study, with young adults with A D H D tending to show slower reaction times than their matched controls. Moreover, when the A D H D group was divided into those with versus without self-reported academic problems, those with academic problems had significantly slower reaction times. Similarily, Johnson et al. (2001) found that adults with A D H D performed more slowly than n o n - A D H D controls on a reaction time task (3RT), as the task became more complex. Results from Iverson and Strangway (2004) using I m P A C T version 2.0 revealed differences on the I m P A C T Processing Speed composite score between adolescents with A D H D and n o n - A D H D controls. Accordingly, it was expected that, in the present study o f young adults, the I m P A C T processing speed composite might also discriminate between young 60 adults with A D H D and n o n - A D H D controls. However, the results did not reveal slower processing speed in the A D H D group. One possibility for the failure of this study to discriminate adults with A D H D from controls on the basis of processing speed is that, similar to many of the investigations of processing speed in the literature, only the child and adolescent groups differed significantly from n o n - A D H D controls, while young and middle aged adults with A D H D did not (Seidman et al., 1998; Walker et al., 2000). Adults with A D H D may have found ways to compensate or adapt to their cognitive deficits with age. This does not \u00E2\u0080\u00A2 necessarily mean that there are \"real\" cognitive improvements, but rather that the neuropsychological tests are unable to capture these subtle deficits in adults. Unexpectedly, the I m P A C T Impulse Control composite did not differentiate the adults with A D H D from the matched controls. However, Iverson and Strangway (2004) using I m P A C T version 2.0 reported significant differences between adolescents with A D H D and controls. Two traditional neuropsychological tests, the C P T and Stroop test, have typically been used to measure impulse/inhibitory control in the A D H D literature. I m P A C T contains tests almost identical to the Stroop and C P T (Color Word Match & Color Cl ick, respectively). Both of these.tests have a large inhibitory control component, and have been frequently and successfully used to differentiate between children with A D H D and n o n - A D H D controls. Within the adult population, the C P T has also been generally successfully at distinguishing adults with A D H D from controls on the basis of number of commission errors (i.e., responding to a target stimuli when withholding o f a response is required) - an index of inhibitory control (Epstein et al., 1998; Murphy et al., 2001). In comparison, the Stroop task has primarily been successful only with the adolescent population (Schmitz et al., 2002; Seidman et al., 1997). It is possible that a similar pattern of results was found on the I m P A C T Impulse Control composite, 61 which is composed of an average of the two scores (i.e., commission errors on Color Word Match & total incorrect/interference on Color Cl ick) . It is possible that good differentiation between groups on one of the scores was diluted to the point of non-significance by a non-discriminating result on the other score. A s mentioned in the context of the Memory Composite, the extraction and examination of the individual test scores was not readily available for this study. More importantly, the test instructions and practice items for one or more of the tasks that comprise the Impulse Control Composite were revised and clarified for Version 2.0 of I m P A C T . This might have increased the usefulness of this composite in the previous study. Finally, as predicted, young adults with A D H D in the present sample reported significantly more symptoms on the Post-Concussion Scale than matched controls. They also reported greater difficulty with the cognitive symptoms on that scale. This finding is consistent with the literature that adults with A D H D report more psychiatric symptoms than healthy controls (e.g., feeling down, feeling irritable, feeling depressed) (Woods, Lovejoy, & Ba l l , 2002). Moreover, the scale includes items that specifically relate to A D H D symptomatology (e.g., trouble concentrating). Throughout this thesis, I have discussed the \"differentiation\" of people with A D H D from n o n - A D H D control subjects. It is important to emphasize that this means statistical differentiation, not practical differentiation. Indeed, as emphasized throughout this thesis, the practical differentiation of people with A D H D based on neuropsychological tests has been difficult and elusive. Finding a statistically significant difference between two groups on a test does not, o f course, mean that the test can differentiate individuals on a case-by-case basis. This is illustrated in Figure 1. 62 Using IQ scores as the metric of interest, average scores are 100 with a standard deviation of 15. Therefore, the vast majority of subjects would score within two SDs from the mean (30 points). Differences between a clinical group and a \"normal\" group are illustrated using effect sizes ranging from .2 (small) to 1.5 (large). A s seen in this figure, even a \"large\" effect size of .8 results in tremendous overlap between a clinical group and a \"normal\" distribution. This, o f course, makes it very difficult to accurately differentiate individuals within groups based on a test score. Figure 1. Overlapping distributions based on effect sizes (using the IQ metric). 140 130 120 110 100 90 80 70 60 50 40 Average 0.2 0.5 0.8 1 1.5 Note: This figure illustrates the theoretical overlap between cl inical groups that differ from \"normal\" by certain magnitudes o f effect sizes. Conclusions The pattern of results produced by this investigation of I m P A C T is consistent with the adult A D H D literature in that the I m P A C T test battery did not consistently demonstrate differences across all neuropsychological tests between individuals with A D H D compared to controls (e.g., Corbett & Stanczak, 1999; Kovner et al., 1998; Walker et al., 2000). Adults with A D H D performed significantly more poorly than controls on the Memory Composite, and 63 showed a nonsignificant trend towards slower reaction times on the Reaction Time Composite, but did not differ on the Processing Speed or Impulse Control Composites. The nature o f these results is similar to the reviewed literature that has found the most promising results on memory tasks (e.g., Seidman et al., 1998; Woods, Lovejoy, Stutts et al., 2002). The lack of , consistent results across the I m P A C T composites likely results from a number of factors. The biggest factor appears to be problems with the test itself. Iverson and Strangway (2004) reported much stronger findings using version 2.0 of the test, especially for the Impulse Control Composite (the effect size was large, d = .93). Improvements to the administration instructions and practice items from version 1.0 to version 2.0 might account for the different results. Another major factor relating to the lesser differences between groups in this study was the nature of the sample. Most of the subjects were in university, whereas in Iverson and Strangway (2004) all were in high school. It stands to reason that those people with A D H D who go to university have less pronounced cognitive difficulty, as a group, than those who do not. Another factor is that as adults with A D H D age, they find strategies to compensate for potential attentional deficits, and therefore some tests become less able to discriminatetheir performance as they get older and adapt better. This does not suggest that attention deficits necessarily disappear with age, but rather that they become more difficult to identify with cognitive measures. Further, the disorder is variable by nature, and adults with A D H D do not manifest cognitive deficits equally under all conditions. One of the noted hallmarks of A D H D is poorer performance as extraneous distractions increase (Woods, Lovejoy, & B a l l , 2002). Completing a test battery under quiet and controlled conditions is likely not a good algorithm for functional 64 performance under \"real world\" conditions. Thus, it is not reasonable to assume that these deficits can always be reliably identified in the testing environment. Further, ADHD is associated with day-to-day variability in performance so one testing session during one period of time may also not capture the full nature of their deficits. Limitations There were a number of limitations to the current study. First, the self-report method of identifying the ADHD group introduces retrospective bias and although the literature appears to support self-report by individuals with ADHD, the bias associated with self-report could not be determined. Using a single diagnostic indicator such as a self-report with no cross-validation of history or symptoms from multiple settings or informants is not ideal. However, given the lack of a gold standard method for diagnosing ADHD, this study, like many others (Epstein et al., 1998; Epstein et al., 2001; Murphy, 2002a, 2002b; Nigg et al., 2002) relied by necessity on self-reported ADHD symptoms for group classification. In the present study, this problem was slightly mitigated because of the young age of the sample. It is reasonable to suppose that young adults aged 15-22 would have been diagnosed more recently than those in other studies where participants were up to 89 years of age. Further, some of the older participants might not have been diagnosed at all in childhood, because the diagnosis might not have been identified or have become well-known to clinicians, given its recent inclusion in the DSM. In contrast, there is a strong likelihood that the present sample of young adults would have received a diagnosis of ADHD within the past 10 years (i.e., in the 1990's). Further, this sample received no obvious gain from participating in this study, and they had no reason to mislead the researcher. 65 Because this was a sample of convenience, derived from a normative database, the individuals with A D H D were not classified or separated according to D S M - I V - T R subtypes. The majority of previous studies have not differentiated their samples on this basis, and research to date has not consistently identified different cognitive profiles among the A D H D subtypes (e.g., Epstein et al., 2001; Kovner et al., 1998; Nigg et al., 2002; Walker et al., 2000). However, further investigation of the subtypes in terms of identifying whether their I m P A C T performance profiles differed might be interesting. A contentious issue is whether or not to control for intelligence. In the present study, this variable could not be analyzed because it was not collected in the normative database. Further, many researchers have suggested that controlling for intelligence might remove meaningful variance associated with A D H D (Seidman et al., 1997). Seidman et al. (1997) contend that using intelligence as a covariate constitutes \"overcontrol,\" thereby limiting the possibility of finding significant differences between adults with A D H D and controls. In the studies reviewed, the majority that controlled for intelligence found no differences on neuropsychological tests between adults with A D H D and controls (Kovner et al., 1998; Walker et al., 2000). A final limitation of the present study relates to the participants' use of medication. In most of the reviewed studies, adults with A D H D were either not taking medication, were taken off stimulant medications prior to testing, or the studies implemented statistical procedures to control for the possible cognitive effects of medication (e.g., Corbett & Stanczak, 1999; Epstein et al., 2001; Seidman et al., 1998). In the present study, the effects of medication (stimulant or other) could not be examined because specific medication information was not collected in the normative database. However, researchers exploring the use of stimulant medications on 66 cognitive performance have found conflicting results. Schmitz et al. (2002) found that unmedicated participants with A D H D performed more poorly than those on stimulant medications. In contrast, Seidman et al. (1997) found no performance differences on cognitive measures between unmedicated and medicated participants with A D H D . Future research investigating whether the profiles of individuals with A D H D differ according to whether they are using stimulant medications would be valuable. Future Directions I m P A C T is a new test. Developing a new test takes time, money, and most importantly feedback from both clinicians and researchers. V i a such feedback the test has undergone revisions resulting in I m P A C T version 2.0. Future researchers using this test should examine the individual subtests, not just the composite scores. Moreover, the ease of administration including time, cost-efficiency, and limited examiner training requirements, suggests that a large number o f individuals with A D H D or other disorders could be tested rapidly. With larger samples some of the current methodological limitations of the reviewed literature could be addressed with ease. Future research could examine the differential neuropsychological performance o f individuals within constrained age ranges in order to identify i f there are primary cognitive differences between the age groups. Longitudinal studies are needed with repeated evaluation over time to improve on the cross-sectional literature that currently dominates the research in this area. One of the advantages of I m P A C T is its automatic randomization of stimuli to allow for repeated testing with minimal practice effects. Larger sample sizes also allow for research relating to cognitive differences between A D H D subtypes. The battery approach appears to improve the ability of cognitive tests to discriminate the performance of adults with A D H D from n o n - A D H D controls. Continued 67 investigation of the battery approach to A D H D differential diagnosis appears to be the most likely route to creating a reliable diagnostic method for the deficits associated with A D H D throughout the lifespan. 68 References Alexander, M . P., & Stuss, D . T. (2000). Disorders of frontal lobe functioning. Seminars in Neurology, 20(4), All'-437. American Psychiatric Association. (1968). Diagnostic and statistical manual of mental disorders (2nd ed.). Washington, D C : Author. American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders (3rd ed.). Washington, D C : Author. American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (Revised). Washington, D C : Author. American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, D C : Author. American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (Text Revision). Washington, D C : Author. Applegate, B . , Lahey, B . B . , Hart, E . L . , Biederman, J., Hynd, G . W. , Barkley, R. A . , et al. (1997). Validity of the age-of-onset criterion for A D H D : a report from the D S M - I V field trials. Journal of the American Academy of Child and Adolescent Psychiatry, 36(9), 1211-1221. Barkley, R. A . (1997). Behavioral inhibition, sustained attention, and executive functions: constructing a unifying theory of A D H D . Psychological Bulletin, 121(1), 65-94. Barkley, R. A . (2000). Genetics of childhood disorders: X V I I . A D H D , Part 1: The executive functions and A D H D . Journal of the American Academy of Child and Adolescent Psychiatry, 39(%), 1064-1068. 69 Barkley, R. A . , & Biederman, J. (1997). Toward a broader definition of the age-of-onset criterion for attention-deficit hyperactivity disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 36(9), 1204-1210. Barkley, R. A . , DuPaul, G . J. , & McMurray, M . B . (1990). Comprehensive evaluation of attention deficit disorder with and without hyperactivity as defined by research criteria. Journal of Consulting and Clinical Psychology, 58(6), 775-789. Barkley, R. A . , Grodzinsky, G . , & DuPaul, G . J. (1992). Frontal lobe functions in attention deficit disorder with and without hyperactivity: a review and research report. Journal of Abnormal Child Psychology, 20(2), 163-188. Benezra, E . , & Douglas, V . I. (1988). Short-term serial recall in A D D H , normal, and reading-disabled boys. Journal of Abnormal Child Psychology, 16(5), 511 -525. Brown, J. (1958). Some tests of decay of immediate memory. Quarterly Journal of Experimental Psychology, 10, 12-21. Carlson, C. L . , Lahey, B . B . , & Neeper, R. (1986). Direct assessment of the cognitive correlates of attention deficit disorders with and without hyperactivity. Journal of Psychopathology and Behavioral Assessment, 8, 69-86. Carte, E . T., Nigg , J. T., & Hinshaw, S. P. (1996). Neuropsychological functioning, motor speed, and language processing in boys with and without A D H D . Journal of Abnormal Child Psychology, 24(4), 481-498. Collins, M . W. , Field, M . , Lovel l , M . R., Iverson, G . , Johnston, K . M . , Maroon, J. , et al. (2003). Relationship between postconcussion headache and neuropsychological test performance in high school athletes. American Journal of Sports Medicine, 31(2), 168-173. 7 0 Collins, M . W. , Lovel l , M . R., Iverson, G . L . , Cantu, R. C , Maroon, J. C , & Field, M . (2002). Cumulative effects of concussion in high school athletes. Neurosurgery, 57(5), 1175-1179; discussion 1180-1171. Conners, C. K . (2002). Couriers' Continuous Performance Test II. Toronto, O N : Mul t i Health Systems. Corbett, B . , & Stanczak, D . E . (1999). Neuropsychological performance of adults evidencing Attention-Deficit Hyperactivity Disorder. Archives of Clinical Neuropsychology, 14(4), 373-387. Culbertson, J. L . , & K r u l l , K . K . (1996). Attention deficit hyperactivity disorder. In R. L . Adams, O. A . Parson, J. L . Culbertson & S. J. N ixon (Eds.), Neuropsychology for practice (pp. 271-330). Washington, D C : American Psychological Association. Delaney, J. S., Lacroix, V . J., Gagne, C , & Antoniou, J. (2001). Concussions among university football and soccer players: a pilot study. Clinical Journal of Sports Medicine, 11(4), 234-240. Delis, D . C , Kramer, J. H . , Kaplan, E . , & Ober, B . A . (1987). California Verbal Learning Test: Adult Version. San Antonio, T X : The Psychological Corporation. Delis, D . C , Kramer, J. H . , Kaplan, E . , & Ober, B . A . (2000). California Verbal Learning Test (CVLT-II) Second Edition: Adult Version. U S A : Psychological Corporation. Douglas, V . I. (1972). Stop, look, and listen: The problem of sustained attention and impulse control in hyperactive and normal children. Canadian Journal of Behavioral Science, 4, 259-282. 71 Douglas, V . I., & Parry, P. A . (1983). Effects of reward on delayed reaction time task performance of hyperactive children. Journal of Abnormal Child Psychology, 11(2), 313-326. Epstein, J. N . , Conners, K . , Sitarenios, G . , & Erhardt, D . (1998). Continuous performance test results of adults with attention deficit hyperactivity disorder. The Clinical Neuropsychologist, 12, 155-168. Epstein, J. N . , Johnson, D . E . , Varia, I. M . , & Conners, C. K . (2001). Neuropsychological assessment of response inhibition in adults with A D H D . Journal of Clinical and Experimental Neuropsychology, 23(3), 362-371. Erlanger, D . M . , Kutner, K . C , Barth, J. T., & Barnes, R. (1999). Neuropsychology of sports-related head injury: Dementia Pugilistica to Post Concussion Syndrome. Clinical Neuropsychology, 13(2), 193-209. Faraone, S. V . , & Biederman, J. (1998). Neurobiology of attention-deficit hyperactivity disorder. Biological Psychatry, 44(10), 951 -958. Fischer, M . , Barkley, R. A . , Edelbrock, C. S., & Smallish, L . (1990). The adolescent outcome of hyperactive children diagnosed by research criteria: II. Academic, attentional, and neuropsychological status. Journal of Consulting and Clinical Psychology, 58(5), 580-588. Gallagher, R., & Blader, J. (2001). The diagnosis and neuropsychological assessment of adult attention deficit/hyperactivity disorder. Scientific study and practical guidelines. Annals of the New York Academy of Sciences, 931, 148-171. 72 Gaub, M., & Carlson, C. L. (1997). Gender differences in ADHD: a meta-analysis and critical review. Journal of the American Academy of Child and Adolescent Psychiatry, 36(8), 1036-1045. Golden, C. J. (1978). Diagnosis and rehabilitation in clinical neuropsychology. Springfield, IL: Charles C. Thomas. Hageman, W. J., & Arrindell, W. A. (1993). A further refinement of the reliable change (RC) index by improving the pre-post difference score: introducing RCID. Behavior Research and Therapy, 31(1), 693-700. Halperin, J. M., Newcorn, J. H., Sharma, V., Healey, J. M., Wolf, L. E., Pascualvaca, D. M., et al. (1990). Inattentive and noninattentive ADHD children: do they constitute a unitary group? Journal of Abnormal Child Psychology, 18(4), 437-449. Halperin, J. M., Sharma, V., Greenblatt, E., & Schwartz, S. T. (1991). Assessment of the Continuous Performance Test: Reliability and validity in a nonreferred sample. Psychological Assessment, 3, 603-608. Heaton, R. K., Chelune, G. J., Talley, J. L., Kay, G. G., & Curtis, G. (1993). Wisconsin Card Sorting Test (WCST) Manual Revised and Expanded. Odessa, FL: Psychological Assessment Resources. Heaton, R. K., Temkin, N., Dikmen, S., Avitable, N., Taylor, M. J., Marcotte, T. D., et al. (2001). Detecting change: A comparison of three neuropsychological methods, using normal and clinical samples. Archives of Clinical Neuropsychology, 16(\), 75-91. Heilbronner, R. L., Henry, G. K., Buck, P., Adams, R. L., & Fogle, T. (1991). Lateralized brain damage and performance on trail making A and B, Digit Span Forward and Backward, and TPT Memory and Location. Archives of Clinical Neuropsychology, 6(4), 251-258. 73 Iverson, G . L . , Gaetz, M . , Lovel l , M . R., & Collins, M . W . (2004a). Cumulative effects of concussion in amateur athletes. Brain Injury, 18(5), 433-443. Iverson, G . L . , Gaetz, M . , Lovel l , M . R., & Collins, M . W . (2004b). Relation between subjective fogginess and neuropsychological testing following concussion. Journal of the International Neuropsychological Society, 10(6), 904-906. Iverson, G . L . , Lovel l , M . R., & Collins, M . W . (2002). Validity of ImPACT for measuring the effects of sports-related concussion. Paper presented at the National Academy of Neuropsychology Annual Conference, M i a m i , F L . Iverson, G . L . , Love l l , M . R. , & Collins, M . W . (2003). Interpreting change on I m P A C T following sport concussion. Clinical Neuropsychology, 17(4), 460-467. Iverson, G . L . , Lovel l , M . R., & Collins, M . W . (2005). Validity of I m P A C T for measuring processing speed following sports-related concussion. Journal of Clinical and Experimental Neuropsychology, 27, \-l. \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 Iverson, G . L . , Love l l , M . R., Collins, M . W. , & Norwig, J. (2002). Tracking recovery from concussion using I m P A C T : Applying reliable change methodology. Archives of Clinical Neuropsychology, 17, 770. Iverson, G . L . , Lovel l , M . R. , Podell, K . , & Collins, M . W . (2003). Reliability and validity of ImPACT. Paper presented at the Annual Conference of the International Neuropsychological Society, Honolulu, H A . Iverson, G . L . , & Strangway, C . L . (2004). Computerized screening in adolescents with ADHD. Paper presented at the National Association of School Psychologist Conference, Dallas, T X . 74 Jacobson, N . S., & Truax, P. (1991). Clinical significance: a statistical approach to defining meaningful change in psychotherapy research. Journal of Consulting and Clinical Psychology, 59(1), 12-19. Johnson, D . E . , Epstein, J. N . , Waid, L . R., Latham, P. K . , Voronin, K . E . , & Anton, R. F . (2001). Neuropsychological performance deficits in adults with attention deficit/hyperactivity disorder. Archives of Clinical Neuropsychology, 16(6), 587-604. Kempton, S., Vance, A . , Maruffi P., Luk, E . , Costin, J., & Pantelis, C . (1999). Executive function and attention deficit hyperactivity disorder: stimulant medication and better executive function performance in children. Psychological Medicine, 29(3), 527-538. Kle in , M . , Ponds, R. W. , Houx, P. J., & Jolles, J. (1997). Effect of test duration on age-related differences in Stroop interference. Journal of Clinical and Experimental Neuropsychology, 19(1), 77-82. Konrad, K . , Gauggel, S., Manz, A . , & Scholl, M . (2000). Inhibitory control in children with traumatic brain injury (TBI) and children with attention deficit/hyperactivity disorder ( A D H D ) . Brain Injury, 14(10), 859-875. Kovner, R., Budman, C , Frank, Y . , Sison, C , Lesser, M . , & Halperin, J. (1998). Neuropsychological testing in adult attention deficit hyperactivity disorder: a pilot study. International Journal of Neurosciences, 96(3-4), 225-235. Kupietz, S. S. (1990). Sustained attention in normal and in reading-disabled youngsters with and without A D D H . Journal of Abnormal Child Psychology, 18(4), 357-372. Leach, L . , Kaplan, E . , Rewilak, D . , Richards, B . , & Proulx, G . B . (2000). Kaplan Baycrest Neurocognitive Assessment. San Antonio, T X : The Psychological Corporation. 75 Levin, M . (1998). Definition of A D H D . Journal of the American Academy of Child and Adolescent Psychiatry, 57(4), 343-345. Lezak, M . D . (1995). Neuropsychological Assessment (3rd Edition) A compendium of neuropsychological tests. New York, N Y : Oxford University Press. Loge, D . V . , Staton, R. D. , & Beatty, W . W . (1990). Performance of children with A D H D on tests sensitive to frontal lobe dysfunction. Journal of the American Academy of Child and Adolescent Psychiatry, 29(4), 540-545. Lovel l , M . R., Collins, M . W. , Iverson, G . L . , Field, M . , Maroon, J. C , Cantu, R., et al. (2003). Recovery from mild concussion in high school athletes. Journal of Neurosurgery, 98(2), 296-301. Lovel l , M . R., Collins, M . W. , Iverson, G . L . , Johnston, K . M . , & Bradley, J. P. (2004). Grade 1 or \"ding\" concussions in high school athletes. American Journal of Sports Medicine, 32(1), 47-54. MacLeod, D . , & Prior, M . (1996). Attention deficits in adolescents with A D H D and other clinical groups. Child Neuropsychology, 2, 1-10. Maroon, J. C , Lovel l , M . R., Norwig, J., Podell, K . , Powell , J. W. , & Hartl, R. (2000). Cerebral concussion in athletes: evaluation and neuropsychological testing. Neurosurgery, 47(3), 659-669; discussion 669-672. McGee, R. A . , Clark, S. E . , & Symons, D . K . (2000). Does the Conners' Continuous Performance Test aid in A D H D diagnosis? Journal of Abnormal Child Psychology, 28(5), 415-424. Mercugliano, M . (1999). What is attention-deficit/hyperactivity disorder? Pediatric Clinics of North America, 46(5), 831-843. 76 Mitrushina, M . N . , Boone, K . B . , Razani, J., & D'El ia , L . F. (2005). Handbook of normative data for neuropsychological assessment, Second edition. N e w York, N Y : Oxford University Press. Morris, J. C , Heyman, A . , Mohs, R. C , Hughes, J. P., van Belle, G . , Fillenbaum, G . , et al. (1989)^ The Consortium to Establish a Registry for Alzheimer's Disease ( C E R A D ) : Part I. Cl inical and neuropsychological assessment of Alzheimer's disease. Neurology, 39(9), 1159-1165. Mota, V . , & Schachar, R. (2001). Validity of D S M - I V A D H D . Child and Adolescent Psychiatry, 40, 503-504. Murphy, K . R., Barkley, R. A . , & Bush, T. (2001). Executive functioning and olfactory identification in young adults with attention deficit-hyperactivity disorder. Neuropsychology, 15(2), 211-220. Murphy, P. (2002a). Cognitive functioning in adults with Attention-Deficit/Hyperactivity Disorder. Journal of Attention Disorders, 5(4), 203-209. Murphy, P. (2002b). Inhibitory control in adults with Attention-Deficit/Hyperactivity Disorder. Journal of Attention Disorders, 6(1), 1-4. Murphy, P., & Schachar, R. (2000). Use of self-ratings in the assessment of symptoms of attention deficit hyperactivity disorder in adults. American Journal of Psychiatry, 157(1), 1156-1159. Nigg , J. T., Butler, K . M . , Huang-Pollock, C . L . , & Henderson, J. M . (2002). Inhibitory processes in adults with persistent childhood onset A D H D . Journal of Consulting and Clinical[Psychology, 70(1), 153-157. 77 Nigg, J. T., Hinshaw, S. P., Carte, E. T., & Treuting, J. J. (1998). Neuropsychological correlates of childhood attention-deficit/hyperactivity disorder: explainable by comorbid disruptive behavior or reading problems? Journal of Abnormal Psychology, 107(3), 468-480. Peterson, L. R., & Peterson, M. J. (1959). Short-term retention of individual verbal items. Journal of Experimental Psychology, 58, 193-198. Pineda, D., Ardila, A., Rosselli, M., Cadavid, C , Mancheno, S., & Mejia, S. (1998). Executive dysfunctions in children with attention deficit hyperactivity disorder. International Journal ofNeurosciences, 9<5(3-4), 177-196. Reitan, R. M., & Wolfson, D. (1993). The Halstead-Reitan neuropsychological test battery. Tucson AZ: Neuropsychology Press. Rey, A. (1964). L'examen clinique en psychologic Paris: Presses Universitaires de France. Romine, C. B., Lee, D., Wolfe, M. E., Homack, S., George, C , & Riccio, C. A. (2004). Wisconsin Card Sorting Test with children: a meta-analytic study of sensitivity and specificity. Archives of Clinical Neuropsychology, 19(8), 1027-1041. Royall, D. R., Lauterbach, E. C , Cummings, J. L., Reeve, A., Rummans, T. A., Kaufer, D. I., et al. (2002). Executive control function: a review of its promise and challenges for clinical research. A report from the Committee on Research of the American Neuropsychiatric Association. Journal of Neuropsychiatry and Clinical Neurosciences, 14(4), 377-405. Salthouse, T. A. (2003). Memory aging from 18 to 80. Alzheimer Disease and Associated Disorders, 17(3), 162-167. 78 Salthouse, T. A., & Fristoe, N. M. (1995). Process analysis of adult age effects on a computer-administered Trail Making Test. Neuropsychology, 9(4), 518-528. Schachar, R., Mota, V. L., Logan, G. D., Tannock, R., & Klim, P. (2000). Confirmation of an inhibitory control deficit in attention-deficit/hyperactivity disorder. Journal of Abnormal Child Psychology, 28(3), 227-235. Schmitz, M., Cadore, L., Paczko, M., Kipper, L., Chaves, M., Rohde, L. A., et al. (2002). Neuropsychological performance in DSM-IV ADHD subtypes: an exploratory study with untreated adolescents. Canadian Journal of Psychiatry, 47(9), 863-869. Seidel, W. T., & Joschko, M. (1990). Evidence of difficulties in sustained attention in children with ADDH. Journal of Abnormal Child Psychology, 18(2), 217-229. Seidman, L. J., Biederman, J,, Faraone, S. V., Weber, W., & Ouellette, C. (1997). Toward defining a neuropsychology of attention deficit-hyperactivity disorder: performance of children and adolescents from a large clinically referred sample. Journal of Consulting and Clinical Psychology, 65(1), 150-160. Seidman, L. J., Biederman, J., Weber, W., Hatch, M., & Faraone, S. V. (1998). Neuropsychological function in adults with attention-deficit hyperactivity disorder. Biological Psychatry, 44(4), 260-268. Smith, A. (1972). Symbol Digit Modalities Test (SDMT). Manual (revised). Los Angeles: Western Psychological Services. Snow, J. H. (1998). Developmental patterns and use of the Wisconsin Card Sorting Test for children and adolescents with learning disabilites. Child Neuropsychology, 4, 89-97. Spreen, O., & Strauss, E. (1998). A compendium of neuropsychological tests: Administration, norms, and commentary (2nd Edition). New York, NY: Oxford University Press. 79 Stearns, C . L . , Dunham, M , Mcintosh, D . , & Dean, R. S. (2004). Attention deficit/hyperactivity disorder and working memory in clinically referred adults. International Journal ofNeurosciences, 114(2), 273-287'. Tabachnick, B . G . , & Fidel l , L . S. (2001). Using multivariate statistics 4th edition. Boston, M A : A l l y n and Bacon. Temkin, N . R., Heaton, R. K . , Grant, I., & Dikmen, S. S. (1999). Detecting significant change in neuropsychological test performance: a comparison of four models. Journal of the International Neuropsychological Society, 5(4), 357-369. Trommer, B . L . , Hoeppner, J. A . , Lorber, R., & Armstrong, K . J: (1988). The go-no-go paradigm in attention deficit disorder. Annals of Neurology, 24(5), 610-614. van der Meere, J. , & Sergeant, J . (1988). Controlled processing and vigilance in hyperactivity: time wi l l tell. Journal of Abnormal Child Psychology, 16(6), 641-655. Walker, A . J. , Shores, E . A . , Trollor, J. N . , Lee, T., & Sachdev, P. S. (2000). Neuropsychological functioning of adults with attention deficit hyperactivity disorder. Journal of Clinical and Experimental Neuropsychology, 22(1), 115-124. Wechsler, D . (1997). Wechsler Adult Intelligence Scale 3rd Edition. San Antonio, T X : Psychological Corporation. Wilens, T. E . , Biederman, J., & Spencer, T. J. (2002). Attention deficit/hyperactivity disorder across the lifespan. Annual Review of Medicine, 53, 113 -131. Woods, S. P., Lovejoy, D . W. , & B a l l , J. D . (2002). Neuropsychological characteristics of adults with A D H D : a comprehensive review of initial studies. Clinical Neuropsychology, 16(1), 12-34. 80 Woods, S. P., Lovejoy, D. W., Stutts, M. L., Ball, J. D., & Fals-Stewart, W. (2002). Comparative efficiency of a discrepancy analysis for the classification of Attention-Deficit/Hyperactivity Disorder in adults. Archives of Clinical Neuropsychology, 17(4), 351-369. 81 "@en . "Thesis/Dissertation"@en . "2005-11"@en . "10.14288/1.0054528"@en . "eng"@en . "School Psychology"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en . "Graduate"@en . "Computerized screening in adolescents and young adults with ADHD"@en . "Text"@en . "http://hdl.handle.net/2429/16827"@en .