UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effect of mood induction on fear reduction Samson, Deborah Christine Veronica 1987

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

Item Metadata

Download

Media
831-UBC_1987_A8 S25.pdf [ 5.07MB ]
Metadata
JSON: 831-1.0097072.json
JSON-LD: 831-1.0097072-ld.json
RDF/XML (Pretty): 831-1.0097072-rdf.xml
RDF/JSON: 831-1.0097072-rdf.json
Turtle: 831-1.0097072-turtle.txt
N-Triples: 831-1.0097072-rdf-ntriples.txt
Original Record: 831-1.0097072-source.json
Full Text
831-1.0097072-fulltext.txt
Citation
831-1.0097072.ris

Full Text

THE EFFECT OF MOOD INDUCTION ON FEAR REDUCTION by DEBBIE SAM50M B.A., The University of British Columbia, 1983 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES Department of Psychology We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA June, 1987 © Debbie Samsom . 1987 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Psychology The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date June, 1987 Abstract The present research examined the relationship between mood and fear. A musical mood induction technique was utilized to induced either a happy mood or a sad mood in eighty-four female university students fearful of spiders or snakes. Following mood induction, subjects underwent in vivo systematic desensitization to reduce their fear. Fear levels were reassessed four weeks later and those subjects who showed a return of fear underwent a second session of exposure therapy. Measures of subjective fear and self-efficacy were taken before and after mood induction and again after fear reduction. The length of time taken to reduce fear was also recorded. The findings showed that an induced sad mood led to greater subjective fear and lower self-efficacy compared with an induced happy mood. In addition, an induced sad mood during fear reduction was associated with greater return of fear four weeks later. No difference was found in the length of time taken to reduce fear for happy and sad subjects. This study also addressed the issue of mood state dependency of fear reduction. It was hypothesized that subjects who underwent their second session of fear reduction while in a mood state congruent with that of their first session would show more rapid habituation than those in an incongruent mood state. The results did not support this hypothesis. If the findings of this investigation prove generalizable to clinical depression, they suggest that the most effective approach in treating individuals who are both clinically depressed and anxious may be to treat the depression before beginning exposure to fearful situations or stimuli. ii Table of Contents Abstract ii Table of Contents iii List of Tables ; v List of Figures •. vi Acknowledgement vii I. INTRODUCTION 1 A. Historical Overview 1 B. Depression and Fear Reduction 4 C. State-Dependency Learning 12 II. EXPERIMENTAL RATIONALE 16 A. Subjects and Recruitment 16 B. Treatment Techniques 19 C. Therapists 19 D. Mood Induction Procedures 20 E. Dependent Measures 27 F. Research Hypotheses 27 III. METHOD ; 29 A. Subjects and Design 29 B. Experimenters and Therapists 30 C. Procedure 32 i. Screening 33 ii. Heart-Rate Measurement 34 iii. Mood and Self-Efficacy Measurement 35 iv. Behavior Avoidance Test 36 v. Handwriting Speed Task 37 i i i v i . Mood Induction 3 8 v i i . Fear Reduct ion 3 9 v i i i . Se s s i on Two 4 0 IV. RESULTS 42 A. M i s s i n g Data 4 2 B. P r e l i m i n a r y Ana l y se s 4 2 i . Mood Man ipu la t ion 4 4 C. Ma jo r Ana l y se s 4 7 i . W i t h i n Se s s i on E f f e c t s 47 i i . Be tween Se s s i on E f f e c t s 5 3 i i i . S t a t e Dependency 5 4 D. A d d i t i o n a l Ana l y se s 5 7 i . S u c ce s s f u l v s P a r t i a l l y S u c ce s s f u l S ub j e c t s 5 7 i i . The rap i s t Compar i sons 5 8 V. DISCUSSION 6 7 REFERENCES 8 2 APPENDICES A Therap i s t Rat ing S ca l e f o r Sub jec t ' s Mood 8 9 B The rap i s t Eva lua t ion Quest ionna i re 91 C Fear Survey Schedule 9 4 D Consent Form 9 7 E Mood Quest ionna i re 100 F S e l f - E f f i c a c y Questionnaire.. 102 G Man ipu la t ion Check 104 H Tab le s of Means and Standard Dev ia t i ons 106 I Un i va r i a te Ana l y se s Summary Tab les 113 J Ana l y se s of Va r i ance Summary Tab le s 118 iv List of Tables Table 1 — Mean Scores on Major Dependent Variables Before Mood Induction 43 2 ~ Intercorrelations Between Items on Mood Questionnaire Before Mood Induction 48 3 — Intercorrelations Between Items on Mood Questionnaire After Mood Induction 49 4 — Mean Scores on Major Dependent Measures After Mood Induction 50 5 ~ Mean Scores on Relevant Variables Based on Return of Fear 55 6 ~ Chi Square Showing the Number of Subjects in Happy and Sad Groups With Return of Fear 56 7 ~ Mean Scores on Relevant Variables for Partially Successful and Successful Subjects 59 8 — Chi Square Showing the Number of Subjects with Return of Fear 61 V List of Figures Figure 1 — Mean Reported Sadness Pre and Post Mood Induction 45 2--Mean Reported Happiness Pre and Post Mood Induction 46 3 — Mean Suds Pre and Post Mood Induction 51 4 — Mean Self-Efficacy Pre and Post Mood Induction 52 5 — Mean Return of Fear 60 6 — Mean Duration: Therapist by Group 63 7 ~ Mean Happiness Ratings: Therapist by Group 64 8 — Mean Sadness Ratings: Therapist by Group 65 vi Acknowledgement I wish to express my appreciation to my research advisor, Dr. S. Jack Rachman, and to my research committee members, Dr. Keith Dobson and Dr. James Steiger, for their guidance during the course of this study. I would also like to thank Susan Cross-Calvert, Hilary Stevens, Bridget Minishka and Gene Flessati for the time and effort they put into this project. A special thank you is extended to John for his encouragement and support. I share this accomplishment with him. vii 1 Introduction Historical Overview Although the effectiveness of exposure techniques for fear reduction has been demonstrated (Marshall, Gauthier & Gordon, 1979), the theoretical underpinnings of this procedure are st i l l uncertain. Several explanations have been put forward, including reciprocal inhibition (Wolpe, 1958), counterconditioning (Davison, 1968), self-efficacy (Bandura, 1977), and habituation (Rachman, 1978). The reciprocal inhibition interpretation as described by Wolpe (1958) holds that a graded hierarchy and an incompatible response (ie. relaxation) are essential for the successful desensitization of fear. The counterconditioning theory also relies upon the importance of an incompatible response. However, this is st i l l a debatable point. Some studies (eg. Davison, 1968; Kass & Gilner, 1974) suggest that some structured relaxation procedure is a necessary ingredient in desensitization. Other findings (Miller & Nawas, 1970; Nawas, Welsch & Fishman, 1970) seem to indicate the opposite. Schubot (1966) found that a structured relaxation procedure was necessary only for subjects who were initially extremely phobic. For moderately fearful subjects, the same degree of improvement occurred whether or not relaxation was paired with anxiety. It could be argued that people who experience intense fear reactions are more likely, although not necessarily, to seek treatment for their difficulty, and would therefore be more like the extreme phobics than the moderate phobics in the Shubot study. This would suggest the importance of utilizing relaxation in a clinical setting. The necessity of presenting hierarchy items in a graded, gradual fashion has also been debated. Using the standard desensitization format, items are ranked according to the amounts of fear each elicits, and when presenting the 2 scenes, the least fear-arousing ones are presented first. While this procedure is clearly effective, Krapfl (1967) found that presenting the hierarchy in descending order was about as effective in reducing snake-phobic behavior as the standard ascending-order procedure. Richardson and Suinn (1973) reported improvement in test anxiety when their subjects were exposed to only the three highest hierarchy scenes. However, these procedures may cause the phobic individual considerably more distress than encountered when the standard method is used. Bandura's self-efficacy theory holds that therapeutic effects of exposure techniques (as well as other treatment strategies) result from an increased conviction in one's ability to perform particular tasks. Some research arising from this theory has demonstrated a strong relationship between self-perceived efficacy and subjective fear (Bandura, 1982; Craske & Craig, 1984) when these variables are assessed at the same time. However, in a recent study by Craske and Rachman (1985), self-efficacy measures collected prior to therapy were not shown to be useful in predicting the return of fear several months post-treatment. Rachman (1978) suggests that exposure-based fear reduction techniques may best be construed as a form of habituation. According to this theory, the presentation of phobic stimuli results in exposure to crucial stimulus elements which provoke anxiety or panic. With repeated or prolonged exposures to the stimuli, fear gradually reduces. Each of the above theories has been the subject of extensive theoretical and experimental analyses. Research arising from these theories has contributed valuable, information to our understanding of fear reduction techniques and suggests that perhaps multiple procedures are at work. Turning to more practical treatment issues, researchers have analyzed 3 various components of exposure therapy to determine their relative contribution to the overall treatment package. Treatment components such as duration of exposure (eg. Sue, 1975; Watts, 1973), prevention of avoidance (deSilva & Rachman, 1984), and order of hierachy presentation (eg. Krapfl, 1967) have been explored. These studies have provided therapists with practical information about the dimensions of exposure therapies that can be used to enhance treatment effects. In addition to treatment components, research in exposure therapy has focused upon subject variables. Here, the main question asked is whether a treatment is more or less effective with certain kinds of participants. Subject variables which have been shown to affect fear-reduction techniques include heart rate (Craske & Rachman, 1985; Grey et al, 1981), self-efficacy (Bandura, 1979), and initial reported fear level (eg. Philips, 1985; Rachman & Levitt, 1985; Grey et al, 1981). The purpose of the present investigation was to determine whether an individual's mood is an important variable in exposure techniques for reducing anxiety responses. The effects of a depressed mood were of particular interest. It was hypothesized that there is a relationship between mood and fear, such that a depressed mood impedes fear reduction and a happy mood enhances fear reduction (this shall be termed the "mood effect"). However, it is possible that this relationship is not so simple. It may be that fear reduction is mood-state dependent. This implies that if subjects undergo successful fear reduction in a depressed mood, then therapeutic gains should be most evident at a later date if subjects have their fear reduced a second time in a congruent mood state (this shall be termed the "state-dependency effect"). To date, there has been no research on the state-dependency of fear 4 reduction techniques. Studies involving state-dependency have focused largely on the effects of drug-induced states on memory retention. Drug induced state-dependent retention has been well established (Eich, 1980). State-dependent retention has also been reported with changes in affective state (Bower et al., 1978; Macht, Spear & Levis, 1977; Weingartner, Miller & Murphy, 1977), although this phenomenon is less clearly established. The present study is concerned with the affective state-dependency of a behavioral task (ie. systematic desensitization). At present, there are no studies in the psychological literature investigating this phenomenon. During the second session of fear reduction, mood effects and state-dependency effects may be in conflict. Consider a situation in which individuals receive their first session of exposure therapy while in a sad mood state. They then undergo a second session of fear reduction in either a happy mood or a sad mood. State-dependency would lead to the prediction that those in a sad mood during session two would show more rapid habituation than those in a happy mood. However, mood effects would suggest the opposite - ie. those in a happy mood would show more rapid habituation, regardless of their emotional state during their first session of fear reduction. At the outset of this study, it was not known which variable -mood or state-dependency - would exert greater influence over fear-reduction. Depression and Fear Reduction Exposure therapy for fear reduction can fail for several reasons. For example, treatment might be too intense or too brief, or the phobic individual may be too agitated or too depressed. Regardless of the reasons for which treatment is a failure, it is often of interest to determine when and by what process failure has occurred. An interested researcher might choose to 5 examine whether fear is declining during a given treatment session. This decline of fear is often interpreted as within-session habituation and is assessed by comparing fear levels at the beginning and end of a session. An investigator might also consider "return of fear" as a factor in treatment failure. This phenomenon is defined as the "reappearance of fear that was earlier present but had undergone a decline" (Rachman, 1979). During the past eight years the return of fear has been studied in circumscribed fears (eg. Grey, Sartory & Rachman, 1979, Grey Rachman & Sartory, 1981, Philips, 1985, Rachman & Levitt, 1985) and in obsessive-compulsive disorders (Foa, 1979; Likierman & Rachman, 1980). It is a robust phenomenon and is typically observed after an interval between the completion of a treatment session and the commencement of a new session, or between completion of a treatment programme and re-test (Craske & Rachman, 1985). In the present investigation both within-session habituation and return of fear were of interest. There are a number of reasons for hypothesizing that a depressed mood impedes fear reduction. To begin, there are examples in the psychological literature of patients suffering from both depression and circumscribed phobias (eg. Philips, 1985; Foa, 1983). When exposure techniques are implemented as a treatment for the fear, depression has reportedly impeded the patient's progress. Such research has focused on the differential effects of depression on within-session habituation and between-session return of fear. For example, Philips (1985) investigated exposure-based treatment of the fear of vomiting. She discovered that she could divide her vomit-phobic subjects into two groups: Fast Respondents (those who completed therapy in eight sessions) and Slow Respondents (those who completed therapy in thirteen sessions). From Philips' study, it appears that the Slow Respondents 6 were slower not because of any delays in within-session habituation, but because of a repeated return of fear between sessions. In addition, the Slow Respondents had slightly higher B.D.I, scores than the Fast Respondents. The investigator therefore hypothesized that the return of fear might have been a consequence of low mood. These results are in need of replication due to the unusual nature of the fear of vomiting and the small sample (N=7) in Philips' investigation. Foa (1979) showed a comparable return of fear in depressed obsessional patients. In this group of subjects, habituation was not seen to occur either within or between sessions. These failures were in contrast to the successes observed in patients who responded positively to comparable forms of treatment, and who showed both forms of habituation. In a later report on a larger group of patients, Foa, Grayson and Steketee (1980) confirmed that highly depressed patients showed less improvement than mildly depressed patients (referring to improvements in obsessions and compulsions). These differences were apparent at the end of treatment and at a follow-up assessment. In regard to specific effects on habituation, these researchers found that there was a nonsignificant trend for the highly depressed patients to show less habituation within the session and a similar nonsignificant trend was observed in respect to habituation between sessions. It is difficult to draw any firm conclusions from the somewhat disparate results of Philips' and Foa's research. In Foa's first study (1979) she found her depressed obsessive compulsives to have poor within and between session habituation. In her second study, (Foa et al, 1980), using a larger sample, she found a non-significant trend in the same direction. Philips* results lend support to the hypothesis that depression leads to poorer between-session habituation. She, however, did not find any within-session effects with her 7 subject population. The present investigation was designed to assess both within-session and between-session levels of fear in order to clarify the effects of mood on fear reduction. Another reason for hypothesizing that a depressed mood impedes fear reduction stems from the literature on the effects of anti-depressive medication on obsessive-compulsive individuals. Using a within-subject design, Capstick (1975) investigated the effects of clomipramine on four patients suffering from obsessive-compulsive symptoms. While these patients may not be representative of obsessional patients as a whole, results of this study suggest that clomipramine is effective in reducing compulsive behaviors. A later study by Rachman et al (1979) utilized a between-groups design to investigate the effects of exposure plus clomipramine for the treatment of obsessive-compulsives. The results of this research indicated that clomipramine was followed by improvements in mood and compulsions. These results are consistent with those of Capstick (1975) and suggest that significant depression helps to maintain compulsive behavior. However, it is difficult to distinguish between the primary and secondary effects of clomipramine. Rachman et al (1979) hypothesize that improvements in mood were primary and improvements in compulsions secondary based on the fact that nondepressed obsessive patients did not improve significantly on measures of compulsive behavior upon receiving clomipramine (compared with a placebo control group). Further clarification of the connection between depression and anxiety would be achieved if it is shown that improvements in anxiety occur after a significant decline in depression. The present study addressed this issue by manipulating mood in the laboratory and then assessing the effect on fear. Certain psychological theories would suggest that depression impedes 8 fear-reduction. Research by Bower (1983) indicates that mood states can affect cognitive processing such that when people are in a sad mood this leads to thoughts of past failings, whereas a positive mood is accompanied by thoughts of personal accomplishments (as well as other pleasant thoughts). Further, these cognitions appear to affect a person's belief in self-efficacy. People judge their capabilities to be higher when under a hypnotically-induced positive mood than when in a neutral state, and they regard themselves as least-efficacious when in a depressed mood (Kavanagh & Bower, 1985). Further, when dealing with cognitive tasks, persons act in accordance with their mood-altered percepts of self-efficacy and thus choose less challenging tasks when in an inefficacious frame of mind (Kavanagh & Bower, 1985). It was hypothesized that if these findings generalize to fear reduction, a depressed mood would result in less willingness to tolerate exposure to the feared stimulus which would, in turn, impede fear reduction. The concept of emotional processing (Rachman, 1979) provides a useful framework for discussing the relationship between depression and fear-reduction. Rachman describes emotional processing as "a process whereby emotional disturbances are absorbed, and decline to the extent that other experiences and behavior can proceed without disruption". Within the emotional processing paradigm, successful fear-reduction methods, with their emphasis on repeated exposures to the fearful stimuli, are viewed as a means of absorbing persistent, disruptive emotional reactions. Unsuccessful fear reduction - either in the form of persistence of fear or in the form of return of fear - is seen as incomplete emotional processing. Rachman describes four groups of factors that may give rise to difficulties in emotional processing. Among these are the State Factors, one of which is dysphoria. The following relationship is hypothesized: depression leads to 9 incomplete emotional processing which is reflected in unsuccessful fear reduction. It is possible that a depressed mood can adversely affect fear reduction because of its relationship to variables already known to influence this process. For example, an elevated heart-rate response at the start of a fear reduction programme is a strong predictor of the return of fear (Craske & Rachman, 1985; Grey et al, 1981). A study by Schwartz et al (1981) showed that an induced sad mood resulted in an increase in heart rate that was significantly higher than a control group, but not significantly different from a fear-induced group. In addition, Ekman, Levenson & Friesen (1983) reported high heart rate associated with a sad emotional state. One might therefore hypothesize that a depressed mood is associated with high heart rate and thus is predictive of return of fear. Initial level of fear has also been investigated as an important variable in fear-reduction procedures (Philips, 1985; Rachman & Levitt, 1985; Grey et al 1979, Grey et al, 1981; Sartory et al, 1982). Evidence from these studies suggests that a high level of initial fear is predictive of return of fear. For example, in a study by Rachman and Levitt (1985), initial fear and return of fear were positively correlated (r=.57). When mildly claustrophobic subjects were divided into two groups according to whether or not they showed return of fear, initial fear scores of the two groups were significantly different. For the return of fear group, initial fear scores averaged 72.75 (out of 100) in comparison to 5417 for those subjects who showed no return of fear. Initial level of fear has also been shown to be related to depression. Philips (1985) found that those subjects who were slow to respond to treatment had a high average initial level of fear, in addition to elevated depression. When first exposed to the feared stimulus, this group reported an 10 average distress level of 60 (out of 100). This is in contrast to the fast respondent (nondepressed) group who had an average level of fear of 30 at treatment onset. Foa (1982) noted that in a sample of obsessive-compulsives the magnitude of reported fear when subjects were first exposed to the feared stimulus correlated positively with depression and negatively with within-session habituation of anxiety. This result was replicated in her 1983 study where initial reported fear and depression had a significant correlation (r- 46). In addition, both of these variables were negatively correlated with treatment outcome (initial fear level: r—.47, depression: r=-.39). Although the direction of causation is difficult to determine, the authors propose a predictive model of treatment outcome for obsessive-compulsive individuals which suggests that depression affects fear reduction via its positive relationship with reactivity (which is indicated by initial level of fear). The latter, in turn, impedes habituation and thus interferes with responsiveness to behavioral treatment. None of the previously mentioned studies included a physiological measure of fear. Craske and Rachman (1985) recorded heart rate and initial fear and found that the two variables were closely related. When initial fear differences were covaried out, it was found that heart rate effects did not dissipate. On the basis of this analysis, the researchers concluded that return of fear does not seem to depend on high levels of initial fear. Sartory et al (1977) and Grey et al (1979) arrived at the same conclusion based on their data which also showed a linear relationship between heart rate and subjective distress. It was expected that this relationship would be observed in the present investigation. A strong negative relationship between self-eff icacy and subjective fear 11 has also been established. Bandura (1982) had phobic subjects report their self-efficacy in performing different tasks varying in threat value. Afterwards, they reported the intensity of fear arousal that they experienced in anticipation of performing each task and again while they were performing the activity. The results indicated that subjects experienced high anticipatory and performance distress on tasks in which they had low perceived self-efficacy. As the strength of their self-efficacy increased, their fear arousal declined. Bandura (1982) again found this relationship when he investigated the amount of subjective distress phobics at different levels of perceived self-efficacy experience while performing the same task. The less efficacious subjects judged themselves to be, the more anticipatory and performance fear they reported experiencing. In a recent study by Craske and Rachman (1985), self-efficacy measures collected prior to therapy were not shown to be useful in predicting the return of fear several months post-treatment. However, it is noteworthy that 71% of the subjects who reported a reduction in self-efficacy between post-treatment and followup showed an increase in subjective fear at followup. These subjects also showed higher heart rates, more anxious thoughts, less skill and fewer performances. Therefore, it would be presumptuous to conclude that self-efficacy was the crucial mediator in the return of fear since other variables were clearly associated with this phenomenon. A negative relationship between self-efficacy and depression has also been observed. In naturally occurring clinical depression, low self-efficacy is a widely reported symptom (Beck, 1976). With respect to mood induction techniques, Teasdale and Spencer (1982) found that subjects given a sad induction gave lower estimates of the probability of future successes on a 12 laboratory task (one measure of self-efficacy). On the basis of research on subject variables and fear-reduction, it was hypothesized that a depressed mood is associated with decreased self-efficacy, increased initial fear and increased heart rate. These effects in turn i mpede f ear reduct i on procedures. State-Dependent Learning As mentioned previously, research on state-dependency is confined to its effects on memory recall (Eich, 1980). State-dependence implies that what one remembers during a given mood is determined in part by what one learned (or focused upon) when previously in that mood. There is considerable evidence demonstrating drug-induced state dependency of recall (Eich, 1980). However, results from studies involving happy or sad states per se are less clear. Weingartner, Miller & Murphy (1977) utilized naturally occurring mood shifts to investigate affective state-dependent learning. In this research subjects were manic depressive in-patients, monitored repeatedly at four-day intervals. They were asked to generate twenty word associations on one day and regenerate them four days later. The results of this study suggest that recall of associations was more complete when subjects were in the same state as when the words were initially generated. This was true for periods of mania, depression and stable mood. The major drawback of this study is that the affective properties of the generated words were not controlled and therefore mood congruency could have accounted for some or all of the effects. Mood congruence assumes that some material by virtue of its affectively valenced content, is more likely to be stored and/or recalled when one is in a particular mood (Blaney, 1986). It is possible that the word associations generated by the subjects in the Weingartner et al. (1977) study 13 varied with mood state, although the investigators do not provide any information in this regard. If this is the case, then subject recall may have been more a function of mood congruency between words and subject state than a function of state dependency. There have also been attempts to show mood state-dependent effects by means of mood manipulations. Leight and Ellis (1981) compared memory recall of depressed and neutral individuals (using the Velten task to manipulate mood). They presented the stimuli to be remembered in one of two ways: varied or constant. A varied presentation has the effect of making words easier to recall and recognize than constant presentation. To summarize, this study entailed a 4 x 2 (mood x presentation) design resulting in eight instances in which state-dependency effects could be assessed for both recognition and recall. For recall tasks, Leight and Ellis (1981) found that the important factor was not the relationship between mood during learning and mood during recall, but rather, simply mood during learning. That is, subjects who were depressed during learning recalled fewer items when tested the second time than subjects who were initially in the neutral mood condition, regardless of mood state during second session recall. An asymmetrical state-dependency effect was observed on the recognition test. Specifically, recognition was state-dependent, but only when original learning occurred under conditions of constant input and a depressed mood. Leight and Ellis explain their findings as a special instance of the more general principle that state-dependency effects become more pronounced as information is more poorly learned (Weingartner, Adefris, Eich and Murphy, 1976). This finding is further supported by research by Bower et al (1978, study 3) in which state-dependency effects were observed under the 14 interference condition of multiple lists (making learning difficult). Specifically, subjects who had learned two lists - one while happy, the other while sad - had more success (when one of these moods was reinstigated) recalling the list learned in the concordant mood than in the disconcordant mood. Schare et al (1984, study 3) also demonstrated state dependence effects on free recall (but not on cued recall) using a similar interference paradigm. However, Bower and Mayer (1985) and Wetzler (1985) recently reported nonreplication of these findings. Finally, Gage and Safer (1985) have reported mood state-dependence for face recognition in an interference context, though only when stimuli were presented to the right hemisphere. In the Bower et al study, no state-dependent recall effects were found when recall of single word lists was tested, either immediately or one day later. Three other studies also fail to show mood-state dependence in a noninterference recall paradigm (Isen, Shalker, Clark & Karp, 1978, study 2; Bower, Gilligan and Monteiro, 1981, study 3; Schare, Lisman and Spear, 1984). The above research on mood state-dependency lends support to the theory that state-dependent retention may depend upon accessibility of the original material which is in part a function of the extensiveness of encoding during learning (Eich, 1980). That is, the stronger the encoding, the less necessity for reliance upon internal cues (such as mood state) for accessing memory. This theory has implications for the present investigation. Unlike the state-dependency studies previously mentioned (in which the sad groups typically had poorer initial learning), all subjects (regardless of their mood) were trained to the same level of fear reduction. By the end of the first session, fear-reduction was well-learned across groups and thus it was questionable whether state-dependency effects would be evident during the second session. 15 In summary, the findings regarding affective state-dependent recall are mixed. The one effect which has been replicated (Bower et al, 1978, study 3; Schare et al, 1984, study 3) involves free recall in an interference context, suggesting that mood state cues prevail only when other retrieval cues are vague. Mood state-dependency learning does not appear to be a robust phenomenon and therefore it was questionable whether it would be observed in the present investigation. 16 Experimental Rationale This investigation was designed to evaluate the effects of a laboratory-based mood-induction procedure on fear-reduction in a sample of female university undergraduates fearful of spiders or snakes. When considering laboratory-based approaches to the study of human fears, a number of issues arise. Such research is often termed "analogue" despite the writings of Kazdin (1978) and others who have pointed out that all experiments are analogues in one sense or another. The so-called analogue approach to the study of human fears is often dismissed on the basis of a lack of generality to clinical populations and settings. However, this implies that psychological research is of value only if it contributes to our knowledge of clinical practice. Clearly, this is a narrow view of research. Experiments can contribute to knowledge about behavior change in general (without reference to clinical modification). With respect to the present investigation, the way in which mood affects fear reduction was of interest regardless of its clinical relevance. The question of generality was not the issue at hand. If the results of this investigation have some clinical application this is, of course, a welcome bonus, but it was not essential for the verification of the specific hypotheses under investigation. Finally, it should be pointed out that the general izabil ity of any experiment - regardless of whether it is so-called analogue research - should not be assumed. This remains an empirical question to be addressed in future research. Experimenters should therefore be cautious in the conclusions they draw from their work. With these considerations in mind, let us proceed with a description and rationale of the present investigation. Subjects And Recruitment The subjects in this experiment consisted of eighty-four female university 17 undergraduate students with small animal fears (spiders or snakes). Females were chosen as participants for two reasons. First, they are more likely to report and display small animal fears (Rachman, 1978). Second, they appear to be more susceptible to mood induction procedures (DiGeso, personal communication). These factors suggested that fewer subjects would be discarded during data collection if participation was restricted to female students. Because this research involved fear reduction, it was essential that subjects exhibit some level of fear before the experiment could begin. To verify that subjects were experiencing fear when exposed to the target stimulus, a Behavior Avoidance Test (BAT) was conducted. This procedure, based upon research pioneered by Lang and Lazovick (1963), consisted of asking subjects to approach and touch the feared target object. Those subjects who made physical contact with the object were excluded from the study, and, for those who remained, the degree of physical approach was taken as an index of anxiety. In order to ensure that the BAT was a useful screening device, a number of procedures as outlined by Bernstein and Paul (1971) were followed. Subjects were not asked to do something with the target stimulus that they would probably not know how to do. For example, asking subjects to handle organisms with which they may be unfamiliar (such as spiders and snakes) may lead to avoidance simply because subjects do not know how to go about handling them. Therefore, the BAT in this investigation consisted of asking subjects simply to touch the stimulus, rather than handle it. Another potential influence upon subjects' behavior during BATs is the content and presentation of instructions. It is likely that subjects told to approach and touch the target may be more likely to do so than those given 18 similar instructions, but asked to do only what is comfortable for them (Bernstein & Paul, 1971). While specific factors such as instructions and target familiarity play an important part in subject performance during a BAT, perhaps the most crucial influences are the context variables. These include mainly the rationale for the subject's presence at the BAT and the demand characteristics operating within it. Most subjects participating in a BAT know that they are there because of their reported fear, that their fear is being measured by the test, and possibly that failure to demonstrate some degree of fear will eliminate them from the project. In order to control for such demand characteristics, an initial screening device (the Fear Survey Schedule) was utilized. This questionnaire consists of a list of eleven items to which the subject is to indicate the level of fearfulness that each elicits (ranging from "not at all fearful" to "terrified".). By using a multi-item questionnaire, researchers can prevent potential subjects from knowing which fears they are interested in. Clearly, this procedure has the advantage of fewer demand characteristics than that of offering a blanket invitation to all individuals who consider themselves fearful of the available stimulus. When subjects were subsequently contacted over the telephone and asked to participate in the experiment, they were not told that they had been selected specifically because they reported being fearful of spiders or snakes. They were told that they would be asked to approach a small animal of which they may or may not be fearful. The BAT was presented as the first step in the experiment, and thus subjects did not know that continuation was contingent upon their eliciting a requisite level of fear. One final point should be made about the BAT. It is almost always repeated after treatment as a measure of improvement. It seems likely that 19 at this point there would be strong pressure exerted on the subjects to touch the previously avoided object. In the present study, these demand variables were reduced in two ways: 1) The post-treatment BAT took place at the beginning of the second session. This session was presented as a replication of the first session so subjects would be less likely to assume that their fear level should have gone down. 2) To avoid therapists placing any outcome expectations on subjects, they did not conduct the pre- or post-treatment BATs, and they were blind to experimental conditions. At the end of each session, therapists rated the mood of the subject in order to determine if they could guess what mood induction each subject had undergone. Treatment Techniques Typically, clinical behavior therapy for spider and snake phobics is done on an Individual basis and involves in vivo presentation of stimuli. Treatment consists of gradual, graded exposure to the feared stimuli and therapist modelling. The patient is in control of the speed of exposure. These elements were incorporated in the fear reduction procedures of the present investigation. However, treatment differed from that of a clinical setting in that it took place in the laboratory under more controlled, truncated conditions. A detailed description of the treatment is presented in the "Methods" section of this paper. For the present purposes, it is important to state that the approach has been used in other investigations (eg. Sartory, Rachman & Grey, 1977; Grey, Sartory & Rachman, 1979) and has been noted to have "remarkable and rapid 'therapeutic' power" (Grey, Sartory & Rachman, 1979). Therapists Two graduate students in clinical psychology were the therapists in this study. Both were trained in in vivo systematic desensitization. They differed 20 from most therapists in a clinical setting in terms of their age and years of experience. Although this limits generality, it did not affect the specific hypotheses under investigation. Mood Induction Procedures Before reviewing the literature on mood induction techniques, a number of statements should be made about the general concept of emotions. Emotions are thought of as being comprised of three components - subjective, behavioral and physiological (Lang, 1968). There is little dispute that various emotions (such as happiness, anger, fear) can usually be differentiated on the basis of the first two components. However, research to date suggests that physiological activity associated with emotions cannot be differentiated except under highly specific conditions. For example, Ekman, Levenson and Friesen (1983) investigated six emotions - surprise, disgust, sadness, anger, fear and happiness. They found only two significant physiological differences among these emotions: 1) Heart rate changes associated with anger, fear and sadness were all greater than those for happiness, surprise and disgust. 2) Finger temperature change associated with anger was greater than that for all other emotions. In an earlier investigation by Schwartz, Weinberger and Singer (1981), only one physiological difference in happy, sad, angry and fearful emotional states was found. Anger led to greater increases in diastolic blood pressure than all other emotions. This study included measures of heart rate, systolic blood pressure and diastolic blood pressure. A second issue related to emotion is the question of whether it is usual or even possible to experience "pure" mood states. For example, when an individual reports feeling sad, it is important to assess whether he/she is experiencing other emotions at the same time. Ekman, Levenson and Freeman 21 (1983) attempted to induce specific mood states by asking subjects to relive a past emotional experience. After each attempt, subjects rated the intensity of any felt emotion on a scale from 0 to 8. 45% of the time, subjects reported feeling the requisite emotion at less than half the possible intensity and/or reported feeling other emotions at a similar strength. In the Schwartz et al investigation (1981),-numerous examples of "mixed" emotion were reported. For instance, ratings of sadness during the anger and fear conditions were high. In addition, anger was occasionally reported during the fear condition, happiness during the relaxation condition, and relaxation during the happiness condition. Finally, Sutherland, et al (1982) found that their induced sad mood was accompanied by broad changes that included anxiety, tiredness and apprehension. It is important to bear in mind that naturally occurring depression has been described as a complex blend of emotions, rather than a "pure" emotion (eg. Izard, 1971). It can therefore be argued that inducing depression calls for increases in other emotions as well. The issues of emotional psychophysiological specificity and "pure" emotional states are relevant to the present investigation because an attempt was made to manipulate mood states. It was, of course, necessary to verify that this manipulation was successful. As previously mentioned, this could not be done on the basis of physiology. At most, we can conclude from such data that an individual is in a state of arousal. This left two emotional components - subjective and behavioral. Subjects were asked to what extent they were feeling the target emotion. If they reported experiencing this at a high intensity while feeling other emotions at a very low intensity (or not at all), this was one indication that their dominant mood was in accordance with the target emotion. A behavioral measure was then used to validate the 22 subjective measure. It is often predicted that certain behaviours are congruent with certain mood states. If, after inducing a specific mood, an individual behaves in a predicted manner, this lends validity to the mood induction. Further, if physiological measures indicate that the subject is in a state of arousal, this adds to the evidence that the individual is experiencing an emotional change. Mood induction procedures (MIPs) have been used as laboratory analogues of clinical depression since 1968 when Velten introduced the first of such techniques. His method involves reading aloud sixty negative self-referent statements (eg. "I'm discouraged and unhappy about myself) and remains the most widely used MIP. The statements progress from neutrality to dysphoria. The overall tone is that of indecisiveness, tiredness, unhappiness, inefficiency and pessimism. Research (which will be outlined in detail later) suggests that Velten's mood induction procedure induces a genuine mood state which has many similarities to a naturally occurring depressed mood (Clark, 1983). However, there are two major drawbacks to this technique: 1) A large number of people fail to respond to Velten's MIP (between thirty and fifty percent of all subjects). This makes it cumbersome for research purposes and raises questions about the generalizability of the results obtained using it. 2) When subjects do show a mood change in response to Velten's MIP, the effects are transient. For this reason, experimenters interested in studying the effects of induced mood on a long task or series of tasks must periodically "boost" the induced mood by returning the subjects to the induction procedure. This makes the technique inconvenient at times. In an attempt to overcome these problems, Sutherland et al (1982) devised an alternative mood induction procedure. This procedure involves playing subjects mood-suggestive music and asking them to use the music as a 23 background to their own efforts. It is stressed that the music by itself will not automatically induce the desired mood state and that they should try very hard to get into the mood; using whatever means they find most effective. The first study of musical MIP allowed subjects to choose between several different pieces of music. However, most subsequent studies have used the same piece of music for all subjects allocated to a particular mood induction. The present investigation utilized the same music as recent research conducted by Eich (1986) on the effects of MIPs on memory. This researcher found these pieces to be successful in inducing the desired mood in 90% of research participants. This success rate is common in musical MIP research. Comparisons between studies using Velten's MIP and musical MIP indicate that the latter affects a larger number of people. Sutherland et al (1982) report two mood induction studies, one using Velten's MIP (Polivy & Doyle, 1980) and one using musical MIP (Clark & Teasdale, 1985). Comparable mood measures, subject population and mood change criteria were used in the two studies. One hundred percent of the subjects met the predetermined mood change criterion in the study employing musical MIP compared to 68% in the study employing Velten's MIP. Clark and Teasdale (1985) asked subjects in a post-experimental questionnaire whether they had experienced a genuine change of mood during the musical MIP; 87% of subjects replied "yes", whereas only 50% of subjects replied affirmatively to a similar question in Pol ivy and Doyle's study of Velten's MIP. These results have been replicated in a comparison study by Clark (1983) which is more impressive because the two studies under investigation utilized identical self-report measures of mood and drew on the same student population for their subjects. Other techniques such as listening to a taped depressing story (eg. Williams, 1980) and failure feedback (Gatchel et al, 1975) have also been 24 used to induce a sad mood. However, these procedures have not been well-researched and there is no indication from the studies that employ them that they are superior to musical mood induction procedures. When considering the generality of this experiment, the question of whether mood induction procedures result in an appropriate analogue to clinical depression arises. A number of aspects of clinical depression have been assessed after the induction of a sad mood. These are described in the next section of this paper. Self-Reported Mood Subjects given the sad musical mood induction report higher levels of despondency or sadness than either an elated induction control group or a neutral induction control group (Clark & Teasdale, 1985; Sutherland et al, 1982; Teasdale & Spencer, 1982). Musical Ml Ps clearly have a strong effect on reported sadness. To guard against the possibility that subjects might, in order to please the experimenter, be reporting changes in mood which they do not actually feel, a number of investigators specifically asked subjects to be honest in their ratings of mood (eg. Clark & Teasdale, 1985). As an additional check, Clark and Teasdale (1985) asked subjects in a post-experiment questionnaire to report honestly whether or not they felt their mood had changed in the desired direction. Under these conditions, Q7% of subjects reported that their mood had genuinely changed. Anxiety Along with elevated levels of sadness, subjects given the depressed induction have shown elevated levels of anxiety (Clark & Teasdale, 1985; Sutherland et al, 1982). This is entirely consistent with the nature of naturally occurring depressed mood states. They are often accompanied by 25 elevated levels of anxiety. Of course, a depressed mood, by definition, is most marked. With regard to the present investigation, it was expected that ratings of anxiety would covary with ratings of happiness/sadness, but that they would not be as extreme. Psychomotor Retardation Psychomotor retardation is a frequently reported symptom of depression (Millar, 1975). A wide range of measures of retardation have been used in studies of mood induction procedures. The musical MIP has been shown to affect several of these. Teasdale et al (1980) have shown that the time taken to count from one to ten is a sensitive measure of changes in a naturally occurring depressed mood. Clark and Teasdale (1985) and Teasdale and Spencer (1982) found that induced sadness was also associated with longer count times. Clark and Teasdale (1985) found that subjects given a sad induction were slower on a handwriting speed task. This task consists of asking subjects to write out numbers in descending order beginning with 100. Subjects are given one minute to work at this task. These researchers also found that sad-induced subjects took longer to sort a deck of playing cards into two piles (black cards and red cards). Loss of Incentive Teasdale & Spencer (1982) gave subjects a list of eight potentially pleasurable activities and asked them to rate each activity in terms of how much they would like to do it "right now". Subjects in the sad induction group gave significantly lower incentive ratings than subjects in the elated induction group. This finding was replicated by Clark and Teasdale (1985). Further analysis revealed that these subjects gave lower incentive ratings to social/active, social/passive, and isolated/active activities, and gave higher 26 incentive ratings to isolated/passive activities. Positive and Negative Word Recall Clark and Teasdale (1985) investigated the effects of the musical MIP on the recall of positive (eg. life) and negative (eg. death) words. Subjects given the sad induction recalled more negative words than positive words, while subjects given the happy induction recalled more positive words than negative words. Estimates of Success Teasdale and Spencer (1982) investigated the effects of musical MIP on subjects' estimates of the probability of future success and the number of past successes on a laboratory task. All subjects were given a decision-making task followed by the induction of either a sad or happy mood. Subjects given the sad induction gave lower estimates of the probability of future success and lower estimates of the number of past successes than subjects given the happy induction. These findings parallel the effects of naturally occurring clinical depression on estimates of future success (Loeb et al, 1971) and on the recall of past successes (De Monbreun & Craighead, 1977; Gotlib, 1981). In summary, this brief review suggests that the musical mood induction procedure produces a good analogue of mild, naturally occurring depression. It also highlights practical procedures which can strengthen this parallel. It is possible that some subjects will report feeling depressed when in fact their mood has not changed. In order to minimize this possibility, experimenters in the present study emphasized to subjects that they should base their mood ratings on how they actually felt, rather than on how they thought the experimenter would like them to feel. As an added precaution, a post-experiment questionnaire was utilized. This form stressed honesty in 27 responding and asked subjects whether or not they thought their mood changed during the induction procedure. A handwriting speed task was included as a variable manipulation check. Because writing speed scores varied considerably from individual to individual, data analyses took into account individual's pre-induction scores when evaluating post-induction scores. Dependent Measures The dependent measures in this investigation were chosen in keeping with Lang's three-systems model of fear reactions. Lang described such reactions as being comprised of three loosely-coupled components - behavioral, physiological and verbal. These components are highly interactive yet also partially independent, and thereby capable of responding differentially at any given time. For this reason, it is essential to assess all three systems in order to completely measure fear reactions. For the present investigation, approach/avoidance served as the behavioral measure, anticipatory heart rate as the physiological measure and subjective units of distress scales (SUDS) as the subjective measure. Self-efficacy was also assessed due to the probable effects of a depressed mood on this variable (Bandura, 1977). In addition, the relationship between self-efficacy and the return of fear was of interest. Resea rch Hypotheses The previously discussed literature led to the formulation of the following hypotheses: "Mood Hypotheses" 1) Subjects in an induced sad mood will find the target stimulus more fearful and experience lower self-eff icacy than subjects in an induced happy mood. 2) Subjects in an induced sad mood will exhibit slower reductions in fear than subjects in induced happy mood. 3) Subjects in an induced sad mood will experience greater return of fear than subjects in an induced happy mood. "State-Dependency Hypothesis" 4) Subjects who undergo their second session of fear re-duction while in a mood state congruent with that of their first session will exhibit more rapid habituation than subjects in an incongruent mood state. 29 Method In order to test these hypotheses, two groups of fearful Individuals were compared - one group which was happy and one group which was sad. Mood was manipulated in the laboratory using the musical mood induction technique. After this procedure, both groups (ie. happy and sad) underwent exposure therapy for fear reduction. The length of time taken to reduce fear was then compared between groups in order to assess differences in within-session habituation. To test return of fear, individuals came back after a period of time to have their fear level reassessed. During their second visit, all individuals who showed a return of fear underwent another session of fear reduction. To test the state-dependency hypothesis, half the subjects in each group experienced a different mood state during the second fear reduction compared with that of the first (ie. half of those who were happy the first time felt sad the second time, and vice-versa). The other half experienced fear reduction in a mood congruent state (ie. for both sessions they were either happy or sad). The difference in duration of fear reduction between session one and session two was then compared across groups. The following sections of this paper provide a detailed description of the way in which the above procedures were carried out. Subjects and Design Eighty-four female U.B.C. undergraduate students with a demonstrated fear of spiders or snakes were the subjects in this experiment. All received course credit in return for their participation. The experiment took place in a laboratory setting. Subjects underwent mood induction (either happy or sad) and then gradual, graded exposure to the feared stimulus. Two such sessions took place, four weeks apart. 30 The experiment was conceptualized as a 2 x 2 between-subject design. The first between-subject factor was the mood (happy vs. sad) in which fear was reduced during the first session, and the second factor was the mood (again, happy vs. sad) in which fear was reduced during the second session. The crossing of these two variables defined four fear reduction conditions as follows: 1) Subjects who underwent a happy mood induction in session one followed by a happy mood induction in session two. 2) Subjects who underwent a happy mood induction in session one followed by a sad mood induction in session two. 3) Subjects who underwent a sad mood induction in session one followed by a happy mood induction in session two. 4) Subjects who underwent a sad mood induction in session one followed by a sad mood induction in session two. Subjects were randomly assigned to each experimental condition. Experimental Design Sess. 1 Happy Sad Happy Sess. 2  Sad Experimenters and Therapists Two experimenters were required to carry out each session of this investigation. The role of the first experimenter was to induce the desired 31 mood (happy or sad). Two undergraduate psychology students and two graduate students in clinical psychology performed the mood inductions. The second experimenter served as therapist during the fear reduction portion of the experiment. There were two therapists involved in this study, both of whom were female graduate students in clinical psychology, trained in exposure and modelling techniques. One therapist reduced the fear of all the subjects afraid of spiders, and the other reduced the fear of all the subjects afraid of snakes. To prevent experimenter bias, the therapists were not told the experimental condition of the subjects (ie. they did not know whether the subject had undergone a happy or sad mood induction prior to fear reduction). As an added precaution, subjects were instructed not to talk to the therapist about earlier procedures. At the end of both sessions, the therapists rated the mood of the subject on a visual analogue scale (see Appendix A). The questionnaire utilized for this is described in detail below. The purpose of the rating procedure was to determine if the therapists could guess the mood of the subjects despite the fact that they had not been told which mood (happy or sad) had been induced. The training of therapists took place in a one hour session, after which the therapists were evaluated on the following items: a) Provision of information about the feared stimulus b) Use of modelling techniques c) Pacing of exposure d) Use of distraction e) Use of positive reinforcement f) Elicitation of fear ratings from subject g) Maintenance of a neutral mood 32 b) Obtaining cooperation of subject i) Overall ability to reduce subject's fear Each of the above items was rated on a 0 to 100 point visual analogue scale (see Appendix B). In order for therapists to participate in this investigation they had to obtain an average rating of at least 80. Therapists were informed that once the experiment was underway, treatment integrity checks would be carried out by an independent observer at random intervals. In actuality each therapist was rated twice during data collection. Therapists had to again obtain an average rating of at least 80 out of 100 during these checks. Had they not achieved this, they would have been retrained or replaced. Procedure This investigation was divided into the following five sections: (a) Screening Fear Survey Schedule (b) Selection Completion of Consent Form Behavior Avoidance Test (c) Session One Mood Induction Behavior Avoidance Test Fear Reduction Behavior Avoidance Test (d) Reassessment Behavior Avoidance Test (e) Session Two Mood Induction Behavior Avoidance Test Fear Reduction Behavior Avoidance Test 33 Debriefing The screening procedure took place on a group basis. After this point subjects were tested individually. Measures of heart rate, subjective distress, avoidance, mood and self-eff icacy were taken at various points throughout the experiment (details below). In addition, the length of time taken to complete fear reduction was recorded. Subjects were required to complete a handwriting task as a manipulation check (Clark, 1982) before and after mood induction procedures. The number of items written during this task was recorded. Presented below is a diagram of the procedures involved in this experiment (after initial screening) and the points at which data were collected: SELECTION SESSIONS 1 AND 2 (REASSESS) 1 DATA 2 DATA 3 DATA 4 DATA 5 Ss seated bat 1 h.r. hand- * o f m.i. mood/ m.i. compl. h.r. suds writing items 3 min. compl. consent f. mood cap task compl. (4 miri. s.e. mood adapt.) s.e. SESSIONS 1 AND 2 (CONTINUED) DATA 6 DATA 7 DATA 8 DATA 9 DATA mood hand- * of bat 2 h.r. f.r. dur. bat 3 h.r. h.r. writ. items suds mood/ suds s.e. task compl. 4 min. s.e. h.r. - heart rate suds - subjective units of distress s.e. - self-efficacy cap - closest approach point m.i. - mood induction f.r. - fear reduction dur. - duration bat - behavior avoidance test Screening 34 The fear survey schedule (F.S.S.) was administered to students enrolled in undergraduate courses in psychology (see Appendix C). Those who indicated on this questionnaire that they were "extremely fearful of" or "terrified of" spiders and/or snakes were contacted by telephone and asked to participate in the experiment. They were told the following: "This is calling from the UBC department of psychology. If you recall, you recently completed a questionnaire for us about fear. At present we are conducting a research project on factors that influence fear and were wondering if you would like to participate. Before you decide, let me tell you a little about our experiment. It consists of two sessions, four weeks apart, each of which will be about one hour in length. You will receive course credit for participating. In each session, we would ask you to do a very short writing task which involves simply writing down numbers. Then you would be asked to spend about twenty minutes listening to music and trying to experience certain emotions like happiness or sadness. We would also be asking you to slowly approach a live, harmless animal, of which you may or may not be afraid, while at the same time telling us how afraid you feel. Throughout all this, we would be recording your heart rate using a small monitor that clips onto your ear. Of course, you would be free to withdraw from the experiment at any time if you felt the need to. Do you have any questions at all? If the subject agreed to participate, the experimenter then arranged a convenient time. Selection Upon arrival at the laboratory, subjects were seated comfortably in a chair where they remained throughout the experiment. The heart rate monitor was explained to them and clipped to their earlobe. Subjects were then shown how to complete the mood and self-efficacy questionnaires. Finally, they were asked to read and sign the consent form (see Appendix D). Heart Rate Monitor Heart rate was assessed with a Sanyo Pulse Meter. This meter has a pulse 35 sensor that clips onto the earlobe. Pulse is displayed with a time count in 7-digit LCD. Mood and Self-Efficacy Measurement The mood scale for the subjects consisted of six divisions in the manner used by Sutherland, Newman and Rachman (1982). A sheet of paper with six lines of 10 cm. in length and marked from 0 to 100 was used as a score sheet. At the zero end of each line was typed a statement saying: "I do not feel at all "j and the 100 mm point on the scale was labelled "I feel extremely ". The six adjectives that were used on the six lines were: tired, anxious, sad, despondent, apprehensive and happy. Subjects were asked to place a mark anywhere along each of the lines to indicate their mood at that particular time (see Appendix E). As mentioned previously, therapists were also asked to rate the mood of the subject. The same visual analogue scale was used on their questionnaire. At the zero end of a 10 cm. line was the statement: "The subject is not at all ", and the 100 mm. point was labelled: "The subject is extremely ". Therapists were asked to rate the subjects on two dimensions - happiness and sadness (see Appendix A). A similar method was used to assess subjects' self-efficacy. On a sheet of paper was typed the following question: "How confident are you that you will be able to tolerate a live but harmless spider (snake) (closest  approach point for each subject) from you with minimal fear or discomfort?". At the zero end of a 10 cm. line was the statement "Not at all confident" and the 100 cm. end was labelled "Totally Confident". Subjects were asked to place a mark anywhere along the line to indicate their level of confidence at that particular time (see Appendix F). The above questionnaires were scored from 0 to 100 by measuring (with a 36 ruler) the distance from the zero cm point on the line to where the subject placed a mark. Behavior Avoidance Test Those subjects who agreed to participate underwent a behavior avoidance test (BAT) to further assess their fear level. Subjects were asked to approach the phobic stimulus (spider or snake) until they could go no further. To begin, subjects were placed eighteen feet from the stimulus and told the following: "Inside the container is a live, harmless garter snake (or live, harmless spider). Can you see it from where you are standing? In a moment I am going to ask you to report the peak amount of fear you are experiencing using a scale from 0 to 100 ("0" being no fear and "100" being terrifying fear). I will move the container toward you and when I reach where you are sitting, I would like you to put one finger near the snake (or spider) for five seconds while continuing to look at it. I will tell you when the five seconds are up. If you are unable to put your fingers near the snake (or spider), could you please let me know when the animal is as close to you as you can possibly tolerate? Do you understand? Please go ahead. (EXPERIMENTER SLOWLY MOVES SNAKE/SPIDER TOWARD SUBJECT) Now, I'd like you to report the peak amount of fear you are experiencing using the 0 to 100 scale." The subject's reported score out of 100 (SUDS) was recorded as a measure of subjective fear. The distance of the subject to the target stimulus at the end of the BAT was also recorded. This distance, which is termed "closest approach point" (CAP), was held constant for subsequent BATs. As a result, SUDS scores always corresponded to this point, allowing for comparison across BATs. Those subjects who on the first BAT were able to touch the phobic stimulus with a SUDS of less than 75 were excluded because they did not show sufficient fear for the purposes of this investigation. Out of 127 subjects, 28 (22%) were excluded on this basis. 37 Session One Handwriting Speed Task Subjects were asked to perform the handwriting speed task before and after the mood induction procedure. Subjects were given a blank sheet of paper and a pen, and were instructed as follows: "In a moment, I am going to ask you to do a short simple task for me. Please do not start the task until I finish explaining what I'd like you to do and indicate that you may begin. On the paper in front of you I would like you to write numbers in descending order beginning with "100" (eg. 100 ...99 ...98 ... 97...) until I ask you to stop. Do you understand? ... You may begin writing now." The experimenter timed the subject with a stopwatch for one minute, indicated to her when she could stop writing, and removed the paper and pen. A score for this task was derived by simply summing the number of items completed by the individual over the one-minute period. A second handwriting speed task was conducted after the mood induction procedure and was scored in the same manner as the first. Each subject's second score was then subtracted from the first resulting in a "change" score for each subject which could have either a positive or negative value. It was necessary to assess handwriting speed in this manner because scores were likely to vary considerably from individual to individual and thus pre-induction scores had to be taken into account when evaluating post-induction scores. It was expected that subjects who underwent a sad mood induction would have a negative change score averaging from -5 to -6. Those subjects who underwent a happy mood induction were expected to obtain change scores averaging from +5 to +6 (Alloy, Abramson & Viscusi, 1981). Differences in scores for these two groups were expected due to psychomotor retardation associated with a depressed mood. 38 Scores on the handwriting task were not involved in the criteria for determining sufficient mood change for continuation of the experiment. Instead, they were analyzed on a group basis at the end of the experiment as an added measure of the validity of the mood manipulation technique. Mood Induction After the behavior avoidance test, subjects were informed of the following: "In today's session you wil l listen to a selection of classical music that should help you develop a happy (sad) mood. However, music alone cannot create the desired mood, so you should try to think about something that makes you happy (sad). You may find it especially helpful to concentrate on happy (sad) events that you have personally experienced. While you are listening to the music, I wil l come in periodically and ask you to f i l l out some brief questionnaires. When I think that you have developed an appropriate mood, I wil l send in another experimenter and she wil l let you know what is going to happen next. You will be listening to the music throughout the rest of the experiment. Do you have any questions so far?" Upon receiving these instructions, headphones attached to a Prosonic Stereo Radio Cassette Player were placed on the subject by the experimenter. Through the headphones was played, at a comfortable listening volume, one of two selections of "happy" music (a segment of Eine Klein Nachtmusik (5 min: 10 sec) or Divertimento * 136 (4:10), both by Mozart) or one of two selections of "sad" music (Albinoni's Adagio in 6 Minor (6:32) or Barber's Adagio pour CjoxxIes_(5:33)). Each of these four selections was repeatedly recorded on one side of a 90 minute cassette tape. This saved the experimenter the trouble of having to stop and rewind the tape, which might have impeded the development of the desired mood state. Three minutes after the music began, and every three minutes thereafter, subjects rated their current mood. The music continued to play while the subjects made these ratings. In fact, the music, once started, did not stop 39 until the subjects were dismissed at the end of the session. Before the experiment continued, subjects were required to give themselves a rating of at least "75" (out of 100) in response to the appropriate item (ie. pertaining to happiness or sadness) on the mood questionnaire. In addition, ratings of the noncongruent mood had to be less than "25". Based on these criteria, 15 subjects out of 99 were excluded from participation. Ratings of all other moods did not play a part in mood induction criteria, but were analyzed after data collection. An additional point concerning the required levels of happiness/sadness is worthy of comment. Subjects did not know that the start of the fear reduction procedure was contingent upon their achieving a certain level of happiness or sadness. If the subjects had been aware of this contingency, it is possible that they would have rated their mood as being more extreme than it actually was in order to hasten the pace of the session. Fear Reduction Once the subject reached the requisite mood, the second handwriting task was completed. The subject then underwent a second behavior avoidance test which led directly into fear reduction. Every eight minutes throughout the fear reduction procedure, subjects completed the mood questionnaire in order to determine whether their mood was changing. Subjects remained seated and the phobic stimulus was presented and advanced, as appropriate, by the therapist. To begin, the therapist assumed position at a distance equal to the closest tolerated point indicated by the subject during the first BAT. The therapist had the phobic stimulus with her in a glass container. She gradually approached the subject at a speed which was comfortable for the subject. Once the subject was able to hold the container, the therapist took the stimulus out of the container and modelled various ways of handling it. 40 The subject was encouraged to look at and handle the phobic stimulus. In addition, she was asked at various points to report levels of fear on the 0 to 100 scale (SUDS) described previously. Fear reduction was considered complete when the subject reported a SUDS level of "10" or less (out of 100) while the stimulus was at a distance which corresponded to the closest approach point from the first BAT. It should be noted here that this criterion implies that subjects were required to express a 65 point reduction (at least) in fear from the beginning to end of the first session. The fear reduction procedure was timed, and the duration before reaching completion recorded by the therapist. Final measurements of heart rate, mood and self-efficacy were then taken. Subjects were asked to complete one last questionnaire which asked them to state honestly whether or not their mood had changed during the experiment (see Appendix G). Subjects who listened to happy music were then discharged, with the request that they return at the same time, four weeks later, to carry on with the research. Subjects who listened to sad music stayed on a while longer - talking to the therapist and listening to happy music - to ensure that their level of happiness/sadness was at least neutral before they left the laboratory. After the subjects had left, the therapists rated their mood exhibited during fear reduction on a scale of 0 to 100 as previously described. Reassessment Subjects returned four weeks later at which time they underwent a behavior avoidance test. Measures of heart rate, mood and self-efficacy were taken. Session two began immediately after this procedure. Session Two Procedures involved in the second session were identical to those in the 41 first. Upon completion of the experiment, subjects were thoroughly debriefed and given course credit. 42 Results For all analyses other than those dealing with state-dependency effects, conditions 1 and 2 (both of which involved a happy mood induction during session one) were combined to form the "Happy group", and conditions 3 and 4 (both of which involved a sad mood induction during session one) were combined to form the "Sad group". Heart-rate was excluded from all analyses because measurement was inaccurate due to movement artifact. Missing Data Four subjects did not return for their second session. They were therefore excluded from analyses involving return of fear and state-dependency effects. Four subjects did not complete the self-efficacy questionnaire at the end of session one and three did not complete it at the beginning of session two. Seventeen subjects were eliminated from the major analyses because they did not meet the fear reduction criterion (subjective fear of ten or less) within the available time period (approximately thirty minutes). The data from these subjects were used in several post hoc analyses presented under the section entitled Additional Analyses. Preliminary Analyses 1) The first analysis of the data was undertaken to determine if the Happy and Sad groups differed on major dependent variables prior to the experimental manipulation. For this purpose, the following measures were collected before the mood induction procedure: a) subjective fear level, b) closest approach point, and c) self-eff icacy. It was not expected that groups would differ on these variables. T-tests were utilized to compare the Happy and Sad groups. No significant differences were found. These results are presented in Table 1. 43 Table 1 Mean Scores on Major Dependent Variables Before Mood Induction Group Variable Happy3 Sad b Significance Marginal Meanc Reported Fear 82.7 81.9 t(82)=0.54 82.3 (7.5) (7.1) (7.3) Closest 9.9 144 1(82)=0.88 11.5 Approach Point (19.3) (22.7) (21.0) Self-Efficacy 30.9 32.0 t(80)=0.85 31.5 (26.2) (24.8) (25.5) *C<.05 **n<.oi NDle. Reported Fear and Self-Efficacy are rated on a 0 to 100 scale. Units for Closest Approach Point are inches. Numbers in parentheses are standard deviations. an=43. bn=4i. CN=84. 44 2) Mood Manipulation In order to verify that the mood induction procedure was effective in producing happy and sad moods, the following four analyses were performed: a) Self-reported measures of happiness and sadness before and after the mood induction procedure were compared for the Happy and Sad groups using T-tests. These scores were collected once prior to mood induction and then every eight minutes (approximately) throughout the fear reduction procedure. In order to control for inflation of Type 1 error, the alpha level was reduced to .01 (.05/5). The two groups did not differ on pre mood induction scores of happiness, i(82)= 1.33, n.s., or sadness, i(82)=0.96, n.s. After mood induction, Happy and Sad groups differed significantly on both happiness and sadness scores throughout the fear reduction procedure (see Appendix H). These results are displayed in Figures 1 and 2. b) Scores on the writing speed task were also analyzed to determine the effects of the mood induction procedure. Pre to post change scores were calculated by subtracting the number of digits written by subjects before mood induction from those written after. These change scores were then compared between groups using a T-test. The results indicate that the Happy group wrote more numbers after mood induction (M.=+2.6,5X2=47) and the Sad group wrote fewer numbers (M.=-0.9, SX2=5.1). The difference between the two groups was significant, i(82)=2.80, p_< 01. c) In order to evaluate the subsidiary effects of the mood induction procedure, subjects were asked to rate their current feelings of sadness, tiredness, anxiety, despondence, apprehension and happiness on the mood questionnaire. Inter-item correlations on responses collected before and after mood induction were calculated using Pearson's correlation coefficients. In order to control for inflation of Type 1 error, the alpha level 100 Figure 1. Mean Reported Sadness Pre and Post Mood Induction 90 H 0 H 1 1 1 1 1 T Pre M.I. 1 2 3 4 5 T I M E • Happy Group + Sad Group R E P O R T E D H A P P I N E S S 9t? 47 was reduced to .0083 (.05/6). Significant correlations are displayed in Tables 2 and 3. d) As a final check on the mood manipulation, the percentage of subjects who indicated that their mood had "honestly changed" during the experiment was calculated. 79 of the 84 subjects (94%) replied in the affirmative to this question. Major Analyses I) Within Session Effects In order to test Hypothesis 1, self-report measures of fear and self-eff icacy were collected immediately after the mood induction procedure. To test Hypothesis 2, the length of time taken to reduce subjective fear to ten or less was recorded for each subject (duration). It was hypothesized that Happy and Sad groups would differ on these three variables (subjective fear level, self-efficacy, and duration of fear reduction). To determine this, a Hotel ling's T^ was utilized. The results indicate that the two groups differed significantly, £(3,63)=3.71, o<.05. Univariate analysis of each factor revealed that the two groups were significantly different on measures of subjective fear and self-efficacy, but not on the length of time taken to reduce fear (see Appendix I). These results are presented in Table 4. To further assess Hypothesis 1, 2 (group) x 2 (pre and post mood induction) analyses of variance were computed with subjective fear and self-efficacy as the dependent measures (see Appendix J). Significant interactions were found for both these variables (subjective fear: £( 1,82)=8.92, p<.01, self-efficacy: £( 1,79)=3.97, p<.05). T-tests were then utilized to evaluate the sources of differences in the means. In order to control for inflation of Type 1 error, the alpha level was reduced to .0125 (.05/4). The results ( as shown in Figures 3 and 4) indicate that fear level decreased significantly after a happy mood 48 Table 2, Intercorrelations Between Items on Mood Questionnaire Before Mood Induction sad tired happy anxious despondent tired .352* happy -.381* -.106 anxious .117 .134 -.061 despondent .401* .239 -.170 .186 apprehensive .134 .172 -.049 .434* .213 *p_<.0083. m&. N=84. . 4 9 Tab le 3 Intercorrelations Between Items on Mood Questionnaire After Mood Induction sad t i r e d happy anxious despondent t i r e d .564* happy -.962* -.571* anxious .214 .312* -156 despondent .789* .584* -.754* .241 apprehens ive .243 .289* -.182 .621* .335* *p.<.0083 Nate. N=84. 50 Table 4 Mean Scores on Major Dependent Measures After Mood Induction Group Variable Happy3 Sad b Significance Marginal Meanc Reported Fear 60.8 72.2 E(1,65)=7.93** 66.4 (18.7) (18.3) (18.5) Self-Efficacy 45.5 33.2 E(1,65)=5.32* 39.2 (25.1) (22.0) (23.6) Duration 17.1 21.3 E(1,65)=1.91 19.2 of Fear Reduction (7.8) (11.0) (9.4) *0<-05 **p_<.01 IMS.. Reported Fear and Self-Efficacy are rated on a 0 to 100 scale. Units for Duration are minutes. Numbers In parentheses are standard deviations. aQ=34. bn=33. CN=67. Figure 3. Mean SUDS Pre and Post Mood Induction PRE M.I. POST M.I. I^XI Happy Group I I Sad Group >• o < u u u UJ Figure 4. Mean S e l f - E f f i c a c y Pra and Pott Mood Induction I//I Happy Group 53 induction, i(42)=6.80, ja<.0125, and after a sad mood induction, i=(40)=3.94, p_<.0125. However, this difference was significantly greater after the happy mood induction, i(82)=2.82, p_<.oi25. Self-efficacy also increased significantly, i(40)=2.97, p_<.0125, after a happy mood induction. There was no change in self-efficacy after a sad mood induction, i(39)=0.52, n.s. 2) Between Session Effects (Return of Fear) To evaluate Hypothesis 3, measures of fear were taken at the end of session one and at the beginning of session two. Return of fear was calculated by subtracting the first measure from the second. The following analyses were then conducted based on the return of fear measure: a) A T-test was utilized to compare the mean return of fear for the Happy and Sad groups. Results indicate that there was no significant difference between the two groups, 1(61 )= 1.52, n.s. The mean return of fear for the Happy group was 16.9 (5X2=19.3). The mean return of fear for the Sad group was 25.8 (SD=22.8). b) To further assess the factors influencing return of fear, subjects were divided post hoc into "Return of Fear" and "No Return of Fear" groups. Subjects who displayed an increase in fear level of more than ten points from the end of the first session to the beginning of the second session were classified as "Return of Fear" subjects (n=29). Those individuals whose fear level did not increase by greater than ten points were classified as "No Return of Fear" subjects (0=30). Hotelling's T 2 was utilized to determine whether the above groups differed on the following dependent measures: a) duration of fear reduction during the first session, b) self-efficacy at the beginning of the first session, c) self-efficacy at the beginning of the second session, d) "natural" mood (happy and sad) before mood induction during the first session, e) induced mood (happy and sad) during the first session, and f) 54 "natural" mood (happy and sad) during assessment of return of fear at the beginning of the second session. The results of this analysis indicate that the two groups differed significantly, E(9,49)=5.12, jai'OI. Univariate analysis of each factor was then conducted (see Appendix I). The results are presented in Table 5. Further analysis was done using a chi-square to compare the number of subjects who showed return of fear in the Happy and Sad groups. Table 6 shows that there were significantly more sad subjects showing return of fear and more happy subjects showing no return of fear, x2(1 )=5.78, p.<.05). 3) State Dependency To test Hypothesis 4, those subjects who showed return of fear (N=29) underwent a second session of fear reduction. The length of time taken to reduce these subjects' fear to ten or less was recorded. A savings score was computed by subtracting the duration of the second fear reduction session from that of the first. To test for state-dependency effects, subjects were divided into two groups based on their induced mood during the two fear reduction procedures. Subjects who had the same mood induced for both sessions formed the "Congruent" group (n=16), and subjects who received different mood inductions during sessions one and two formed the "Incongruent" group (n=13). A T-test was then done comparing the mean savings scores of the Congruent group (£3=12.3, 50=8.3) and Incongruent group (£1=6.4, 5Q-1I.4). There was no significant difference between groups, i(27)=1.60, n.s. To further assess savings scores, independent of state-dependency, two more T-tests were done with subjects regrouped in the following ways: 1) happy mood during session one (H=7.6, SD=10.2) compared with sad mood during session one (L1=l 1.3, Table 5 Mean Scores on Relevant Variables Based on Return of Fear 55 Group Variable No Return of Fear3 Return of Fear5 Significance Marginal Meanc Duration of Fear Reduction (sess. l) 15.3 (9.3) 21.6 (11.5) E(1.57)=5.33* 18.5 (10.4) Self-Efficacy (sess. 1) 31.9 (26.5) 36.7 (24.2) E(1.57)«0.52 34.3 (25.4) Self-Efficacy (sess. 2) 77.4 (17.2) 56.0 (21.8) £( 1,57)= 17.82** 66.7 (19.5) Natural Mood (sess. 1) Happy 65.7 (15.7) 51.7 (18.0) E( 1,57)= 10.20** 58.7 (16.9) Sad 12.5 (17.2) 19.8 (21.8) E(l)57)=2.24 16.2 (19.5) Induced Mood (sess. 1) Happy 59.0 (32.8) 38.3 (33.6) E(l,57)=5.72* 48.7 (33.2) Sad 33.2 (35.6) 54.0 (36.4) £(1,57)=4.92* 43.6 (36.0) Natural Mood (sess. 2) Happy 68.8 (16.4) 54.1 (21.7) E(l,57)=8.72** 61.6 (19.0) Sad 15.6 (17.7) 23.2 (20.6) E(l,57)=2.33 19.4 (19.2) *C<05 **p<01 Note Numbers in parentheses are standard deviations. Units for duration are minutes. Eight cases not processed due to missing data. aD=30. bD=29. CN=59. 56 Table 6 Chi Square Showing the Number of Subjects in Happy and Sad Groups With Return of Fear Group* Happyb Sadc No Return of Fear 21 11 Return of Fear 11 20 x2(1)=5.78, n<.05 Note Four cases were not processed due to missing data. aU=63. bn=32. cn=3l. 57 SJi=12.2),i(27)=0.96, n.s. 2) happy mood during session two (LM 1.6, SIM 2.2) compared with sad mood during session two (H=8.3, 50=8.4), t(27)=0.87, n.s. Finally, a 2 (mood during session one) x 2 (mood during session two) analysis of variance with savings score as the dependent measure was calculated, £(3,25)= 1.42, n.s. (see Appendix J). In summary, no significant differences in saving scores were found in any of the above analyses. The mean savings score for the entire sample (N=29) was 9.2 minutes (SD= l O.O). 4) Additional Analyses a) Successful vs Partially Successful Subjects A post hoc analysis was done comparing those individuals whose subjective fear was successfully reduced to ten ("Successful** group, n=67) with subjects whose fear could be only partially reduced within a thirty minute period ("Partially Successful" group, n=17). The latter group displayed a mean level of fear of 43.3 (SJQ=18.8) after an average of thirty-four minutes of exposure to the feared stimulus (SJQ=3.2 minutes). Eight subjects in the Partially Successful group underwent a happy mood induction, and nine underwent a sad mood induction. Eight received therapy from Therapist 1, and nine received therapy from Therapist 2. Hotelling's T 2 was utilized to compare the Successful and Partially Successful groups on the following set of dependent measures: a) self-eff icacy at the beginning of session one, b) self-eff icacy at the end of session one, c) self-efficacy at the beginning of session two, and d) return of fear. This analysis indicated that the groups differed significantly, E(4,68)=7.46, p_<OI. Univariate analysis of each factor was then conducted 58 (see Appendix I). The results are presented in Table 7. Further analysis was done to examine the interaction between induced mood during fear reduction in session one and natural mood during testing for return of fear at the beginning of session two. Natural happy mood at the beginning of session two was divided into two groups ("Happy" and "Not Happy") based on a median split (median=65). A 2 (induced mood during fear reduction in session one) x 2 (natural mood at beginning of session two) analysis of variance was then computed with return of fear as the dependent measure (see Appendix J). A significant interaction was found, £(1,59)=469, P/.05. Tukey tests were performed to investigate the sources of differences in the means. The results suggest that those individuals who had their fear reduced in an induced happy mood showed less return of fear when they were reassessed in a naturally happy mood (£1=3.5, SJQ= 140) compared with an unhappy mood (M.=28.7, SQ=25.2), Q(4,59)=4.61, p.<.01 • No s u c n interaction was found for subjects who had their fear reduced in a sad mood, Q(4,59)=0.46, n.s. These findings are displayed in Figure 5. A chi-square was calculated comparing the number of subjects in the four groups (divided according to induced mood and natural happiness) who showed return of fear. These results were significant, x.2(3)=8.36, p<.05, and are displayed in Table 8. The significant difference appears to lie in the "Induced Happy - Natural Happy" cell. For this group, thirteen subjects showed no return of fear, while only two subjects showed return of fear, b) Therapist (*or Stimulus) Comparisons Two analyses were done to investigate whether there were any differences (^Therapist effects are confounded with feared stimulus effects.) 59 Table 7 Mean Scores on Relevant Variables for Partially Successful and Successful Subjects Group Variable Partially Successful5 Significance Marginal Successful3 Meanc Self-Efficacy 28.1 36.2 F( 1,7 0=0.28 32.2. (begin.of sess.l) (18.3) (26.9) (22.6) Self-Efficacy 59.1 85.3 F(l, 70=23.72** 72.2 (end of sess.l) (20.7) (15.5) (18.1) Self-Efficacy 57.4 67.5 F( 1,70=3.99* 62.5 (begin.of sess.2) (14.7) (22.0) (18.4) Return of 10.7 21.3 F(1, 70=5.48* 16.0 Fear (15.5) (23.5) (19.5) *U<.05 **rj<.01 Note. Self-Efficacy variables are rated on a 0 to 100 scale. Numbers in parentheses are standard deviations. Eleven cases not processed due to missing data. an=i5. CN=73. Figure 5. Mean Return of Fear \S S\ Happy Group I 1 Sod Group NOT HAPPY HAPPY N A T U R A L H A P P I N E S S Note. Groups divided by induced mood during fear reduction (sess.l) and by median split of natural happiness scores during testing for return of fear (sess.2). 61 Table 8 Chi Square Showing the Number of Subjects With Return of Fear Group3 I.Happy I. Sad I.Happy I. Sad N. Happy N. Happy N. Not Happy N. Not Happy No Return of Fear 13 6 8 5 Return of Fear 2 10 9 10 x2(3)=8.36, fi<.05. Note, Groups divided by Induced mood during fear reduction (session 1) and by median split of natural happiness scores during testing for return of fear (session 2). Mndueed (session 1), N=Natural (session 2). Four cases not processed due to missing data. aN=63. 62 between therapists on major dependent variables. First, a 2 (therapist) x 2 (group) analysis of variance was computed with return of fear as the dependent measure (see Appendix J). No significant main effects for therapist, £( 1,59)=2.33, n.s., or for group, £( 1,59)=2.70, n.s., were found. No interactions were present, £( 1,59)=.0.43, n.s. This analysis was then repeated with duration of fear reduction (minutes) as the dependent measure (see Appendix J). A significant main effect for therapist was found, £( 1,63)= 10.03, p_<.01. There was no main effect for group, £(1,63)=2.83, n.s., and there was no significant interaction, £( 1,63)=2.55, n.s.. These results (as shown in Figure 6) suggest that Therapist 1 reduced fear more quickly (£1= 14.5, 50=8.3) than did Therapist 2 (£1=22.1,50= 15.3). To evaluate whether therapists remained blind to the experimental condition of subjects, they were asked to rate each subject's level of happiness and level of sadness at the end of the fear reduction procedure. To analyze this data, two 2 (therapist) X 2 (group) analyses of variance were computed with a) therapist ratings of happiness and b) therapist ratings of sadness as the dependent measures (see Appendix J). Therapist and group main effects were found for both these analyses. No interactions were present (happy ratings: £( 1,75)= 1.61, n.s., sad ratings: E( 1,75)=0.02, n.s.). The results of these analyses (shown in Figures 7 and 8) indicate that both therapists judged the subjects in an induced happy mood to be happier (£1=56.9, 5Ji=20.9) than those in an induced sad mood (£1=40.7, 50=17.3), £(1,75)=11.69, c<.001. In addition, those in an induced sad mood were judged to be sadder (£1=54.7,50= 19.9) than those in an induced happy mood (£1=41.7, 50=19.3), £(1,79)=7.84, p.<.01. The results further suggest that Therapist 1 rated subjects as sadder (£1=57.1, 50=20.2) than did Therapist 2 (£1=39.3, 50= 17.4), E( 1,79)=22.98, p_<.001 • Conversely, Therapist 2 rated subjects as l/X I Hoppy Croup I 1 Sod Group Figure 7. Mean Happiness Ratings Therapist by Group THERAPIST 1 THERAPIST 2 l/Xl Happy Group Sad Group Figure 8. Mean Sadness Ratings Th•rapist by Group TOO -i 90 H 80 THERAPIST 1 THERAPIST 2 L/X I Happy Group I I Sod Group 6 6 happier (£1=55.2, SD= 19.8) than did Therapist 1 (£1=42.3, SD=15.5), E(l,79)=20.38,p_<.001. 67 Discussion This study evaluated the effects of a mood induction procedure on fear reduction in a sample of eighty-four female university undergraduates fearful of spiders or snakes. It was predicted that mood would have an impact on subjective levels of fear, reported self-eff icacy, within-session habituation and between-session return of fear. Mood was also expected to have an effect on heart-rate - a physiological measure of fear. However, as mentioned previously, heart-rate could not be adequately assessed because of movement artifact, and thus this component of fear will not be addressed in the following discussion. Major findings of this study will be evaluated in terms of experimental, manipulation, within-session changes, between-session effects, and state-dependency effects. Additional findings related to the success of the fear reduction procedure and to therapist differences will then be addressed. Finally, conclusions and recommendations for future research will be discussed. Before turning to the major hypotheses, it is essential to verify that the mood induction procedure was effective in producing happy and sad moods. For this purpose, self-reported mood measures were taken immediately after mood induction and approximately every eight minutes during the fear reduction procedure. Happy and sad groups differed significantly throughout this time period indicating that the manipulation was effective in producing the desired mood for the required length of time. In addition, the self-reported measure of mood change (which stressed honesty in responding) indicated that almost all subjects (94%) felt that their mood had changed in the desired direction during the mood induction procedure. This value is similar to the 87% reported by Clark and Teasdale (1985). 68 The results of the handwriting task show that as predicted subjects wrote more quickly after a happy mood induction, and more slowly after a sad mood induction. Although the magnitude of this difference was somewhat less than that reported by other researchers (eg. Alloy et al, 1981), these findings show that a sad mood was associated with psychomotor retardation. This provides further evidence that the mood manipulation was effective. To assess subsidiary effects of the mood induction procedure, intercorrelations between items on the mood questionnaire were examined. Before mood induction, a sad mood was positively correlated with feeling tired and despondent, and negatively with feeling happy. After mood induction, these same relationships existed, and were of a much stronger magnitude. In addition, several other significant correlations emerged. A tired mood was positively correlated with anxiety, despondence and apprehension, and negatively with happiness. Despondence was also correlated negatively with happiness, and positively with apprehension. From the results of the inter-item correlations on the mood questionnaire, it appears that individuals who underwent a sad mood induction became sadder, more tired, more despondent and much less happy. Those individuals who had a happy mood induced became happier, less tired, less despondent, and less sad. Neither happiness nor sadness was correlated significantly with anxiety or apprehension. However, there are low correlations between these variables suggesting that a sad mood is accompanied by some anxiety, while a happy mood is not. These results are consistent with the nature of naturally occurring depressed mood states which are often accompanied by elevated levels of anxiety. Moreover, these results are similar to those obtained by Sutherland et al (1982) with their use of the musical mood induction procedure. 69 In summary, the experimental manipulation checks suggest that the musical mood induction procedure resulted in the desired mood changes. With this reassurance, the major hypotheses of this investigation can now be addressed. Hypothesis 1 proposed that individuals who underwent a sad mood induction would report greater subjective levels of fear and lower levels of self-efficacy than would subjects who underwent a happy mood induction. This hypothesis was confirmed. After mood induction, subjects in both the happy and sad groups reported lower levels of fear during the second behavior avoidance test compared with the first. However, individuals in a happy mood showed a greater decrease in fear than individuals in a sad mood. In addition, these subjects reported having more confidence in their ability to tolerate the presence of the feared animal (self-efficacy). Confidence increased after the happy mood induction while there was no change in confidence after the sad mood induction. One cannot determine from this experiment whether a sad mood leads to an increase in fear and a decrease in self-eff icacy, or whether a happy mood decreases fear and increases self-efficacy. In order to draw such conclusions, a control group of subjects who did not undergo mood induction would be necessary for purposes of comparison. It was predicted in Hypothesis 2 that the increased fear and decreased self-efficacy of sad subjects would delay within-session habituation and therefore necessitate a longer period of fear reduction in order for this group to meet the experimental criterion. This hypothesis was not supported. There was no difference between the happy and sad groups on the length of time taken to reduce fear. The average duration of fear reduction was nineteen minutes. This finding is consistent with the duration (eighteen 70 minutes) in an experiment by Rachman and Lopatka (personal communication), but is somewhat lower than that of thirty minutes reported in a different experiment by Rachman, Robinson and Lopatka (1987). It is possible that there was no difference between the happy and sad groups in within-session habituation because all subjects' fears reduced rapidly. As a result, there may have been insufficient variance in the duration measure to allow significant group differences to emerge. If this study is replicated on a clinical population of phobics, for whom fear reduction typically takes much longer than thirty minutes, a difference in within-session habituation might be found. However, it should be noted that the absence of a delay in within-session habituation in the Sad group is consistent with Philips' (1985) results. She found no differences between her depressed and nondepressed phobics on this measure. In addition, Foa's original (1982) research, in which she reported delays in habituation in depressed obsessive-compulsive individuals, was not replicated when a larger sample was used (Foa, 1983). Taken together, the findings of this and other investigations suggest that there is no delay in within-session habituation associated with sadness. It is possible that the presence of a reassuring therapist who is a competent model overrides the higher fear and lower self-efficacy associated with a sad mood. One might therefore hypothesize that once these individuals are away from the positive influence of the therapist, they remember the fear and lack of confidence they initially felt during fear reduction and experience a return of fear. More precisely, Hypothesis 3 proposed that subjects who had their fear reduced in a sad mood would show more return of fear than subjects whose fear was reduced while in a happy state. When a direct comparison of the magnitude of return of fear was made for Happy and Sad groups, no 71 significant differences emerged. However, when the subjects were divided post hoc according to whether they showed return of fear, differences in mood were found. Of the thirty-one subjects who showed a return of fear four weeks after fear reduction, twenty (65.5%) had had their fear reduced while in a sad mood. In general, the individuals who showed return of fear in session two were in a sadder mood during fear reduction in session one. Also of interest is the finding that these individuals were less happy at the time of testing for return of fear. It appears that the group which showed the least return of fear was the one in which subjects had their fear reduced in an induced happy mood and were retested four weeks later while in a naturally happy mood. This interaction is not surprising. When subjects returned for their second session, it is likely that they recalled their first session of fear reduction. Theories of mood and memory relationships (eg. Bower, 1985) predict that if the subjects were in a happy mood while thinking back, they would remember this experience more positively than if they were in a sad mood. This would result in a greater expectation of control and lower reported fear during the second session. Further, if subjects initially had their fear reduced in a happy mood, they would likely have experienced less fear and greater confidence leading to even more favourable recall during retest ing. Interestingly, it was found that a sad mood at the time of retesting had no effect on the return of fear. One would expect the same principles outlined above to be in operation resulting in the greatest return of fear in individuals who had their fear reduced in a sad mood and then were retested in a sad mood. However, reported sadness at retest appears to be unrelated to return of fear. The reason for this may be that too few subjects reported feeling sad at the beginning of the second session to properly assess the relationship 72 between sadness and return of fear. Another unexpected finding was that those subjects who did not show return of fear were also naturally happier before mood induction during session one. This finding may be partially explained by the fact that natural happiness during session one was modestly correlated (£=0.30, p_<0.01) with natural happiness during session two. This suggests that a significant proportion of the subjects who were happy during the beginning of session one were also happy during retesting in session two. This would result in both measures being related to the return of fear. Another variable that was found to be associated with return of fear was self-efficacy. The relationship between return of fear and self-efficacy was restricted to measurements of self-efficacy taken immediately before retesting of fear. There was no difference in self-efficacy at the beginning of the experiment when comparing Return of Fear and No Return of Fear groups. In other words, subjects were able to predict how tolerant they would be immediately before exposure to the feared stimulus in session two, but their feelings of self-efficacy four weeks earlier were unrelated to return of fear. These findings are consistent with those obtained by Craske and Rachman (1985). In their investigation, self-efficacy measures collected prior to therapy were not shown to be useful in predicting the return of fear several months post-treatment. However, those who reported a reduction in self-efficacy between post-treatment and followup showed an increase in subjective fear at followup. Given that there are significant differences in both self-efficacy and happiness at the beginning of session two for Return of Fear and No Return of Fear groups, it is not surprising that these two variables are significantly correlated (r=0.40, p_<.001). However, self-efficacy and mood are not 73 significantly correlated at the beginning of session one. This is probably because almost all subjects reported low self-efficacy after having undergone a behavior avoidance test during which they felt very frightened. This restricted range may also be the reason that initial self-efficacy was not related to return of fear. The final variable related to return of fear in this investigation is the duration of fear reduction in session one. The subjects who showed a return of fear during session two spent more time undergoing fear reduction during session one. The reasons for this finding are not immediately clear. Traditional theories of exposure would predict the opposite - ie. that the longer one is exposed to the feared stimulus, the less likely one would be to experience return of fear. However, it may be that a fear which takes longer to reduce is more resistant than one which reduces quickly. Further, it may be that these resistant fears are more difficult to eliminate completely and thus are more likely to manifest at some later date. This relationship between duration of fear reduction and return of fear has been recently repl icated by Rachman and Lopatka (1987). The fourth major hypothesis of this investigation proposed that there would be a state-dependency effect of mood on fear reduction. This hypothesis applied to subjects who showed a return of fear at the beginning of session two and therefore underwent a second session of fear reduction. It was predicted that the second fear reduction would take less time if subjects were in a mood state congruent with that during their first session. If they were in an incongruent mood state, fear reduction was expected to take longer. This hypothesis was not supported by the results of this study. Both Congruent and Incongruent groups showed a more rapid decrease in fear during their second session compared with their first session. However, there was 74 no significant difference between the groups. One reason for this may have been that the sample size was relatively small (N=29). In addition, results may have been affected by the nonclinical sample used. Perhaps if this study is replicated with clinical phobics, fear reduction during the second session would not been so rapid, and group differences in duration would emerge. Finally, mood state dependency is not a robust phenomenon, and appears only under highly specific conditions involving a lack of other memory cues (Eich, 1980). In this investigation, there were many salient cues (ie. laboratory surroundings, therapists) that the subjects could utilize to aid in recalling their previous responses to the feared stimuli. In summary, two of the major hypotheses in this investigation were supported and two were not. As predicted, sad individuals showed greater subjective fear and lower self-efficacy compared with happy individuals (Hypothesis 1). However, these differences did not have any impact on within-session habituation (Hypothesis 2). Sad individuals showed more between-session return of fear (Hypothesis 3). Finally, subjects who underwent their second fear reduction in a mood state congruent with that of their first session did not show more rapid habituation than subjects in an incongruent mood state (Hypothesis A). The post hoc comparisons of Successful and Partially Successful subjects led to some interesting findings regarding self-efficacy and return of fear. Those subjects who still reported fear after thirty minutes of gradual exposure also reported lower self-efficacy at the end of session one and at the beginning of session two. However, as with the return of fear findings, self-efficacy at the beginning of session one had no impact on whether fear would be completely reduced. Once again, self-efficacy appears to be related to the immediate experience of fear, but not to future levels of fear. 75 An unexpected finding was the lower level of return of fear in the Partially Successful group. This finding has also been observed in two other investigations by Rachman et al (1987) and Rachman and Lopatka (1987), and is contradictory to what would be predicted by traditional learning theorists. Return of fear is similar to spontaneous recovery as described by Skinner (1938) who claimed that this phenomenon reflects incomplete extinction. One would therefore predict that the Partially Successful group should show more return of fear as a function of incomplete fear reduction. The lower return of fear in the Partially Successful group is also inconsistent with the previously mentioned explanation that highly resistant fears are more vulnerable to return of fear. The Partially Successful group was the most resistant to fear reduction techniques in this investigation, yet it exhibited less return of fear than the Successful group. Perhaps the difference in the partially successful subjects lies in their self-report of fear. Many did not appear to be fearful to the therapist observing them, yet they continued to report fear levels of approximately forty. This suggests that there may have been desynchrony in the three systems of fear. According to Rachman and Hodgson (1974), the usual pattern of desynchrony in fear reduction is behavioral change, followed by physiological change, and finally by subjective change. This cognitive lag could have been present in the partially successful subjects resulting in their continued report of fear after behavioral indications were no longer evident. It was not until sometime after the first session that the subjective report of fear became concordant with the behavioral index. When these subjects returned after four weeks, they did not show as much return of fear because it was off-set by the "catching up" of the subjective component with the behavioral component of fear. Unfortunately, without accurate assessment of 76 physiological indices of fear, it is impossible to form a clear picture of the pattern of fear reduction displayed in the partially successful subjects. Thus, the explanation for the difference in return of fear between the Successful and Partially Successful groups remains speculative. The final set of analyses in this investigation was carried out to determine if there were any differences between the two therapists on a number of dependent variables. It was not predicted that there would be any therapist differences. However, the results of the analysis on duration of fear reduction indicated that Therapist 1 reduced fear more quickly than did Therapist 2. This difference does not imply that one therapist was better than the other. It is important to note that both therapists were within one standard deviation of the mean reported duration in this investigation and in the Rachman and Lopatka (1987) and Rachman et al (1987) studies. Moreover, each therapist was rated twice during the course of data collection on their use of fear reduction techniques and overall ability to reduce fear. Both received excellent ratings. There are a number of possible explanations for the observed therapist difference in the length of time to reduce fear. There was a therapist/animal confound in this experiment because Therapist 1 reduced the fear of all the subjects afraid of snakes and Therapist 2 reduced the fear of all the subjects afraid of spiders. It is therefore possible that the observed difference in duration between therapists actually reflects a difference in the amount of time it takes to reduce the fear of spiders and of snakes. However, there are no reports of such a difference in the literature on fears. Moreover, research by Rachman and Lopatka, in which the identical fear reduction procedure was used on subjects from the same student population, shows no difference in the length of time it takes to reduce fears of spiders and of snakes (personal 77 communication). This suggests that the duration difference observed in this investigation is associated with the therapists, rather than with the feared stimuli. It is possible that the two therapists differ in the style in which they conduct therapy (ie. Therapist I may be more demanding than Therapist 2). Unfortunately, this aspect of the therapists' performance was not evaluated. There is also the possibility that the therapists differed in their perspective regarding the present study. Therapist 1 was the primary investigator and thus may have been more strict about following the experimental protocol of reducing fear as rapidly as possible. Therapist 2, on the other hand, may have been more responsive to the discomfort of the subjects and therefore slowed the exposure process when subjects appeared quite fearful. Although the relationship between therapist and duration of fear reduction was not expected, it had no impact on the major hypotheses. There was no interaction between therapist and condition showing that neither therapist reduced fear more rapidly or slowly as a function of the mood of the subject. In addition, there was no difference in return of fear related to therapist. An attempt was made to ensure that therapists were blind to the treatment condition of subjects. To check on this, therapists were asked to rate the mood of the subjects following exposure. Analysis of these scores revealed that both therapists were able to differentiate the happy subjects from the sad subjects. These results were not predicted prior to carrying out this investigation. Moreover, both therapists were surprised to find that they were able to recognize the mood of the subjects and reported that they had felt very uncertain when making their happiness and sadness ratings. Although this finding complicates the interpretation of the results, it adds to the validity of the mood induction procedure. The ratings are a source of 78 information about the behavioral component of emotions. Not only could happy and sad subjects be differentiated on the basis of their subjective reports of happiness and sadness and on their psychomotor activity, but also on the basis of their behavior. Although therapists could not state definitively how they knew whether subjects were happy or sad, they did mention some behavioral clues. For example, the therapists reported that sad subjects did not talk or smile as much as happy subjects. Given that the therapists were not blind to treatment conditions, one must ask what impact their awareness had on the outcome of this study. Of the four major hypotheses, it is unlikely that two could have been affected by the therapists' knowledge of the subjects' mood. Hypothesis 1 involved measures of self-efficacy and subjective fear ratings, both of which were taken before the therapist had any contact with the subject. This rules out any possibility of therapist influence. Hypothesis 3 involved return of fear measures which were taken by individuals other than the therapists. Two of these individuals were undergraduate students in psychology who were unaware of the experimental hypotheses, and two were clinical graduate students. All four experimenters were blind to treatment condition when testing for return of fear. The two hypotheses that could have been affected by therapists' awareness of subjects' mood are Hypotheses 2 and 4, both of which are based on the duration of fear reduction measurement. However, neither of these hypotheses were supported in this investigation. One final point should be made about the therapists in this study. Therapist 2 was blind to the experimental hypotheses, while Therapist 1 was not. Despite this difference, there were no therapist x condition interactions. This suggests that knowing the hypotheses under investigation had no effect 79 on the results of this study. In summary, there does not appear to be any evidence that therapist awareness of the mood of subjects, or knowledge of the research hypotheses, had any impact on the results. The findings of this investigation suggest that mood affects 1) the subjective experience of fear, 2) confidence in one's ability to tolerate a feared stimulus, and 3) whether an individual shows return of fear four weeks after having fear reduced. What this experiment cannot tell us is whether a happy mood has a positive impact on these variables, whether a sad mood has a negative impact, or whether the effects are due to some combination of these two possibilities. However, there is some evidence which suggests that a happy mood can have a positive effect on fear. This is apparent in the finding that subjects who had their fear reduced in a happy mood and were tested for return of fear in a naturally happy mood showed the least return of fear. There were no differences in the other three groups based on combinations of mood during fear reduction and mood during retest ing. Unfortunately, the effect of a sad mood could not be evaluated since very few subjects arrived at the lab in a naturally sad mood. Clearly, more research is needed to clarify the specific effects of happiness and sadness on fear. There are a number of practical implications for this research if the results prove to be generalizable to other fearful situations. For example, consider the use of music in dentists' offices. On the basis of the present study, one would predict that if fearful patients listen to pleasant music (with the instruction to think about events that make them feel happy), this would result in less subjective fear, greater confidence in one's ability to tolerate the dentist, and less return of fear on subsequent visits. In partial support of this prediction, Booth (1983) found that individuals fearful of the 80 dentist who listened to pleasant music during their visit reported the experience to be less aversive than those who did not listen to music. This study may also have implications for clinical depression and anxiety disorders. The first step is to show more clearly that a sad mood impedes fear reduction, rather than simply showing that it leads to differences in fear relative to a happy mood. If it is shown that a sad mood has negative effects on fear, the next step is to determine if such results generalize to clinical depression. If this is shown to be the case, it suggests that the most effective approach in treating individuals who are both clinically depressed and who suffer from anxiety disorders might be to treat the depression before exposing them to fearful situations or stimuli. Throughout this discussion, a number of limitations of this investigation have become apparent. First, the lack of a neutral control group limits the conclusions which can be drawn about the specific effects of happy and sad moods on fear. Second, therapists were not blind to mood conditions, although every attempt was made to ensure this. Fortunately, this had no impact on the results. It may be impossible for therapists to remain unaware of the mood of the individuals undergoing treatment. Third, it is unfortunate that therapist effects were confounded with feared stimuli. This makes it impossible to analyse the differences between therapists with respect to duration of fear reduction. Finally, it was not possible to obtain accurate measurements of the physiological components of fear. This has placed a great deal of emphasis on subjective reports in interpreting the effects of mood on fear. Future research in this area should focus on more clearly defining the specific effects of a happy mood and sad mood on fear. In addition, the psychophysiological components of fear need to be assessed with adequate 81 equipment. Finally, the generalization of the results of this study to other populations and settings remains an empirical question worthy of investigation. 82 References Bandura, A. (1982). Self-efficacy mechanism in human agency. American Psychologist ,37, 122-147. Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioural change. Psychological Review, 84, 191-215. Beck, AT. (1976). Cognitive Jherapy and the emotional disorders. New York: International Universities Press. Bernstein, D.A. and Paul, G.L. (1971). Some comments on therapy analogue research with small animarphobias". Journal of Behaviour Therapy and Experimental Psychiatry, 2,225-237. Blaney, P.H. (1986). Affect and memory: A review. Psychological Bulletin, 99, 229-246. Booth, R. (1983). Psychological approaches to dental anxiety. (Unpublished ri. Psych. Sc. Thesis). University College: Dublin. Bower, G.H. and Mayer, J.D. (1985). Failure to replicate mood-dependent retrieval. Bulletin of the Psychonomic Society, 23, 39-42. Bower, G.H., Gilligan, S.G. and Monteiro, K.P. (1981). Selectivity of learning caused by affective states. Journal of Experimental Psychology: General, 451 -473. Bower, G.H., Monteiro, K.P. and Gilligan, S.G. (1978). Emotional mood as a context for learning and recall. Journal of Verbal Learning and Verbal Behavior, 17, 573-585. Capstick, N. (1975). Clomipramine in the treatment of the true obsessional state: A report on four patients. Psychosomatics, 16, 21 -25. Clark, D.M. (1983). On the induction of depressed mood in the laboratory: Evaluation and comparison of the velten and musical procedures. Advances in Behaviour Research and Therapy, 5, 27-49. 83 Clark, D.M. and Teasdale, J.D. (1985). Constraints on the effects of mood on memory. Journal of Personality and Social Psychology, 48, 1595-1608. Craske, M. and Craig, K.D. (1984). Musical performance anxiety: The three-systems model and self-eff icacy theory. Behaviour Research & Therapy, 22, 267-280. Craske, M. and Rachman, S. (1985). Return of fear: perceived skill & heart rate responsivity. Unpublished doctoral dissertation. Department of Psycho 1 ogy, Un ivers i ty of Bri t i sh Co 1 umb i a. Davison, G.C. (1968). Systematic desensitization as a counter-conditioning process. Journal of Abnormal Psychology,73,91-99. DeMonbreun, B.G. and Craighead, W.E. (1977). Distortion of perception and recall of positive and neutral feedback in depression. Cognitive Therapy and Research, 1,311 -329. Eich, J.E. (1986). Context, memory and integrated item/context imagery. Journal of Experimental Psychology: Learning, Memory & Cognition, 11, 764-770. Eich, J. E. (1980). The cue-dependent nature of state-dependent retention. Memory & Cognition, 8, 157-173. Ekman, P., Levenson, R.W.and Friesen, W.V. (1983). Autonomic Nervous System Activity Distinguishes Among Emotions. Science, 221, 1208-1210. Foa, E. (1979). Failure in treating obsessive-compulsives. Behaviour Research & Therapy, 17, 169-176. Foa, E., Grayson, J.B. and Steketee, G.S. (1982). Depression, habituation and treatment outcome in obsessive-compulsives. In Practical applications of learning theories in psychiatry. (Edited by Boulougouris, J.C). New York: Wiley. Foa, E., Grayson, J.B., Steketee, G.S., Doppelt, H.G., Turner, R.M. and Latimer, P.R. (1983). Success and failure in the behavioral treatment of obsessive-compulsives. Journal of Consulting and Clinical Psychology, 51,287-297. 84 Gage, D.F. and Safer, MA (1985). Hemisphere differences in the mood state-dependent effect for recognition of emotional faces. Journal of Experimental Psychology: Learning, Memory, and Cognition,11, 752-763. Gatchel, R.J., Paulus, P.B. and Maples, CW. (1975). Learned helplessness and self-reported affect. Journal of Abnormal Psychology,84,732-734. Gotlib, I.H. (1981). Self-reinforcement and recall: differential deficits in depressed and non-depressed psychiatric inpatients. Journal of Abnormal Psychology, 90, 521 -530. Grey, S., Rachman, S. and Sartory, G. (1981). Return of fear: The role of inhibition. Behaviour Research & Therapy, 19, 135-143. Grey, S., Sartory, G. and Rachman, S. (1979). Synchronous and desynchronous changes during fear reduction. Behaviour Research & Therapy, 17, 137-147. Isen, AM., Shalker, T.E., Clark, M. and Karp, L. (1978). Affect, accessibility of material in memory, and behavior A cognitive loop? Journal of Personality and Social Psychology, 36, 1-12. Izard, C.E. (1971). The face of emotion. New York: Appleton-Century -Crofts. Kass, W. and Gilner, S.H. (1974). Drive level, incentive conditions and systematic desensitization. Behaviour Research and Therapy, 12, 99-106. Kazdin, A.E. (1978). Evaluating the generality of findings in analogue therapy research. Journal of consulting and clinical psychology, 46, 673-686. Krapfl, J.E. (1967). Differential ordering of stimulus presentation and semi-automated versus live treatment in the systematic desensitization of snake phobia. Unpublished doctoral dissertation, University of Missouri. Lang, P. (1968). Fear reduction and fear behavior. In Research in Psychotherapy (Ed. J. Shi ien). A.PA Washington. 85 Lang, P.J. and Lazovik, A.D. (1963). Experimental desensitization of a phobia. Journal of Abnormal and Social Psychology ,66,519-525. Leight, K.A. and Ellis, H.C. (1981). Emotional mood states, strategies, and state-dependency in memory. Journal of Verbal Learning and Verbal Behavior, 20, 251 -266. Likierman, H. and Rachman, S. (1980). Spontaneous decay of compulsive urges: Cumulative effects. Behaviour Research & Therapy, 18, 387-394 Loeb, A., Beck, A.T. and Diggory, J. (1971). Differential effects of success and failure on depressed and non-depressed patients. Journal of Nervous Mental Disease, 152, 106-114 Macht. Ml., Spear, N.E. and Levis, D.J. (1977). State-dependent retention in humans induced by alterations in affective state. Bulletin of the Psychonomic Society, 10,415-418. Marshall, W.L., Gauthier, J. and Gordon, A. (1979). The current status of flooding therapy. In Progress in Behavior Modification, Vol. 7 (Edited by Hersen, M.), pp. 205-275. Miller, W.R. (1975). Psychological deficits in depression. Psychological Bulletin, 82, 238-260. Miller, H.R. and Nawas, M.M. (1970). Control of aversive stimulus termination in systematic desensitization. Behaviour Research and Therapy, 8, 57-61. Nawas, M.M., Welsch, W.V. and Fishman, S.T. (1970). The comparative effectiveness of pairing aversive imagery with relaxation, neutral tasks and muscular tension in reducing snake phobia. Behaviour Research and Therapy, 6, 63-68. Philips, H.C. (1985). Return of fear in the treatment of a fear of vomiting. Behaviour Research & Therapy, 23, 45-52. Polivy, J. and Doyle, C. (1980). Laboratory induction of mood states through the reading of self-referent mood statements: Affective changes or demand characteristics. Journal of Abnormal Psychology, 89, 286-290. 8 6 Rachman, (\97d). Fear and courage. San Francisco: Freeman. Rachman, S. (1979). Emotional Processing. Behaviour Research & Therapy, 18,51-60. Rachman, S., Cobb, J. Grey, S., McDonald, B., Mawson, D., Sartory, G. and Stern, R. (1979). The behavioural treatment of obsessional-compulsive disorders, with and without clomipramine. Behaviour Research & Therapy, 17,467-478. Rachman, SJ. (1979). The return of fear. Behaviour Research & Therapy, 18,51-60. Rachman, S.J. and Levitt, K. (1985). Panics and their consequences. Behaviour Research & Therapy 23,585-600. Rachman, S.J. and Lopatka, C. (1987). Return of fear: underlearning and overlearning. Behavior Research & Therapy, (accepted for publication). Rachman, SJ., Robinson, S., and Lopatka, C. (1987). Is incomplete fear-reduction followed by a return of fear? Behavior Research & Therapy, (accepted for publication). Richardson, F.C. and Suinn, R.M. (1973). A comparison of traditional systematic desensitization, accelerated massed desensitization, and anxiety management training in the treatment of mathematics anxiety. Behavior Therapy, A, 212-218. Sartory, G., Rachman, S., Grey, S. (1977). An investigation of the relation between reported fear and heart rate. Behaviour Research & Therapy, 15,435-438. Sartory, G., Rachman, S. and Grey, S. (1982). Return of fear: The role of rehearsal. Behaviour Research & Therapy, 20, 123-133. Schare, Ml., Lisman, S.A and Spear, N.E. (1984). The effects of mood variation on state-dependent retention. Cognitive Therapy and Research, 8, 387-408. 87 Schubot, E.D. (1966). The Influence of hypnotic and muscular relaxation in systematic desensitization of phobic behavior. Unpublished doctoral dissertation, Stanford University, 1966. Schwartz, G.E., Weinberger, DA, Phil, M. and Singer, J A (1981). Cardiovascular differentiation of happiness, sadness, anger, and fear following imagery and exercise. Psychosomatic Medicine, 43, 343-363. de Silva P. and Rachman S. (1984). Does escape behaviour strengthen agoraphobic avoidance? A preliminary study. Behaviour Research & Therapy, 22, 87-93. Skinner, B.F. (1938). The behavior of organisms. Appleton-Century, New York. Sue, D. The effect of duration of exposure on systematic desensitization and extinction. Behaviour Research and Therapy, 13,55-60. Sutherland, 6., Newman, B. and Rachman, S. (1982). Experimental investigations of the relations between mood and intrusive unwanted cognitions. British Journal of Medical Psychology ,55, 127-138. Teasdale, J.D., Fogarty, S.J. and Williams, J.M.G. (1980). Speech rate as a measure of short-term variation in depression. British Journal of Social and Clinical Psychology, 19,271-278. Teasdale, J.D. and Spencer, P. (1982). Effects of induced elation-depression on estimates of past success and subjective probability of future success. Unpublished manuscript, Department of Psychiatry, University of Oxford. Velten, E. (1968). A laboratory task for induction of mood states. Behaviour Research and Therapy, 6, 473-482. Watts, F.N. (1973). Desensitization as an habituation phenomenon: Studies of interstimulus interval lengths. Psychological Reports, 33, 715-718. Wetzler, S. (1985). Mood state-dependent retrieval: A failure to replicate. Psychological Reports, 56, 759-765. 88 Weingartner, H, Miller, H. and Murphy, D.L. (1977). Mood-state-dependent retrieval of verbal associations. Journal of Abnormal Psychology,86, 276-284 Weingartner, K, Adefris, W., Eich, J.E. and Murphy, D.L. (1976). Encoding-imagery specificity in alcohol state-dependent learning. Journal of Experimental Psychology: Human Learning andrfemeory,2,QZ-Q7. Williams, J.M.6. (1980). Generalization in the effects of a mood induction procedure. Behaviour Research and Therapy, 18,565-572. Wolpe, J. (1958). Psychotherapy by reciprocal inhibition. Stanford, Calif: Stanford University Press. 89 Appendix A Therapist Rating Scale for Subject's Mood 90 Mood Scale (To Be Completed By Therapist) Subject No Session No Therapist: Instructions: Please place a slash (/) anywhere along the continuum you feel is appropriate. How sad do you feel this subject was during fear reduction? 0 100 NOT AT SAD EXTREMELY SAD How happy do you feel this subject was during fear reduction? 0 100 NOT AT ALL HAPPY EXTREMELY HAPPY Appendix B Therapist Evaluation Questionnaire 92 Therapist Evaluation Questionnaire Please rate the therapist on the following nine items: 1) Provision of information about the feared stimulus POOR 100 EXCELLENT 2) Pacing of exposure POOR 100 EXCELLENT 3) Use of distraction POOR 100 EXCELLENT 4) Use of positive reinforcement 100 POOR EXCELLENT 5) El ici tat ion of fear ratings from subject POOR 100 EXCELLENT 6) Maintenance of a neutral mood 0 100 POOR EXCELLENT 93 7) Obtaining cooperation of subject POOR 100 H EXCELLENT 8) Use of modelling techniques POOR 100 EXCELLENT 9) Overall ability to reduce subject's fear 100 POOR EXCELLENT Appendix C Fear Survey Schedule 96 Fear Survey Schedule Please check (V) the appropriate level of fear for each of the following Items: ITEM 1 AM NOT AT ALL FEARFUL OF: 1 AM SLIGHTLY FEARFUL OF: 1 AM MODERATELY FEARFUL OF: 1 AM EXTREMELY FEARFUL OF: 1 AM TERRIFIED OF: SNAKES CATS BIRDS SPIDERS WORMS DOGS INSECTS HORSES THE DARK HEIGHTS CLOSED SPACES Name: Phone Number: Best Time to Call: 97 Appendix D Consent Form 99 I have read the attached information, consent to participate in this research, and have received a copy of this consent form. Signature: ; : Name: Date: Student I.D.*: Appendix E Mood Questionnaire 101 Mood Scale Instructions: Please place a slash (/) anywhere along the continuum you feel is appropriate. How sad do you feel? 0 I AM NOT AT ALL SAD 100 I AM EXTREMELY SAD How tired do you feel? 0 I I AM NOT AT ALL TIRED 1 0 0 AM EXTREMELY TIRED How happy do you feel? 0 I AM NOT AT ALL HAPPY 100 AM EXTREMELY HAPPY How anxious do you feel? 0 I I AM NOT AT ALL ANXIOUS 100 H AM EXTREMELY ANXIOUS How despondent do you feel? 0 100 I AM NOT AT ALL DESPONDENT I AM EXTREMELY DESPONDENT How apprehensive do you feel? 0 100 \ 1 I AM NOT AT I AM EXTREMELY ALL APPREHENSIVE APPREHENSIVE Appendix F Self-Eff icacy Questionnaire 103 S-E Qestionnaire Instructions: Please place a slash (/) anywhere along the continuum you feel is appropriate. How confident are you that you will be able to tolerate a live but harmless spider/snake you with minimal fear or discomfort? 0 100 NOT AT ALL TOTALLY CONFIDENT CONFIDENT Appendix 6 Manipulation Check M-C Questionnaire Please answer the following question honestly: Do you think your mood changed when you listened to the music during today's experiment? • • YES NO Appendix H Tables of Means and Standard Deviations 107 Means and Standard Deviations for Reported Happiness Pre and Post Mood Induction by Group Group Variable Happy Sad M SD U SD SJfl. Pre Mood Induction 63.26 (18.09) 58.07 (17.64) t(82)=1.33 Post Mood Induction: Time 1 81.09 (6.07) 16.93 (5.01) t(82)=27.46* Time 2 (8 min) 68.90 (14.99) 41.22 (19.96) U81K7.13* Time 3 (16 min) 70.94 (18.14) 47.82 (21.92) t(69)=4.87* Time 4 (24 min) 71.12 (12.44) 53.62 (21.46) t(36)=3.14* Time 5 (32 min) 69.88 (9.08) ' 51.25 (13.22) t(14)=3.28* *B<.01 Means and Standard Deviations for Reported Sadness Pre and Post Mood Induction by Group Group Variable Happy Sad tl SD tl 5X2 Pre Mood Induction 14.26 (16.35) 18.27 (21.37) t(82)=0.96 Post Mood Induction: Time 1 9.34 (8.30) 78.63 (3.02) t(82)=26.47* Time 2 (8 min) 11.03 (12.24) 41.12 (19.80) t(8!)=8.31* Time 3(16 min) 9.85 (10.92) 30.26 (20.31) t(69)=5.37* Time 4 (24 min) 10.65 (11.16) 28.25 (18.13) t(36)=3.67* Time 5 (32 min) 12.38 (9.29) 31.75 (17.86) t(14)=2.96* *C<.01 Means and Standard Deviations for Return of Fear (Group x Natural Mood) 109 Group Happy Sad Natural Mood M SQ h SQ (session 2) Not Happy 28.65 (0=17) (25.16) 3.53 (n=15) (13.95) Happy 26.47 (Q=15) (24.18) 25.19 (n=16) (21.68) 110 Means and Standard Deviations for Savings Scores Savings score (minutes) Groupings LI SD Congruent (0=16) 12.26 (8.31) Incongruent (n=l3) 6.41 (11.38) Happy (session 1) (n-13) 7.64 (10.18) Sad (session 1) (n=l6) 11.27 (9.99) Happy (session 2) (n=12) 11.58 (12.16) Sad (session 2) (n=17) 8.28 (8.42) Condition 1 (n=6) 12.87 (10.72) Condition 2 (n=6) 3.11 (7.68) Condition 3 (n=7) 10.30 (14.48) Condition 4 (n=10) 11.92 (7.07) Means and Standard Deviations for Duration of Fear Reduction (Group x i n Group Happy Sad tl SD LI SD Therapist 1 10.47 (5.99) 18.40 (10.63) <D=15) (n-17) Therapist 2 22.08 (6.24) 22.09 (9.73) (n=17) (0=14) Means and Standard Deviations for Return of Fear (Group x Therapist) Group Happy Sad tl SD tl SD Therapist 1 10.13 (21.57) 23.53 (22.68) (0=15) (0=17) Therapist 2 22.82 (25.10) 28.57 (22.91) (0=17) (0=14) Means and Standard Deviations for Happiness Ratings (Group x Therapist) Group Happy Sad tl SD tl SD Therapist 1 52.89 (22.31) 31.78 (16.04) (11=20) (n=20) Therapist 2 60.82 (15.50) 49.62 (13.70) (0=23) (Q=21) Means and Standard Deviations for Sadness Ratings (Group x Therapist) Group Happy Sad tl SD tl SD Therapist 1 53.06 (20.89) 61.06 (19.16) (0=20) (0=20) Therapist 2 30.27 (12.25) 48.29 (17.48) (0=23) (0=21) Appendix I Univariate Analyses Summary Tables 114 Univariate Analysis of Variance Results for Groups (Happy and Sad) Source df rJS E Subjective Fear 1 2,706.97 7.93** Error 65 341.57 Self-Efficacy 1 3,112.96 5.32* Error 65 585.01 Duration of Fear 1 27,646.38 1.91 Reduction Error 65 14,482.51 *U<01 **C<.05 Univariate Analysis of Variance Results for Return/No Return of Fear Groups 115 Source di US Duration of 1 57,976.25 5.33* Fear Reduction Error 57 10,877.89 Self-Efficacy 1 338.44 0.52 (begin, of session I) Error 57 645.29 Self-Efficacy 1 6,753.70 17.82** (begin, of session 2) Error 57 379.05 Natural Mood: Happy 1 2,894.44 10.20** (begin, of session 1) Error 57 283.70 Natural Mood: Sad 1 784.57 2.24 (begin, of session 1) Error 57 349.57 Induced Mood: Happy 1 6,312.10 5.72* (session 1) Error 57 1,103.34 116 Univariate Analysis of Variance Results for Return/No Return of Fear groups (continued) Source di fclS Induced Mood: Sad 1 6,380.31 492* (session 1) Error 57 1,296.60 Natural Mood: Happy 1 3,21438 8.72** (begin, of session 2) Error 57 368.74 Natural Mood: Sad 1 853.52 2.33 (begin, of session 2) Error 57 366.92 *p<.05 **p_<.01 117 Univariate Analysis of Variance Results for Successful/Partially Successful! Groups Source. di LIS £ Self-Efficacy 1 163.97 0.28 (begin, of session 1) Error 71 578.94 Self-Efficacy 1 6,990.16 23.72** (end of session 1) Error 71 294.76 1,761.35 3.99* 441.22 Return of Fear 1 1,255.63 5.48* Error 71 229.13 Self-Efficacy 1 (begin, of session 2) Error 71 *U<05 **B<.01 Appendix J Analyses of Variance Summary Tables Analysis of Variance Results of Suds Ratings for Groups Pre and Post Mood Induction Source df m E Groups 1 1,155.17 5.28* Error 82 218.91 Pre/Post Induction 1 10,535.53 59.93** Group x Pre/Post Induction 1 1,567.55 8.92** Error 82 175.79 Analysis of Variance Results of Self-Efficacy Ratings for Groups Pre and Post Mood Induction Source df M5_ E Groups 1 2,422.20 2.46 Error 79 984.11 Pre/Post Induction 1 877.58 3.52 Groups x Pre/Post Induction 1 988.11 3.97* Error 79 248.77 *e<.05 **p_<.oi 120 Analysis of Variance Results of Savings Scores for Conditions (1 -4) Source di MS E Conditions 3 13,765.50 1.42 Within Groups 25 9,694.34 Analysis of Variance Results of Return of Fear for Natural Mood and Induced Mood Source dl MS E Induced Mood 1 2,031.92 4.27 Natural Mood 1 2,807.76 5.90* Induced Mood x Natural Mood 1 2,230.79 4.69* Error 59 475.93 Analysis of Variance Results of Return of Fear for Group and Therapist Source di MS E Group 1 1,454.96 2.71 Therapist 1 1,249.75 2.33 Group x Therapist 1 228.73 0.43 Error 59 536.27 *0<.05 **u<.01 Analysis of Variance Results of Duration for Group and Therapist Source di LIS E Group 1 28,898.89 2.85 Therapist 1 102,585.83 10.03** Group x Therapist 1 26,126.94 2.55 Error 63 10,228.00 Analysis of Variance Results of Happiness Ratings for Group and Therapist Source di MS F Group 1 2,261.86 7.84** Therapist 1 6,626.72 22.98** Group x Therapist 1 5.80 0.02 Error 75 288.39 Analysis of Variance Results of Sadness Ratings for Group and Therapist Source ns E Group 1 3,571.17 11.69** Therapist 1 6,228.16 20.38** Group x Therapist 1 491.03 1.61 Error 75 305.55 *p<.05 **p<.01 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0097072/manifest

Comment

Related Items