UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The three-systems model and self efficacy theory : piano performance anxiety Craske, Michelle Genevieve 1982

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

Item Metadata

Download

Media
831-UBC_1983_A8 C73.pdf [ 8.83MB ]
Metadata
JSON: 831-1.0095688.json
JSON-LD: 831-1.0095688-ld.json
RDF/XML (Pretty): 831-1.0095688-rdf.xml
RDF/JSON: 831-1.0095688-rdf.json
Turtle: 831-1.0095688-turtle.txt
N-Triples: 831-1.0095688-rdf-ntriples.txt
Original Record: 831-1.0095688-source.json
Full Text
831-1.0095688-fulltext.txt
Citation
831-1.0095688.ris

Full Text

THE THREE-SYSTEMS MODEL AND SELF EFFICACY THEORY: PIANO PERFORMANCE ANXIETY BY MICHELLE GENEVIEVE CRASKE B.A., University of Tasmania, 1980 B.A.(HONS), University of Tasmania, 1981 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS THE FACULTY OF GRADUATE STUDIES (Department of Psychology) We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA December, 1982 (c) Michelle Genevieve Craske, 1982 i n In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head o f my department or by h i s or her r e p r e s e n t a t i v e s . I t i s understood t h a t copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be allowed without my w r i t t e n p e r m i s s i o n . Michelle G. Craske Department of Psychology The U n i v e r s i t y of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date December 20, 1982 DE-6 (3/81) ABSTRACT This study examined contrasting p r e d i c i t i o n s from Self E f f i c a c y theory and the Three-Systems model of fear and anxiety, i n the context of musical performance anxiety. Further experimental evidence was sought for Hodgson and Rachmans' (1974) hypotheses derived from the three-systems model, and for Bandura's predicted r e l a t i o n s h i p s between the construct of s e l f e f f i c a c y and behavioural, p h y s i o l o g i c a l and verbal response systems. P i a n i s t s , who rated themselves as either ' r e l a t i v e l y anxious' or ' r e l a t i v e l y nonanxious' solo performers, were asked to play a musical piece under two conditions. F i r s t , the p i a n i s t s played alone under non-evaluative conditions. Second, they played before an audience and videocamera under evaluative conditions. Measures were taken of each response system during both performance conditions; behavioural measures included performance q u a l i t y and a timed c h e c k l i s t of observable signs of anxiety; s e l f report measures included the State Anxiety scale, SUD scales, S e l f Statement scales and several s e l f e f f i c a c y scales; and autonomic measures, recorded continuously v i a telemtry, included heart rate, r e s p i r a t i o n and skin conductance. The audience condition was found to e l i c i t more intense emotional responses i n r e l a t i v e l y anxious p i a n i s t s ; t h i s condition e l i c i t e d l e s s intense emotional responses i n r e l a t i v e l y nonanxious p i a n i s t s . These group e f f e c t s were enhanced when extreme scorers were analysed. In the audience condition, the r e l a t i v e l y anxious group exhibited increased l e v e l s of anxiety i n each of the response systems ( i . e . synchrony), while desynchrony was observed i n the r e l a t i v e l y nonanxious group. Correlations among the dependent measures were generally weak, but concordance was most often observed when intense emotional responses were e l i c i t e d . The r e s u l t s obtained l e n t support to the three-systems model. In contrast to Bandura's p r e d i c t i o n s , s e l f e f f i c a c y did not c o r r e l a t e with l e v e l s of anxiety i n each response system. Issues raised by the study included the importance of multiple-system measurement, treatment implications, contrasts between 'analogue' and ' c l i n i c a l ' populations, and the complexity of the phenomenon of anxiety. i v TABLE OF CONTENTS Page ABSTRACT i i TABLE OF CONTENTS i v LIST OF TABLES v i i LIST OF FIGURES v i i i LIST OF APPENDICES ix ACKNOWLEDGEMENTS x INTRODUCTION 1 LITERATURE REVIEW 5 Theoretical Accounts of Fear 5 Behavioural Conceptualisations 5 C l a s s i c a l Conditioning 5 Two-Stage Theory 5 Incubation Model 7 Cognitive-Behavioural Conceptualisations 9 Se l f E f f i c a c y Theory 10 Cr i t i c i s m s of Se l f E f f i c a c y Theory 14 The Nature of Fear 16 Formulation of the Three-Systems Model 16 The Three Systems of Fear 17 Interdependency of the Three Systems 19 Synchrony and Desynchrony 20 Theoretical Implications 21 Supporting Evidence 22 Measurement Issues 25 Verbal Response System 25 Autonomic Response System 26 Behavioural Response System 27 Treatment Issues 31 Response Treatment Matching 31 Response System Generalization i n Reaction to Treatment 34 STATEMENT OF THE PROBLEM 38 Se l f E f f i c a c y Versus The Three-Systems Model 38 The o r e t i c a l Contrast 38 Treatment Choice Contrast 42 Experimental Evidence 43 Se l f E f f i c a c y as an Hypothetical Construct 43 Purpose of the Study 45 Experimental Hypotheses 46 METHOD 48 Subjects 48 Experimental Setting 49 Self-Report Measures 49 Behavioural Measures 52 Autonomic Measures 53 Apparatus and Equipment 55 Assessment Conditions 57 Procedure 58 Selection of Subjects 58 Alone Condition 59 Audience Condition 61 Debriefing 61 Design and S t a t i s t i c a l Analyses 61 RESULTS 63 Summary of Findings 63 Preliminary Variables 66 V a l i d i t y of the Screening Index 69 Interrater R e l i a b i l i t i e s 69 'Vigour', 'Video* and 'Respiration' Design Control Measures 72 Group Means Analyses 72 Extreme Group Means Analyses 82 Correlations Among Dependent Measures 96 Se l f E f f i c a c y 107 DISCUSSION 117 Experimental Manipulations 118 Three Response Systems 124 Se l f E f f i c a c y 136 Conclusion 140 REFERENCES vi 143 APPENDICES 153 v i i LIST OF TABLES Table I Means, Standard Deviations and F Values for Males and Females on Each Dependent Variable Table II Interrater R e l i a b i l i t i e s Assessed by Pearson Correlations Table III Phi-Correlations for the Timed Che c k l i s t of Observable Signs of Anxiety Table IV Pearson Correlations Between Vigour and Both Respiration and Heart Rate Table V Group Frequencies of Reported Awareness, Suspicion and Unawareness of the Camera Table VI Pearson Product Moment C o e f f i c i e n t s Between Respiration Rate and Heart Rate Table VII Means and Standard Deviations for Each Dependent Measure Table XV Means and Standard Deviations of S e l f E f f i c a c y Measures Table XVI Table XVII Means and Standard Deviations for Item 2 of the Self E f f i c a c y Scale Pearson Correlations Between S e l f E f f i c a c y and Dependent Variables Page 67 70 71 73 74 75 77 Table VIII Means and Standard Deviations for Each Dependent Measure for the Most Anxious and Least Anxious Groups 83 Table IX Pearson Correlations Between Behavioural Measures 98 Table X Pearson Correlations Between Self-Report Measures 99 Table XI Pearson Correlations Between Autonomic Measures 101 Table XII Pearson Correlations Between Behavioural and Self-Report Measures 102 Table XIII Pearson Correlations Between Behavioural and Autonomic Measures 104 Table XIV Pearson Correlations Between Self-Report and Autonomic Measures 106 109 111 113 Table XVIII Pearson Correlations Between S e l f E f f i c a c y Strength at Three Levels and Heart Rate 115 v i i i LIST OF FIGURES Page Figure 1 Mean Performance Quality for R e l a t i v e l y Anxious, R e l a t i v e l y Nonanxious, Most Anxious and Least Anxious P i a n i s t s 86 Figure 2 Mean Timed Che c k l i s t Scores for R e l a t i v e l y Anxious, R e l a t i v e l y Nonanxious, Most Anxious and Least Anxious P i a n i s t s 87 Figure 3 Mean Self E f f i c a c y Strength for R e l a t i v e l y Anxious, R e l a t i v e l y Nonanxious, Most Anxious and Least Anxious P i a n i s t s 89 Figure 4 Mean Tot a l SUDS for R e l a t i v e l y Anxious, R e l a t i v e l y Nonanxious, Most Anxious and Least Anxious P i a n i s t s 90 Figure 5 Mean State Anxiety for R e l a t i v e l y Anxious, R e l a t i v e l y Nonanxious, Most Anxious and Least Anxious P i a n i s t s 91 Figure 6 Mean Number of Negative S e l f Statements for Re l a t i v e l y Anxious, R e l a t i v e l y Nonanxious, Most Anxious and Least Anxious P i a n i s t s 93 Figure 7 Mean Heart Rate (Range Corrected) Prior to Playing for R e l a t i v e l y Anxious, R e l a t i v e l y Nonanxious, Most Anxious and Least Anxious P i a n i s t s 94 Figure 8 Mean Heart Rate (Range Corrected) During Performances for R e l a t i v e l y Anxious, R e l a t i v e l y Nonanxious, Most Anxious and Least Anxious P i a n i s t s 95 LIST OF APPENDICES Page Appendix 1 I n i t i a l Contact 153 Appendix 2 Report of Confidence as a Performer 154 Appendix 3 Subject Data 156 Appendix 4 General T r a i t Anxiousness Scale for P i a n i s t s 157 Appendix 5A State Scale (State-Trait Anxiety Inventory) 158 Appendix 5B SUD Scale 159 Appendix 5C Retrospective SUD Scale 160 Appendix 5D Expectations of Personal E f f i c a c y Scale 161 Appendix 5E Self E f f i c a c y Scale (Alone Condition) 163 Appendix 5F Self E f f i c a c y Scale (Audience Condition) 164 Appendix 5G Performance Anxiety Self Statement Scale (Alone Condition) 165 Appendix 5H Performance Anxiety S e l f Statement Scale (Audience Condition) 168 Appendix 6 D e f i n i t i o n s of Performance Quality Dimensions 171 Appendix 7 Performance Quality Rating Sheet 172 Appendix 8 Timed Behavioural C h e c k l i s t (Rating Sheet and D e f i n i t i o n of Terms) 173 Appendix 9 Audience Che c k l i s t ' 174 Appendix 10 Phone Contact Instructions 175 Appendix 11 Subject Data Sheet 176 Appendix 12 Performance Schedule 177 Appendix 13 F i r s t Set of Subject Instructions - Alone Condition 178 Appendix 14 Consent Form 1 179 Appendix 15 Debriefing Instructions i n Alone Condition 180 Appendix 16 Consent Form 2 181 Appendix 17 F i r s t Set of Subject Instructions - Audience Condition 182 Appendix 18 Video Camera Questionnaire 183 Appendix 19 Second Set of Subject Instructions - Audience Condition 184 Appendix 20 Debriefing Letter 185 Appendix 21 Absolute Values of Heart Rate and Skin Conductance Level 188 Appendix 22 Summary of Repeated Measures Univariate Analyses of Variance 189 Appendix 23 Summary of Group x Condition x Time of Measurement Analyses of Variance 192 Appendix 24 Summary of S i g n i f i c a n t Univariate Analyses of Variance for Most Anxious and Least Anxious P i a n i s t s 194 X ACKNOWLEDGEMENTS I would l i k e to thank my supervisor, Professor K.D. Craig, whose guidance and help throughout t h i s work have been greatly appreciated. I am also indebted to Professor D. Reubart and Dr. M. Kendrick for th e i r h e l p f u l suggestions concerning musical performance. My thanks are also extended to Dr. W. Iacono and Michael Kozak for t h e i r i n s t r u c t i o n i n psychophysiology and to Dr. Peter Graystone for h i s consultation i n the use of telemetry. I also want to thank Dr. J . Steiger and Marsha Schroeder for the i r s t a t i s t i c a l supervision. F i n a l l y , I would l i k e to thank my family for t h e i r "long distance" encouragement, and my fellow students for t h e i r support. I am e s p e c i a l l y g r a t e f u l to Charl i e for h i s u n f a i l i n g help, advice and understanding. 1 INTRODUCTION The behavioural conceptualisation of fear and anxiety has recently undergone two separate developments. F i r s t , a multicomponent theory, which challenges the unitary model of fear, has emerged. Second, cognitive v a r i a b l e s have been assigned causal status within behavioural frameworks. However, the e s s e n t i a l features of each of these developments are disparate. Gradual accumulation of empirical evidence contradicting features of Mowrer's two-factor theory, and related unitary models of fear, led Lang (1968) to suggest a new model which described fear reactions as comprising three loosely coupled components, or, response systems. This "three-systems" analysis was further elaborated by Rachman and Hodgson (1974). The model proposed that the three major responses systems - behavioural, autonomic and s e l f report/verbal - are i n t e r a c t i v e , yet also p a r t i a l l y independent. As such, the systems are able to respond with d i f f e r e n t i n t e n s i t i e s at any given time ( i . e . discordance) and with d i f f e r e n t rates of change over time ( i . e . desynchrony) (Lang, 1971; Rachman and Hodgson, 1974). The three-systems model has been widely accepted and i s supported by considerable empirical evidence. The model i s able to explain observed i n d i v i d u a l d i f f e r e n c e s i n fear reactions, and has important treatment imp-l i c a t i o n s . In addition, new concepts have been generated by t h i s model, such as the 'pathways of fear a c q u i s i t i o n ' (Rachman, 1976), c l i e n t by treatment matching, and the 'return of f e a r ' (Rachman, 1979a). Bandura's S e l f E f f i c a c y theory (1977) remains the most s i g n i f i c a n t cognitive-behavioural conceptualisation of fear and anxiety. Bandura proposed a s i n g l e superordinate cognitive mechanism to account for changes i n each response system. That i s , conviction i n one's a b i l i t y to perform p a r t i c u l a r 2 tasks i s believed to mediate subjective, autonomic and behavioural anxiety. S e l f E f f i c a c y theory o f f e r s a parsimonious i n t e r p r e t a t i o n of fear enhancement and fear reduction, regardless of s p e c i f i c treatment s t r a t e g i e s . In the search for a s i n g l e underlying factor that accounts for the effectiveness of the various fear reduction techniques, Rachman (1980) proposed the notion of 'emotional processing' while Lang (1980) emphasised the role of 'emotional imagery'; both incorporate a three-systems concept-u a l i s a t i o n of fear. In contrast, Bandura asserted that s e l f e f f i c a c y i s the active mediating agent i n a l l response system changes; s e l f e f f i c a c y theory i s p r i m a r i l y based on a unitary model of fear. A three-systems approach allows considerable v a r i a t i o n in response patterns, and states conditions under which concordance and/or synchrony are more l i k e l y to occur (Hodgson and Rachman, 1974). On the other hand, s e l f ' e f f i c a c y theory s i m p l i f i e s the fear reaction be de-emphasising such v a r i a t i o n . In recognising the merits of both approaches, Wilson (1978) suggested that they be combined: Hodgson and Rachman (1974) have thoughtfully set out the various c l i n i c a l and experimental observations that have been reported. However, an e x p l i c i t theory to explain the findings i s s t i l l needed. S e l f e f f i c a c y theory i s one a l t e r n a t i v e . Instead of ordering the desynchrony data in terms of the e f f e c t s of d i f f e r e n t techniques, a f i n e r grained analysis may generate predictions on the basis of the strength and generality of the e f f i c a c y expectations induced by the d i f f e r e n t methods. Inevitably, i t w i l l require a sophisticated and complex analysis of t h i s sort in accounting for highly variable patterns of behaviour that are observed i n therapeutic outcomes. (p. 225) However, Wilson appears to have overlooked the f a c t that s e l f e f f i c a c y theory does not contain elements that account for the independence of the response systems. Indeed, as a r e s u l t of their d i f f e r i n g fundamental suppositions, many c r i t i c a l p redictions of the three-systems model and of s e l f e f f i c a c y theory are contradictory. Although t h i s c o n f l i c t has been c i t e d in the l i t e r a t u r e (Borkovec, 1978; Wolpe, 1978), l i t t l e research has been 3 conducted i n order to e x p l i c i t l y test the c o n f l i c t i n g p r e d i c t i o n s . Ban-dura's research i n t h i s area has been l a r g e l y r e s t r i c t e d to analogue populations (snake and spider phobics within college students). However, assessments of analogue fears are susceptible to extraneous s i t u a t i o n a l influences that l i m i t g e n e r a l i z a b i l i t y (Borkovec, 1976). As an a l t e r n a t -ive, performance anxiety i s e s p e c i a l l y suitable for t h i s type of research as i t can be severely d e b i l i t a t i n g , and groups representing d i f f e r e n t l e v e l s of anxiety are d i f f e r e n t i a b l e on autonomic, behavioural and verbal measures (Galassi, F r i e r s o n and Sharer; Holroyd, Westbrook, Kirkland, Jones and Van Norman, 1979; Sherry and Levine, 1980). S p e c i f i c a l l y , musical performance anxiety, l i k e speech anxiety (Bork-ovec and O'Brien, 1976) i s of a pervasive nature and often p e r s i s t s despite repeated exposures. Musical performance anxiety generally e n t a i l s worry about performance, preoccupation with f e e l i n g s of inadequacy, and heightened somatic and autonomic arousal (Kendrick, Craig, Lawson and Davidson, 1982). Kendrick (1979) i d e n t i f i e d several important d i s t i n g u i s h -ing c h a r a c t e r i s t i c s of musical performance anxiety. F i r s t , unlike many other performance test s i t u a t i o n s , temporary lapses of concentration are very detrimental and have immediate e f f e c t s on performance. Also, errors are more conspicuous i n a well-known set piece of music as opposed to speech performance where "ad l i b b i n g " i s pos s i b l e . Moreover, the complexity of the motor s k i l l s involved renders musical performance very vulnerable to the e f f e c t s of autonomic arousal. S e l f e f f i c a c y theory predicts that high l e v e l s of perceived s e l f e f f i c a c y produce performance mastery, and l i t t l e , i f any, autonomic arousal or subjective fear. The converse i s predicted for preceived i n e f f i c a c i o u s n e s s . Therefore, Bandura implies that, i n general, the three response systems are l i k e l y to be concordant at any given time in d i r e c t correspondence to the l e v e l and strength of s e l f e f f i c a c y , and that changes in s e l f e f f i c a c y w i l l be accompanied by synchronous changes in the response systems. In contrast, the three-systems model predicts that the degree of concordance depends upon the i n t e n s i t y of the emotional response and demand features of the s i t u a t i o n . Also, the systems are believed to respond d i f f e r e n t i a l l y to contrasting environmental manipulations (DeSilva and Rachman, 1981). More s p e c i f i c a l l y , under conditions which evoke strong emotional responses, where, according to the three-systems model, f e a r f u l i n d i v i d u a l s are most l i k e l y to respond with comparably high l e v e l s of anxiety in each response system ( i . e . concordance), the presence of low s e l f e f f i c a c y would most l i k e l y c o r r e l a t e with each response system. As such,' the causal role of either s e l f e f f i c a c y and/or the i n t e n s i t y of the emotional response could not be determined. However, Bandura's theory would not be supported when s e l f e f f i c a c y i s high and the response systems concordantly indicate fear. Under conditions that do not evoke strong emotional responses f e a r f u l i n d i v i d u a l s are most l i k e l y to respond with discordance according to the three-systems model. In d i r e c t contrast, Bandura suggests that under such conditions, s e l f e f f i c a c y would be higher, and accompanied by compar-ably low l e v e l s of anxiety in each response system ( i . e . concordance). In t h i s study the d i f f e r e n t predictions of the three-systems model and s e l f e f f i c a c y theory have been examined in the context of musical performance anxiety in p i a n i s t s . 5 LITERATURE REVIEW THEORETICAL ACCOUNTS OF FEAR The i n v e s t i g a t i o n of fear reactions over the past f o r t y years has resulted i n s u b s t a n t i a l t h e o r e t i c a l developments from early traumatic conditioning models. Those e a r l i e r models, unable to account for many c h a r a c t e r i s t i c s of fear and anxiety, have been replaced by more elaborate cognitive-behavioural conceptualisations that acknowledge the complexity of fear and mediational processes. By r e j e c t i n g the notion that fear i s a unitary response system, and by recognising the role of cognitive processes as important causal mediators of behaviour, recent theories have proven more adequate. This chapter describes the major behavioural models and their weaknesses and the newly evolved cognitive-behavioural conceptualisations exemplified by Bandura's (1977) Self E f f i c a c y theory. Behavioural Conceptualisations C l a s s i c a l Conditioning In the e a r l y 1920's, Watson and Rayner postulated that neutral s t i m u l i (or, conditioned s t i m u l i (CS)) associated with i n t r i n s -i c a l l y p a i n f u l or frightening events (or, unconditioned s t i m u l i (UCS)), developed f e a r f u l q u a l i t i e s ; the strength of the fear (or, conditioned response (CR)) being determined by the number of r e p e t i t i o n s of the assoc-i a t i o n , and the i n t e n s i t y of the aversive experience. Also, s t i m u l i resembling the fear evoking stimulus were believed to acquire fear evoking properties (Rachman, 1978). However, t h i s model has been unable to account for the often observed strong resistance of fears to e x t i n c t i o n . Two-Stage Theory Mowrer (1939) l a t e r developed the Two-Stage theory in an attempt to explain the persistence of fear during e x t i n c t i o n t r i a l s . An operant/instrumental conditioning stage was added to the o r i g i n a l c l a s s i c a l 6 conditioning model. According to Mowrer, the a s s o c i a t i o n of neutral s t i m u l i with aversive events creates an anxiety drive that motivates defensive (avoidance/escape) behaviour. Such defensive behaviour i s then reinforced by reducing anxiety arousal. Hence, CS are c o n t i n u a l l y reinforced even when UCS are absent, and, therefore, CRs p e r s i s t (Bandura, 1977; C o s t e l l o , 1970; Rachman, 1976). Various sources of evidence for Mowrer's theory included; induction of fear in animals v i a conditioning, development of anxiety states i n s o l d i e r s , induction of fear i n c h i l d r e n , i n c i d e n t a l findings with aversion therapy, and the e f f e c t s of traumatic stimulation (Rachman, 1978). However, research findings c i t e d as support for the two-stage theory are of questionnable v a l i d i t y and recently strong arguments against the theory i t s e l f have developed. These include the f a i l u r e of people to acquire fear in n a t u r a l l y occurring, t h e o r e t i c a l l y fear provoking s i t u a t i o n s , and the d i f f i c u l t y encountered in attempting to a r t i f i c i a l l y produce fear conditioned responses in people (Rachman, 1978). It i s now believed that fear may not develop in reaction to traumatic events unless a c o r r e l a t i o n a l or contingent r e l a t -ionship i s formed between the s i t u a t i o n and the traumatic event. That i s , c l a s s i c a l conditioning i s no longer believed to simply involve CS and UCS pairings on the basis of temporal c o n t i g u i t y alone (Rachlin, 1976). Also, the two-stage theory rests upon the e q u i p o t e n t i a l i t y premise, one which i s now untenable in view of the apparent nonrandom d i s t r i b u t i o n of fear objects. Seligman's preparedness theory suggests that humans are g e n e t i c a l l y prepared to develop c e r t a i n s p e c i f i c fears as a function of evolutionary past. Thus, prepared phobias are viewed as possessing p a r t i c u l a r b i o l o g i c a l s i g n i f i c a n c e , r e l a t e d to s u r v i v a l of the species. Such prepared learning i s s e l e c t i v e , highly r e s i s t a n t to e x t i n c t i o n , probably noncognitive and may be acquired in one t r i a l (Emmelkamp, 1979; Hugdahl and Karker, 1981; Rachman and Seligman, 1976). However, the v a l i d i t y of the preparedness theory has been quest-ionned (see Emmelkamp, 1979). Further arguments c i t e d by Rachman included the d i f f i c u l t y i n r e c o n c i l i n g both the d i s t r i b u t i o n of fears i n normal and neurotic populations, and case h i s t o r i e s that have described a gradual development of fear in the absence of traumatic experiences, with the two-stage theory. Rachman (1977) has suggested that conditioning constitutes only one of three main fear induction processes (or, "pathways of f e a r " ) . He claimed that v i c a r i o u s experience and transmission of information (through s o c i a l modelling or instruction) also e l i c i t fear. In addition, b i o l o g i c a l d ifferences in the propensity to develop fears are believed to e x i s t . Moreover, Rachman (1978) hypothesised that some in d i v i d u a l s are p a r t i c u l a r l y prone to develop conditioned fears (probably of a prepared nature) while others are more susceptible to s o c i a l l y transmitted fears (of an unprepared nature). However, Emmelkamp (1979) commented upon the lack of d e f i n i t i v e data for either the v i c a r i o u s or informational hypothetical a c q u i s i t i o n processes. Mowrer's model s t i l l f a i l s to adequately account for the persistence of fear. That i s , although avoidance responses serve to maintain f e a r f u l reactions, some nonreinforced exposure (where the stimulus i s not avoided) s t i l l occurs. Moreover, evidence c l e a r l y shows that avoidance behaviour i s not c a u s a l l y mediated by an underlying drive state of autonomic arousal (Wilson and O'Leary, 1980). F i n a l l y , the two-stage theory cannot explain the increase i n fear that sometimes occurs over e x t i n c t i o n t r i a l s (Costello, 1970) . Incubation Model Recognition of the paradoxical enhancement of fear over e x t i n c t i o n t r i a l s led Eysenck (1976) to develop hi s incubation model of 8 human neurosis. I t states that most neuroses r e s u l t from r e p e t i t i v e , b r i e f , intense exposure to i n i t i a l l y moderately fear-arousing s t i m u l i v i a symbolic or actual contact. In b r i e f , the theory presumes that during Pavlovian conditioning, UCS e l i c i t complete UCRs. When substituted for the UCS, the CS then e l i c i t s the major components of the UCR. As in the two-stage theory, the CR i s believed to acquire drive properties, hence providing reinforcement for the CS even when the UCS i s absent. Eysenck hypothesised that the net e f f e c t of the CS-only presentation depends on the balance between the forces of e x t i n c t i o n and enhancement of the CR, which i n turn depend on the point a at which the CS presentation i s terminated. Each i n d i v i d u a l i s believed to have a r e l a t i v e l y constant c r i t i c a l point i n CR strength, and i f the CR i s below that point at CS termination, then e x t i n c t i o n occurs. If the CR i s above that point, i t i s enhanced. Therefore, i t i s possible for the CR to exceed the strength of the UCR, r e s u l t i n g i n a slow growth of the neurotic response over time. Lengthy exposures to the feared stimulus would there-fore r e s u l t i n e x t i n c t i o n of the CR (Eysenck, 1976; Bersch, 1980). Eysenck offered some explanation for i n d i v i d u a l d ifferences i n fear a c q u i s i t i o n and fear persistence by l i n k i n g tearfulness to personality c h a r a c t e r i s t i c s ; introversion/extraversion and neuroticism. He postulated that excessively f e a r f u l i n d i v i d u a l s are l i k e l y to be highly introverted and neurotic, as in t r o v e r t s acquire conditioned responses more r e a d i l y than extraverts, and neurotics possess l a b i l e nervous systems (Eysenck, 1976; Rachman, 1978). Eysenck's model, however, has received considerable c r i t i c i s m (Bersch, 1980). The two-stage theory and the incubation model have been very i n f l u e n t i a l i n fear research, but recent evidence has provided reason to question the v a l i d i t y of these theories. Both theories postulate a synchronous and causal r e l a t i o n s h i p between fear arousal and avoidance behaviour. The evidence, however, suggests that while arousal may be a cause for avoidance 9 behaviour, i t i s unnecessary either for learning or for maintaining avoidance behaviour (Bandura and Adams, 1977; Rachman, 1976). Experimental and c l i n i c a l evidence c l e a r l y indicate the absence of a consistent causal r e l a t i o n s h i p between avoidance and arousal. Cognitive-Behavioural Conceptualisations Rachman's (1977) concept of three "pathways of f e a r " acknowedges the role of cognitive v a r i a b l e s . Most psychologists now agree that cognitions are e s s e n t i a l i n the analysis of complex behaviour, and are examining the r e l a t -ionship between verbal reports of private events and behaviour (Wilson and O'Leary, 1980). An important development i n the s h i f t from "S-R psychology" to more c o g n i t i v e l y oriented frameworks was s o c i a l learning theory (Bandura, 1969). This theory views behaviour, i n t e r n a l - i n t e r p e r s o n a l factors and environmental influences as r e c i p r o c a l l y i n t e r l o c k i n g determinants of "action" (Bandura, 1978, 1981; Coyne, 1982). As such, reinforcement i s not concept-ualised as automatically strengthening behaviour, but rather as involving the informative and incentive value of rewards. Simple temporal c o n t i g u i t y of events i s considered i n s u f f i c i e n t to explain c l a s s i c a l conditioning, as some r e a l i z a t i o n that the two events are correlated i s required. Moreover, behaviour change due to the l i n k between a response and the environment (operant conditioning) i s believed to depend on cognitive representations of contingencies (Bandura, 1969; Wilson and O'Leary, 1980). Thus, Moore, Mischel and Ziess (1976) found that the manner i n which c h i l d r e n c o g n i t i v e l y represented rewards was a s i g n i f i c a n t l y more potent determinant of c o n t r o l behaviour than the nature of the actual reward stimulus (in Wilson and O'Leary, 1980) . S i m i l a r l y , Meichenbaum (1973) suggested that we do not react to the environment per se, but to our perceptions of the environment. The nature of the perception i s thus believed to determine the emotional 1 0 response ( M a r z i l l i e r , 1980). Various models of s p e c i f i c cognitions that e l i c i t f e a r f u l responses have developed, including E l l i s ' s theory of " i r r a t i o n a l - b e l i e f s " and notions of discomfort anxiety, Beck's theory of "cognitive d i s t o r t i o n s " , and Bandura's Self E f f i c a c y theory. Self E f f i c a c y Theory S e l f e f f i c a c y theory was developed by Bandura i n 1977. Within t h i s theory, i n accordance with the s o c i a l learning theory framework, cognitive processes are at t r i b u t e d a prominent role i n the a c q u i s i t i o n and retention of behaviour ( M a r z i l l i e r , 1980; Rosenthal, 1978). Motivation, which influences the a c t i v a t i o n and persistence of behaviour, i s influenced by the cognitive processes of representing future consequences i n thought, goal s e t t i n g and s e l f - e v a l u a t i o n . Because s e l f reward i s co n d i t i o n a l upon achieving s e l f determined standards of performance ( i . e . personally set goals), i t s e f f e c t i s a t t r i b u t e d to s e l f induced persistence u n t i l the goal i s reached. F a i l u r e to a t t a i n performance standards motivates c o r r e c t i v e changes i n behaviour i n order to reach those standards. When goals are reached, s e l f rewards become contingent upon higher standards, and through such progressive e l e v a t i o n , motivation c o n t i n u a l l y influences behaviour (Bandura, 1977). Se l f e f f i c a c y theory claims that the effectiveness of various fear reduction techniques i s mediated through a common cognitive mechanism; each technique, to varying degrees, creates and strengthens expectations of personal e f f i c a c y . E f f i c a c y (or, mastery) expectations refer to one's "judgement of how well one can execute courses of action required to deal with prospective s i t u a t i o n s " (Bandura, 1982, p. 123). E f f i c a c y "involves a generative c a p a b i l i t y i n which component cognitive, s o c i a l and behavioural s k i l l s must be organized into integrated courses of action to serve innum-erable purposes" (Bandura, 1982, p. 123). E f f i c a c y expectations are d i s t i n g -uished from outcome expectancies, which refer to the estimate that a given 11 behaviour w i l l produce a c e r t a i n outcome. Se l f e f f i c a c y i s characterised by three dimensions. Magnitude refe r s to the l e v e l of s e l f e f f i c a c y measured against tasks of increasing d i f f i c u l t y . Generality r e f e r s to the notion that expectancies may vary from engaging i n only a l i m i t e d number of tasks to a n t i c i p a t i n g success at a wide v a r i e t y of tasks and s i t u a t i o n s . Once s e l f e f f i c a c y i s established, i t may generalise to other s i t u a t i o n s . However, s e l f e f f i c a c y e s s e n t i a l l y remains s i t u a t i o n a l l y s p e c i f i c , and does not represent a global personality t r a i t . The f i n a l dimension i s strength, or c o n v i c t i o n , of s e l f e f f i c a c y at each l e v e l (Bandura, 1977). Bandura stated that e f f i c a c y expectations influence the choice of both a c t i v i t i e s and environmental s e t t i n g s , and determine the degree of both e f f o r t expended and persistence i n the face of obstacles or aversive experiences. Self e f f i c a c y , therefore, influences coping behaviour in threatening s i t u a t i o n s . Further refinement of the theory has led Bandura to suggest that while high perceived s e l f e f f i c a c y maintains the degree of e f f o r t required to master challenges, i t also dampens preparatory e f f o r t when a task i s viewed as nonchallenging; some uncertainty i n terms of the challenge of the task spurs preparatory a c q u i s i t i o n of knowledge and s k i l l s (Bandura, 1982) . Self e f f i c a c y i s also believed to mediate thought patterns and emotional reactions during a n t i c i p a t o r y and actual transactions with the environment. Poor s e l f e f f i c a c y i s believed to r e s u l t i n excessive dwelling upon personal d e f i c i e n c i e s i n turn creating s t r e s s and impairing performance by reducing e f f o r t and concentration upon the task at hand (Bandura, 1980, 1982). That i s , fear provoking thoughts of ineptitude are believed to produce emotional arousal. However, s k i l l s and incentives are also e s s e n t i a l determinants of action i n addition to perceived s e l f e f f i c a c y . These determinants are i n t e r -a c t i v e . P o s i t i v e incentives foster performance accomplishments which enhance 12 s e l f e f f i c a c y , and, gaining knowledge and s k i l l s that enable one to f u l f i l l personal standards also heighten s e l f e f f i c a c y (Bandura, 1982) . Bandura has provided evidence for his theory by examining the degree of congruence between s e l f percepts of e f f i c a c y and performance at the l e v e l of i n d i v i d u a l tasks. Research has shown that increasing l e v e l s of perceived s e l f e f f i c a c y both across groups and within subjects c o r r e l a t e with progress-i v e l y higher performance accomplishments (Bandura, 1982). The extent to which s e l f e f f i c a c y i s a l t e r e d by experience depends on the p a r t i c u l a r source of e x p e r i e n t i a l information and the way i n which that information i s c o g n i t i v e l y appraised. .Of the four main sources of s e l f e f f i c a c y information "performance accomplishment" i s the most i n f l u e n t i a l . Accomplishment provides evidence of personal achievement, and develops s k i l l s that enable i n d i v i d u a l s to cope with threatening s i t u a t i o n s (Bandura, 1977). Fear reduction techniques that are based upon t h i s source of s e l f e f f i c a c y information include p a r t i c i p a n t modelling, d e s e n s i t i z a t i o n and flooding, as each includes enactive mastery. "Vicarious experience" - observing others perform s u c c e s s f u l l y - can also r a i s e e f f i c a c y expectations. In addition to providing a means of s o c i a l comparison, the model also conveys information about the nature and p r e d i c t a b i l i t y of events, and teaches s t r a t e g i e s for dealing with threatening s i t u a t i o n s . Self i n s t r u c t i o n t r a i n i n g and suggestive therapies focus on the "verbal persuasion" source of s e l f e f f i c a c y information by convincing i n d i v i d u a l s that they possess the c a p a b i l i t i e s to cope with threatening events. F i n a l l y , " p h y s i o l o g i c a l state" serves as a basis for judgements concerning one's l e v e l of anxiety from which inferences are made about performance c a p a b i l i t i e s : "Because high arousal usually d e b i l i t a t e s performance, people are more i n c l i n e d to expect success when they are not beset by aversive arousal than i f they are tense and v i s c e r a l l y agitated" (Bandura, 1982, p. 129). Treatment approaches that enhance s e l f e f f i c a c y 13 through the p h y s i o l o g i c a l information source include relaxation and biofeed-back (Bandura 1977; Bandura and Adams, 1977). Bandura and his colleagues have tested t h e i r predictions through a serie s of experiments using snake and spider phobics (Bandura, 1977; Bandura and Adams, 1977; Bandura, 1980; Bandura, 1982) . Bandura concluded that their experiments provide confirmation for the hypothesis that d i f f e r e n t therapeutic modes a l l r a i s e and strengthen s e l f percepts of e f f i c a c y . Also, i t was concluded that behaviour corresponds c l o s e l y to the l e v e l of s e l f e f f i c a c y change regardless of the method by which i t i s enhanced: the higher the l e v e l of s e l f e f f i c a c y , the greater the performance accomplishments; and the stronger the s e l f e f f i c a c y , the more l i k e l y i n d i v i d u a l s w i l l p e r s i s t u n t i l they succeed. Moreover, as predicted, enactive mastery provides the highest, strongest and most generalised increases in s e l f e f f i c a c y . Bandura (1982) also found that s e l f e f f i c a c y i s a superior predictor of performance i n behavioural approach tests than i s the l e v e l of performance accomplishment attained during treatment. S i m i l i a r r e s u l t s were obtained when the behavioural domain was extended to agoraphobia (Bandura, Adams, Hardy and Howells, 1980). However, the influence of s e l f e f f i c a c y information sources i s mediat-ed by cognit i v e a p p r a i s a l processes; including appraisals of task d i f f i c -u l t y , amount of e f f o r t expended and external a id received, s i t u a t i o n a l c i r c -umstances and temporal patterns of success and f a i l u r e (Bandura, 1977, 1982). For instance, external a t t r i b u t i o n may mitigate the s e l f e f f i c a c y enhancing e f f e c t of performance success. Moreover, several factors a f f e c t the strength of the r e l a t i o n s h i p between s e l f e f f i c a c y and subsequent action; f a u l t y self-knowledge, misjudgement of task requirements, unforeseen s i t u a t i o n a l c o n s t r a i n t s , d i s i n c e n t i v e s to act upon s e l f e f f i c a c y and new experiences that prompt reappraisals of s e l f e f f i c a c y i n the i n t e r v a l between assessment of s e l f e f f i c a c y and action (Bandura, 1982). 14 Self e f f i c a c y theory has been mainly applied to fear and avoidance behaviour. The theory states that, because s e l f e f f i c a c y influences the i n i t i a t i o n and persistence of coping behaviour, i n d i v i d u a l s fear and avoid those s i t u a t i o n s they believe to be beyond th e i r coping s k i l l s , thereby r e t a i n i n g s e l f d e b i l i t a t i n g expectations and defensive behaviour (Bandura and Adams, 1977). In contrast, "those who p e r s i s t i n s u b j e c t i v e l y threatening a c t i v i t i e s that are i n f a c t r e l a t i v e l y safe w i l l gain c o r r e c t i v e experiences that rei n f o r c e t h e i r sense of s e l f e f f i c a c y , thereby eventually eliminating th e i r defensive behaviour" (Bandura, 1977, p. 194). Self e f f i c a c y theory therefore explains the effectiveness of exposure therapy i n a much d i f f e r e n t manner from the incubation model. Eysenck, however, maintains that s e l f e f f i c a c y theory i s e s s e n t i a l l y a noncognitive theory which merely translates the major features of the incubation model into c o g n i t i v e / m e n t a l i s t i c language (Eysenck, 1978). C r i t i c i s m s of S e l f E f f i c a c y Theory Bandura's theory has been the subject of extensive t h e o r e t i c a l and experimental examination; both support and c r i t i c i s m of the theory e x i s t . Several c r i t i c s have challenged the causal s i g n i f i c a n c e given to cognitive processes by s e l f e f f i c a c y theory. Eysenck (1978) claimed that the treatment e f f e c t s i n Bandura et a l s ' research can be be as e a s i l y explained by constructs r e l a t i n g to conditioned autonomic responses, and that s e l f e f f i c a c y i s merely a phenomonological by-product of underlying p h y s i o l o g i c a l change. S i m i l a r l y , Borkovec (1978) viewed s e l f e f f i c a c y as a by-product of simple conditioning. Despite Bandura's attempts to e s t a b l i s h a causal r e l a t i o n s h i p between s e l f e f f i c a c y and action by m i c r o a n a l y t i c a l l y analysing congruence rates, i t i s not s u f f i c i e n t to merely demonstrate c o r r e l a t i o n s i n order to e s t a b l i s h a causal r e l a t i o n s h i p (Eysenck, 1978; Kazdin, 1978). Hence, Bandura's r e s u l t s remain open to a l t e r n a t i v e i n t e r p r e t a t i o n s , such as those by Eysenck and Borkovec. 15 S t i l l other c r i t i c i s m s focus upon the ambiguities within s e l f e f f i c a c y theory i t s e l f . Lang (1978) noted that the q u a l i f i c a t i o n entered into the r e l -a tionship between sources of information and subsequent l e v e l s of s e l f e f f i c a c y by the "cognitive a p p r a i s a l " aspect allows "an e x c e l l e n t post hoc i n t e r p r e t -ation of any behaviour outcome" (p. 190). Moreover, the elements of both the cognitive appraisal and incentive factors have not been s u f f i c i e n t l y spec-i f i e d . S i m i l a r l y , factors postulated to influence the r e l a t i o n s h i p between s e l f e f f i c a c y and subsequent action also allow i n t e r p r e t a t i o n of the f a i l u r e to observe predicted behaviour patterns. Bandura therefore emphasises the importance of m i c r o a n a l y t i c a l analyses i n order to f i n d predicted r e l a t -ionships. However, t h i s requirement severely r e s t r i c t s the a p p l i c a b i l i t y and g e n e r a l i z a b i l i t y of the theory. Also, Kazdin (1978) suggested that the three postulated requirements for action - s e l f e f f i c a c y , s k i l l and incentive -cannot be assessed independently. I t i s possible that, at optimal l e v e l s , competence and incentives alone may account for behavioural outcomes without the need to incorporate notions of s e l f e f f i c a c y . Therefore, Kazdin (1978) suggested the development of a more precise means of e m p i r i c a l l y s p e c i f y i n g appropriate l e v e l s of competence and incentive. In addition, other researchers have not been successful in r e p l i c a t i n g Bandura's f i n d i n g s . F e l t z (1982), for example, found that performance on previous t r i a l s was a more r e l i a b l e predictor of back diving performance than s e l f e f f i c a c y judgements. Another major c r i t i c i s m of s e l f e f f i c a c y theory i s the juxtaposition i t presents to the i n c r e a s i n g l y prominent three-systems model of fear and avoidance (Lang, 1968; Rachman and Hodgson, 1974). The c o n f l i c t between the predictions derived from s e l f e f f i c a c y theory and the three-systems model was the major focus of the study presented herein. 16 THE NATURE OF FEAR ... i t i s c r y s t a l clear that the problems we deal with are considerably more complex than we o r i g i n a l l y believed. (Hersen, 1981, p. 18) Such a statement i s e s p e c i a l l y apt i n describing our present understanding of fear and anxiety, these being no longer conceptualised as g l o b a l , u n d i f f -erentiated phenomena (Freedman, Dornbush and Shapiro, 1981). I t has been these advances, based on c l i n i c a l observations, research and t h e o r e t i c a l developments, that rendered the two-stage theory of fear and avoidance untenable. Inherent in Mowrer's theory was the assumption that subjective fear, autonomic arousal and avoidance behaviour together represent an unidimensional anxiety reaction; that autonomic responses both regulate avoidance behaviour, and must be reduced in order to eliminate avoidance. However, as previously mentioned, fear and avoidance are not always c l o s e l y t i e d . Formulation of the Three-Systems Model In reaction to repeated c l i n i c a l observations of the lack of c o r r e l a t i o n between fear and avoidance, Lang (1968) presented a new conceptualisation of fear: Emotional behaviours are multiple response systems -verbal-cognitive, motor and p h y s i o l o g i c a l events - that i n t e r a c t through interoceptive (neural and hormonal) and exteroceptive channels of communication. A l l systems are c o n t r o l l e d or influenced by brain mechanisms, but the l e v e l of the important centres of influence ( c o r t i c a l or s u b - c o r t i c a l , limbic or brain stem) are varied, and l i k e the r e s u l t i n g behaviours, p a r t i a l l y independent. Because of t h i s imperfect coupling, i t i s possible and even usual to generate emotional cognitions without autonomic arousal, aggressive behaviour without a h o s t i l e motive, or the autonomic and avoidant behaviour of fear without i n s i g h t (proper l a b e l l i n g ) . (Lang, 1971, p. 108) Thus, low c o r r e l a t i o n s (or, discordance) found upon sampling i n each system were taken to indicate that fear i s not a unitary phenomena (Kozak and M i l l e r 17 in press). The emotional response was viewed as a complex of three major measurable systems which are modulated by neural brain centres, and whose outputs have low i n t e r c o r r e l a t i o n s , yet are highly i n t e r a c t i v e (Lang, 1971) . However, a l l three systems are generally responsive to moderate to intense degrees among ' c l i n i c a l l y ' f e a r f u l patients (Rachman, 1978). Rachman and Hodgson (1974) described eight possible combinations of the response systems: Verbal Behavioural P h y s i o l o g i c a l 1 + + + 2 + + 3 + - + 4 + 5 - + + 6 - + -7 - - + 8 - - -The f i r s t represents true fear fulness, while the l a s t represents true fearlessness. The l a t t e r four types do not usually come to therapeutic attention, as the response i s not l a b e l l e d as anxiety or fear by the i n d i v -id u a l (Rachman, 1976). The Three Systems of Fear Since the o r i g i n a l formulation of the three-systems model, Lang (1978) has r e s t r i c t e d the verbal-cognitive component to refer only to verbal reports of fear, i n order to overcome the d i f f i c u l t i e s encountered in making assumpt-ions about the nature of cognitive events from verbal reports (Kozak and M i l l e r , in press). However, the verbal component has been interpreted i n many ways. Rachman (1978) described t h i s subjective aspect as "an alarming f e e l i n g of intense fear or panic" (p. 140) i n phobic patients. The verbal system has been measured i n terms of either "subjective fear and discomfort" r a t i n g s , apprehension or f r i g h t (Rachman and Hodgson, 1974). S p e c i f i c a n x i e t y - e l i c i t i n g thoughts and s e l f statements have also been i d e n t i f i e d ; 18 erroneous i n t e r p r e t a t i o n s , negative s e l f statements, i r r a t i o n a l b e l i e f s s e l f preoccupation and catastrophising statements (Boulougouris, Rabavilas and Stefanis, 1977; Elder, E d e l s t e i n and Fremow, 1981; Kanter and G o l d f r i e d , 1979; Hollandsworth, Glazeski, Kirkland, Jones and Van Norman, 1979; Kendrick, 1979; Lehrer, Schoicket, Carrington and Woolfolk, 1980; Mathews and Shaw, 1977; Meichenbaum, 1978; Ollendick and Murphy, 1977; Rabavilas and Bouloug-o u r i s , 1976; Sutton-Simon and G o l d f r i e d , 1979; Woodward and Jones, 1980). Another dimension of the verbal system has been i d e n t i f i e d as the degree of awareness of i n t e r n a l p h y s i o l o g i c a l cues (Borkovec, 1973b). Borkovec, Weerts and Bernstein (1977) suggest that "autonomic perception" i s somewhat indep-endent of actual arousal and i n t e r a c t s with the behavioural and autonomic systems. The behavioural system also e n t a i l s several elements, including avoidance or escape behaviour upon confrontation with a feared stimulus. Learning and performance d e f i c i t s are more relevant behavioural indices of s o c i a l and performance an x i e t i e s . Observable signs of arousal l e v e l - such as trembling, sweating and so on - have also been included i n the behavioural system. Response in the " p h y s i o l o g i c a l " system r e f e r s to sympathetic innervation of the autonomic system, and w i l l henceforth be referred to as the autonomic response system. Fear e l i c i t s various changes i n the autonomic system; for example, a reduction in blood flow to the bodily extremities and increased blood flow to the brain, suppression of d i g e s t i v e and sexual processes, and increased r e s p i r a t i o n , e l e c t r i c a l conductance properties of the skin, sweat gland a c t i v i t y and muscle tension. Those changes are manifested i n heart p a l p i t a t i o n s , increased heart rate, sweaty palms, dry mouth and 'knotty stomach', re s p i r a t o r y d i f f i c u l t i e s and so on (Bellack, 1980). As Lang o r i g i n a l l y claimed, and as Schwartz (1978) emphasised, the 19 verbal, behavioural and autonomic systems are a l l the observable outcomes of various l e v e l s of p h y s i o l o g i c a l processes. Interdependency of the Three Systems Lang (1971) described both the i n t e r a c t i v e and independent nature of the three systems of fear. The i n t e r a c t i v e nature of the response systems i s evidenced by the blocking influence of i n s t r u c t i o n a l sets upon arousal resp-onses. S i m i l a r l y , i n t e l l e c t u a l i s i n g or denying the p a i n f u l circumstances of watching a s t r e s s f u l f i l m may decrease l e v e l s of tonic autonomic arousal that would otherwise have been present. In more general terms, Lang (1971) described a mutually augmenting feedback loop between the systems: an i n d i v i d u a l possessing a " d i s t r e s s " set can become highly aware of autonomic feedback, which then confirms the cognitive 'set', and which acts as a d i s -criminative cue for further anxious cognitions; the autonomic a c t i v i t y may then feed back i n t e r o c e p t i v e l y to influence s t r i a t e muscle reflexes and c o r t i c a l a c t i v i t y ; such feedback may then dis r u p t the behavioural component by producing u n s k i l l f u l performance or by blocking motor a c t i v i t y ; and such f a i l u r e of psychomotor functioning may then be exteroceptively perceived and i t s negative valence may add to the ascending s p i r a l of verbal, behavioual and autonomic a c t i v a t i o n . The i n t e r a c t i o n of the verbal-cognitive and autonomic systems has been e s p e c i a l l y emphasised by others, such as Borkovec, Grayson and Hennings (1979) and Meyer and Reich (1978) . As a r e s u l t of the i n t e r a c t i v e nature of the response systems, various theories have been able to describe anxiety i n terms of i n d i v i d u a l response systems. Some t h e o r i s t s , such as E l l i s (1979) and Meichenbaum (1978), postulate that changes i n s e l f statements and thinking s t y l e s r e s u l t i n fading of gross motor and autonomic fear responses. Others, such as Borkovec (1978) and Wolpe (1978) , believe that reduction of autonomic arousal 20 f a c i l i t a t e s the reporting of less anxiety and behaviour change. A l t e r n -a t i v e l y , Marks (1972) claimed that behaviour modification can a l t e r verbal reports and v i s c e r a l a c t i v i t y . Therefore, Lang's conceptualisation provides support for each approach. Borkovec (1976) has s p e c i f i e d nine such i n t e r -a c t i o n a l patterns between the response systems. Independence of the three systems becomes e s p e c i a l l y evident when fear i s attenuated; mild f e e l i n g states may merely involve a verbal report of anxiety (Lang, 1971). Independence i s a t t r i b u t e d f i r s t l y to the f a c t that none of the systems uniquely defines the emotional states; that each i s influenced by various other f a c t o r s . Moreover, the systems are not equally s e n s i t i v e to stress s t i m u l i (Lang, 1977). By using a standardized index of s e n s i t i v i t y , Agras and Jacob (1981) found that while verbal and behavioural systems were approximately equally s e n s i t i v e , autonomic indices were les s s e n s i t i v e . In general, s e l f reports of fear c o r r e l a t e moderately well with avoidance behaviour, but only modestly well with autonomic indices (Rachman, 1978). Also, system-sensitivity may d i f f e r across i n d i v i d u a l s (Lang, 1977). F i n a l l y , inadequate measurement may also account for some degree of indep-endence of the systems (Lang, 1977). Synchrony and Desynchrony Rachman and Hodgson (1974) extended the three systems model on the basis of the observation that not only may the three systems be imperfectly c o r r e l a t e d at any given point, but they may also respond to treatment with d i f f e r e n t rates of change. Lang's o r i g i n a l formulation centred around notions of concordance and discordance, whereas Rachman and Hodgson coined the terms of synchrony and desynchrony to refer to the degree of covariance of the changes among the response systems. Desynchronous change may involve either independent or inverse r e l a t i o n s h i p s . In general, the temporal order 21 of change i n response to treatment proceeds from autonomic to behavioural and then to verbal systems (Rachman, 1976b, 1978). Rachman and Hodgson (1974) i d e n t i f i e d two common patterns of synchrony; the f i r s t involving moderate subjective fear and mild avoidance behaviour ( c h a r a c t e r i s t i c of s o c i a l anxiety), and the second involving intense fear and active avoidance followed by a reduction i n fear while avoidance continues unchecked ( c h a r a c t e r i s t i c of obsessional neurosis). Theoretical Implications Implications of t h i s model include the need to seek separate explanations for the o r i g i n s and course of each response component. Rachman's (1978) predictions concerning th e i r respective mechanisms of a c q u i s i t i o n draws upon the notion of "three pathways of fear" and b i o l o g i c a l d i f f e r e n c e s in the propensity to develop fears. He suggested that some i n d i v i d u a l s may be p a r t i c u l a r l y prone to develop fears by conditioning, while others may be more susceptible to the s o c i a l transmission of fears. Rachman (1978) incorp-orated the three systems model into those ideas by speculating that fears acquired by conditioning are l i k e l y to mostly involve autonomic and behav-i o u r a l responses, with the subjective element remaining minor i n comparison. In accordance Hugdahl (1981) reported that subjects with high autonomic arousal were more prone to autonomic conditioning. Further, Rachman stated that s o c i a l l y transmitted fears are more l i k e l y to comprise predominantly subjective responses. Rachman's predictions concord with Seligman's prep-aredness theory. Prepared fears are l i k e l y to be acquired by conditioning and are l a r g e l y noncognitive, whereas unprepared fears that are s o c i a l l y transmitted contain a larger cognitive element (Rachman, 1978) . In addition to b i o l o g i c a l propensities, Borkovec (1976) suggested that d i f f e r e n c e s i n i n d i v i d u a l learning h i s t o r i e s may also account for the r e l a t i v e i n t e n s i t i e s , or, f u n c t i o n a l importance of each response system i n emotional reactions. 22 As well, Rachman (1978) has suggested that the systems are effected by d i f f e r e n t decremental processes. On the basis of empirical evidence showing that habituation weakens unlearned responses, while e x t i n c t i o n weakens learned responses, he suggested that the autonomic system may be p a r t i c u l a r l y susceptible to habituation, while the behavioural system resp-onds more to e x t i n c t i o n , with the subjective system responding to both decremental processes. Rachman and Hodgsons' model also has important treatment implications. According to the model, fear reduction techniques are hypothesised to produce d i f f e r e n t patterns of covariance among the response systems. For example, systematic d e s e n s i t i z a t i o n i s believed to produce a reduction i n subjective and autonomic responses, followed by a s l i g h t reduction i n avoidance, which i s eventually eliminated. P a r t i c i p a n t modelling r e s u l t s i n a high degree of synchrony v i a a steady, gradual decline of each response system. Flooding therapy decreases avoidance behaviour before influencing autonomic and subjective responses. F i n a l l y , spontaneous remission i s characterised by gradual autonomic and subjective fear reduction, while avoidance continues. Supporting Evidence In order to provide a framework for p r e d i c t i n g the degree of concordance and/or synchrony, Hodgson and Rachman (1974) developed f i v e hypotheses on the basis of past research. These have directed subsequent research. F i r s t , they hypothesised that concordance i s more l i k e l y to occur under conditions that evoke strong emotional responses, while discordance i s more l i k e l y under conditions that evoke weak emotional responses. Hence, i n d i v i d u a l s presenting with c l i n i c a l l y severe fear reactions are more l i k e l y to respond to the presence of the feared o b j e c t ' i n each system, whereas mildly f e a r f u l i n d i v i d u a l s from analogue populations may respond i n only the more s e n s i t i v e systems. This discrepancy i n patterning between c l i n i c a l and analogue fears 23 i s important when generalising from analogue research. Sartory, Rachman and Grey (1977) and Marks, Marset, Boulougouris and Huson (1971) have provided empirical support for the f i r s t part of t h i s hypothesis, and studies examining analogue populations, or mildly f e a r f u l i n d i v i d u a l s , generally indicate an absence of autonomic responding (Borkovec, 1973b, 1976). S a l l i s , L i c h s t e i n and McGlynn (1980) reviewed 41 methodologically diverse i n d i v i d u a l and group c l i n i c a l studies of phobics and anxiety states, and 54 analogue fear studies, f i n d i n g that congruence between autonomic and subjective systems measures was most often reported i n i n d i v i d u a l c l i n i c a l studies (87%) and l e a s t often in analogue studies (24.7%). S i m i l i a r patterns appeared when comparing autonomic and behavioural systems measures (65.4% and 41.7% r e s p e c t i v e l y ) . Behavioural-subjective congruence was also most frequent in i n d i v i d u a l c l i n i c a l studies (72%). S a l l i s et a l . note that analyses of group means may r e s u l t i n spuriously low estimates of congruence. The pattern of generally increasing congruence from analogue to group c l i n i c a l to i n d i v i d u a l c l i n i c a l studies supports Hodgson and Rachmans' hyp-othesis. The second hypothesis stated that concordance i s more l i k e l y under low demand conditions (such as d i s t r i b u t e d or graduated practice) than under high demand conditions (such as massed prac t i c e or f l o o d i n g ) , where di s c o r d -ance i s more l i k e l y . Grey, Sartory and Rachman (1979) have provided some empirical support for t h i s hypothesis. T h i r d l y , Hodgson and Rachman hypoth-esised that the degree of synchrony r e s u l t i n g from treatment depends on the p a r t i c u l a r therapy employed; for instance, flooding i s believed to produce desynchrony whereas p a r t i c i p a n t modelling i s believed to produce synchrony. However, the evidence for t h i s i s contradictory; some studies report syn-chrony during flooding (Connolly, 1973; Emmelkamp and Emmelkamp-Benner, 1975; Rabavalis e t . a l . 1976; Stern and Marks, 1973); others report desynchrony 24 (Foa, Jameson, Turner and Paynes, 1980; Benjamin, Marks and Huson, 1972; Watson, Gaind and Marks, 1972; Lande, 1982; Nunes and Marks, 1975; Marshall, Gauthier, C h r i s t i e , Currie and Gordon, 1977): d i f f e r e n t i a l patterns across i n d i v i d u a l subjects have also been observed (Andrasik et a l . , 1980; Barlow, Mavissakalian and S c h o l f i e l d , 1980; Leitenberg, Agras, Butz and Wincze, 1971). It appears that the e f f e c t i s much more complex than a one-to-one r e l a t i o n s h i p between treatment type and outcome. The fourth hypothesis, that the degree of concordance following treatment increases during follow-up, has received d i r e c t empirical support (Foa et a l . , 1980). Also, i n d i r e c t support for that hypothesis derives from the frequent observation that s e l f report changes occur more slowly than changes i n the other response systems (Lande, 1982; Watson et a l . , 1982; Nunes and Marks, 1975). F i n a l l y , Hodgson and Rachman hypothesised that desynchrony between the autonomic system and other response systems w i l l be greater when skin conductance i s measured rather than heart rate. The evidence for t h i s i s as yet inconclusive (Grey et a l . , 1979). Despite the wide ranging implications of the three-systems model for fear assessment and treatment, Agras and Jacobs (1980) claimed that the model i s only slowly permeating c l i n i c a l research. In a review of 48 recent studies, Barlow et a l . (1980) found that only two gathered data in each response system. 25 MEASUREMENT ISSUES The measurement of each response system i s very important, as l e v e l s of discordance and desynchrony r e f l e c t the independence of the systems, as well as the influence of extraneous variables upon measurement devices. Accurate measurement i s e s s e n t i a l i n subje c t - s e l e c t i o n procedures (to ensure that the " f e a r f u l " sample i s t r u l y f e a r f u l ) and in treatment evaluation. Although a considerable amount of research has been dire c t e d towards assessing the adequacy of various measures, the findings have not generally been applied i n treatment outcome studies. The measurement of each response system, and related problems, are described below. Verbal Response System General questionnaires measuring the presence, absence or degree of anxiety f e l t i n response to a v a r i e t y of s i t u a t i o n s include the Fear Survey Schedule and the S-R Inventory of Anxiousness (the l a t t e r i s more s p e c i f i c than the former) (Hersen, 1973; Borkovec et a l . , 1977). More s p e c i f i c scales are also a v a i l a b l e to measure verbal reports of experienced fear i n d i r e c t response to the feared stimulus, and are administered immediately p r i o r to, during or immediately following exposure. One such scale i s the Fear Therm-ometer, although i t i s found to be vulnerable to demand influences. The A f f e c t Adjective C h e c k l i s t , Anxiety D i f f e r e n t i a l and the State scale of the S t a t e - T r a i t Anxiety Inventory are less subject to such i n v a l i d a t i n g influences, as the i r format obscures, or makes i t d i f f i c u l t for the subject to determine, the meaning of item responses (Borkovec et a l . , 1977). As with a l l verbal report measures, various problems a r i s e with the use of these questionnaires; unintentional or purposeful d i s t o r t i o n of the truth, 26 memory f a i l u r e , habitual response s t y l e s , i n a b i l i t y to make verbal d i s c r i m i n -ations, inexact or confusing t e s t items and demand influences (Borkovec et a l . , 1977; L i c k , 1977; Lick and Katkin, 1976). However, these verbal measures are s t i l l capable of i d e n t i f y i n g groups of i n d i v i d u a l s who d i f f e r on the other measures of fear ( M a r z i l l i e r , C a r r o l l and Newland, 1979). Autonomic Response System Inaccuracy of measurement i n the autonomic system i s l a r g e l y a r e s u l t of current lack of knowledge (Rachman, 1978). Also, most p h y s i o l o g i c a l responses are generated by a great v a r i e t y of i n t e r n a l and external s t i m u l i , so any autonomic event i s u n l i k e l y to be an exact substitute as an index of the psychological state (Lang, 1971). P h y s i o l o g i c a l measurement i s further complicated by the lack of correspondence between d i f f e r e n t indices of auton-omic arousal, which may vary as a function of the e l i c i t i n g s t i m u l i (Lang, 1971). Therefore, autonomic arousal should be sampled i n various channels (such as cardiac, electrodermal and so on) in reaction to a p a r t i c u l a r stim-ulus (Hasset, 1978). Additional problems a r i s e from i n d i v i d u a l d i f f e r e n c e s i n the l a b i l i t y of d i f f e r e n t response channels (Lang, 1971). Range c o r r e c t -ion can be u t i l i s e d to remove t h i s confounding influence upon the i n t e r p r e t -ation of the data, although i t i s also important to examine absolute values when comparing groups (Lang, 1971). Differences between and within i n d i v -iduals may be reduced by c o n t r o l l i n g for obvious b i o l o g i c a l v a r i a t i o n s ; time of day, menstrual month, di e t a r y considerations, fatigue and so on (Borkovec et a l . , 1977). However, such extraneous factors may be assumed to be e f f e c t i v e l y c o n t r o l l e d through random d i s t r i b u t i o n within a large population and by repeated measures (Lang, 1971; Borkovec et a l . , 1977). Results indicate that heart rate i s f a i r l y r e l i a b l y elevated when ind-i v i d u a l s d i r e c t l y confront feared s t i m u l i (Sartory et a l . , 1977). However, 27 the s p e c i f i c cognitive and motor demands of the mode of stimulus presentation influence heart rate l e v e l s . Hence, i t i s suggested that responses to f e a r f u l s t i m u l i should be compared to responses to neutral s t i m u l i under equivalent s e t t i n g conditions (Borkovec, Stone, O'Brien and Kaloupek, 1974). Measures of skin properties are useful because the skin receives only sympathetic and not parasympathetic innervation (Borkovec et a l . , 1977). However, skin properties are also influenced by the a t t e n t i o n a l and motor demands of the s e t t i n g , and by stimulus complexity, novelty and cognitive d i f f i c u l t y . Therefore, skin response should also be measured under equiv-alent conditions with both neutral and fear e l i c i t i n g s t i m u l i . Behavioural Response System The behavioural response system i s t y p i c a l l y measured by either the Behaviour Approach Test (BAT) in which the subject performs a s e r i e s of progressively more fear evoking tasks, or, by duration of exposure to the feared stimulus. Measures of both performance di s r u p t i o n and o v e r a l l perf-ormance effectiveness are u t i l i z e d when assessing s o c i a l , performance and t e s t anxiety. The advantage of performance measures over the BAT i s that autonomic and cognitive measures can be obtained during exposure without introducing the confound of s e l f c o n t r o l over the amount of exposure and extent of avoidance. That i s , during performance tests the subject has no opportunity to avoid the feared s i t u a t i o n (Borkovec and O'Brien, 1976). However, performance measures do introduce an a d d i t i o n a l confounding factor of evaluation anxiety, which can be accounted for by including a neutral condition i n the experimental design (Borkovec et a l . , 1977). The BAT was i n i t i a l l y considered to be the most objective measure a v a i l -able. However, the influence of extraneous s i t u a t i o n a l variabl es upon BAT performance i s now recognised. When the BAT i s introduced to the subject as 28 a method of fear assessment, thereby becoming contextually r e a c t i v e , the i m p l i c i t demand for avoidance at pretest i s very apparent, and may override the e x p l i c i t demand for approach (Bernstein, 1973) . Several s i t u a t i o n a l variables contribute to the i m p l i c i t avoidance demand; the subject has usually previously informed the therapist of his fear i n verbal reports, such that avoidance behaviour i s considered to be appropriate; the presence of the therapist i n i t s e l f may produce reactive e f f e c t s ; and where the assess-ment precedes therapy, subjects may r a t i o n a l i s e that unless avoidance i s demonstrated, therapy w i l l not be forthcoming. Conversely, during post-test following therapeutic e f f o r t , approach behaviour i s considered to be approp-r i a t e (Bernstein, 1973). Therefore, assessments may be inaccurate and t r e a t -ment e f f e c t s may be i n f l a t e d , or, at l e a s t , confounded by d i f f e r e n t i a l react-ions across subjects to the s i t u a t i o n a l influences. Though s t i l l involving some i m p l i c i t demand fa c t o r s , performance-type behavioural measures are l e s s l i k e l y to be subject to s i t u a t i o n a l influences, because, as mentioned, subjects have l i t t l e c o n t r o l over the s i t u a t i o n a l parameters. Borkovec (1973a, 1973b, 1976) has demonstrated that r e l i a n c e upon very s i t u a t i o n a l l y - v u l n e r a b l e verbal and behavioural measures can r e s u l t i n i n f l a t -ion of the estimates of fear s e v e r i t y . Since strong emotion i s usually characterised by concordance amongst the response systems, very f e a r f u l i n d i v -iduals are l i k e l y to become autonomically aroused, s u b j e c t i v e l y distressed and show avoidance behaviour when confronted with feared s t i m u l i . There-fore, Borkovec argued that without p h y s i o l o g i c a l responsiveness, and hence probably without intense fear, the behavioural and verbal systems are more susceptible to s i t u a t i o n a l demand influences. This hypothesis has received empirical support (Borkovec, 1973a, 1973b, 1976; Hicks and Sheinber, 1976; Bernstein and Paul, 1971; Smith, Deiner and Beaman, 1974; Levine, Gorman and Sherry, 1978; Trudel, 1979). The implication of these findings i s that studies which do not obtain p h y s i o l o g i c a l measures or do not c o n t r o l for s i t u a t i o n a l influences i n behavioural measures may s e l e c t subjects whose fears are not c l i n i c a l l y severe (where " c l i n i c a l " r e f e r s not to the psychiat-r i c status of the i n d i v i d u a l , but to the se v e r i t y of the f e a r ) . Analogue studies, therefore, are only considered u s e f u l when such measures are taken (Borkovec and Rachman, 1979; Kazdin, 1978). Extraneous s i t u a t i o n a l influences can be minimised i n several ways. Presenting the BAT under conditions of high demand to approach the feared stimulus overcomes s i t u a t i o n a l influences to some extent (Bernstein and Paul, 1971; Trudel, 1979). That i s , subjects who show s i g n i f i c a n t l y and c l i n i c a l l y relevant increases i n autonomic arousal and cognitive d i s t r e s s as a r e s u l t of the presence of the feared stimulus, or who are s t i l l unable to approach the feared object under high demand conditions are more l i k e l y to be c l i n i c a l l y f e a r f u l . Subjects should also view th e i r emotional responses as a source of concern which i n t e r f e r e s with normal functioning (Bernstein, 1973) . If mildly to moderately f e a r f u l subjects are used, th e i r performances can be compared with that of a demand change c o n t r o l group (Bernstein and Paul, 1971). A d d i t i o n a l l y , behavioural measurement (of any sort) can be presented i n a non-reactive context by not being i d e n t i f i e d as fear assessment. Beiman et a l . (1978) developed a nonreactive, i n d i r e c t measure of avoidance c a l l e d the Behavioural Committment Test for a group of mutilation phobics. The BCT involved phoning subjects unobtrusively and rating t h e i r willingness to donate blood, which the authors viewed as a high demand condition. Another strategy i s to e x p l i c i t l y emphasise the a c c e p t a b i l i t y of both avoidance and approach by informing subjects that they are not a l l expected to e x h i b i t strong fear, to disregard t h e i r previous s e l f report responses, and to respond n a t u r a l l y and spontaneously to the te s t i n g condition (Bernstein and Paul, 1971; Bernstein, 1973; Borkovec and O'Brien, 1976). Additional problems with most behavioural measures include the r e s t r i c t e d range of s i t u a t i o n s or s t i m u l i , and the a r t i f i c i a l i t y which may l i m i t g e n e r a l i s a b i l i t y of the r e s u l t s (Lick and Katkin, 1976; Cohen, 1977) . Moreover, autonomic measures should be taken continuously and not, as t y p i c a l l y occurs, immediately following behavioural measures when sets characterised by ' r e l i e f are l i k e l y to confound the p h y s i o l o g i c a l response (Borkovec and O'Brien, 1976) . Behavioural and verbal measures should also be of approximately equivalent s p e c i f i c i t y to ensure that d i f f e r e n t i a l s e n s i t i v i t i e s of the measurement devices do not cause discordance (Rachman and Hodgson, 1974). 31 TREATMENT ISSUES The three-systems model also has very important treatment implications, as any given anxiety disorder w i l l have d i f f e r e n t i a l features that require d i f f e r e n t treatment modalities. (Freedman et a l . , 1981). Borkovec et a l . (1977) stated that: Our concern in the assessment of anxiety i s to i d e n t i f y the extent to which each response component contributes to the c l i e n t ' s presenting problem and to apply therapeutic interventions appropriate to the type of response c h a r a c t e r i s i n g the i n d i v i d u a l ' s reaction to the feared s i t u a t i o n , (p. 403) S i m i l a r l y , If i n d i v i d u a l s load d i f f e r e n t i a l l y on the various components, i d e n t i f i c a t i o n of t h i s should then become a routine procedure in c l i n i c a l diagnosis i n order to be able to t a i l o r an appropriate method of treatment to the i n d i v i d u a l ' s p r o f i l e of responding. (Hugdahl, 1981, p. 82) Response Treatment Matching Borkovec (1973) suggested that the most e f f i c i e n t method of fear red-uction i s to modify that response system(s) most f u n c t i o n a l l y involved i n the o v e r a l l emotional response, by implementing a therapy that focuses on that response system. Autonomically focussed treatments include r e l a x a t i o n , systematic d e s e n s i t i z a t i o n , biofeedback and implosion. The verbal system i s the focus of r a t i o n a l emotive therapy, semantic conditioning, s e l f i n s t r u c t -ion, thought stopping, placebo, expectancy and stress innoculation. Rein-forcement for approach behaviour, response prevention, modelling, contact d e s e n s i t i z a t i o n and flooding therapy focus on the behavioural response system (Borkovec, 1976; Ost, Jerramalm and Johansson, 1981). The f o c a l aspect r e f e r s not to the mechanisms through which the various treatments operate, but to the response which they modify. Evidence for the superior effectiveness achieved by matching response p r o f i l e to treatment i s accumulating. Progressive relaxation has been shown to be more e f f e c t i v e with problems for which autonomic a c t i v a t i o n i s a major contributing factor (Borkovec and Sides, 1979) . S o c i a l s k i l l s t r a i n -ing seems to be more e f f e c t i v e with s o c i a l l y anxious i n d i v i d u a l s who resp-ond to stress predominantly in the behavioural system, as compared to the effectiveness of relaxation with those who respond predominantly i n the autonomic system (Ost et a l . , 1981). P a r t i a l support has been obtained for the superior effectiveness of applying cognitive r e s t r u c t u r i n g to c l i e n t s who are highly aware of i n t e r n a l arousal cues, and cognitive r e s t r u c t u r i n g plus systematic d e s e n s i t i z a t i o n to c l i e n t s not only highly aware of i n t e r n a l cues but who also respond with actual autonomic a c t i v a t i o n to s o c i a l s i t -uations (Shahar and Merbaum, 1981). Also, cognitive r e s t r u c t u r i n g seems to be more e f f e c t i v e with 'low' rather than 'high' autonomic responders (Emmelkamp, Kuipers and Eggeraat, 1978). Lehrer et a l . (1980) have also shown that meditation leads to more cognitive r e l a x a t i o n , whereas relaxation t r a i n i n g leads to more autonomic relaxation (on the basis of s e l f reports of arousal). In a related sense, Chambless, Foa, Groves and Goldstein (1979) have argued that, as agoraphobics fear not only external s t i m u l i but also i n t e r n a l p h y s i o l o g i c a l cues, maximal treatment effectiveness would derive from exposing agoraphobics to both fears. In confirmation, they demonstrated that exposure to external fear s i t u a t i o n s only (achieved by eliminating aware-ness of autonomic arousal through sedative medication) was l e s s e f f e c t i v e than flooding in which arousal was not eliminated. On the other hand, others have shown that the degree of anxiety evoked during exposure procedures does not influence outcome (Hafner and Marks, 1976; Marks, 1978; Mathews, 1978). Ollendick and Murphy (1977) found some experimental support for t h e i r pred-i c t i o n s that i n t e r n a l l y oriented i n d i v i d u a l s (on the locus of c o n t r o l dimens-ion) would be more responsive to techniques incorporating personal c o n t r o l , such as cognitive r e l a x a t i o n , whereas ex t e r n a l l y oriented i n d i v i d u a l s would be more responsive to muscular rel a x a t i o n , which i s more structured and d i r e c t i v e . Davidson and Schwartz have developed a psychobiological model based on the somatic and cognitive dimensions of anxiety. Their treatment sugg-estions are based on the p r i n c i p l e of psychobiological s p e c i f i c i t y : If we conceptualise p a r t i c u l a r behaviour systems as having a f i n i t e amount of channel space and i f we further assume that cognitive or somatic anxiety, for example, represents a r e c y c l i n g of unwanted information in each of the respective systems, then further a c t i v a t i o n of a neural nature in a p a r t i c u l a r system w i l l compete with the ongoing behaviour sequence for channel space. This competition r e s u l t s i n attenuation of the ongoing behaviour. (Schwartz et a l . , 1978, p. 324) Despite i t s d i f f e r e n t t h e o r e t i c a l o r i e n t a t i o n , Davidson and Schwartzs' res-earch s t i l l supports the notion of matching response system and treatment. Hence, the i r research indicates that i n d i v i d u a l s p r a c t i c i n g meditation possess l e s s cognitive and more somatic anxiety than i n d i v i d u a l s p r a c t i c i n g p h y s i c a l exercise, who possess more cognitive and less somatic anxiety (Davidson, 1978; Schwartz, Davidson and Goleman, 1978). In contrast to the above, some studies have f a i l e d to demonstrate a response system by treatment i n t e r a c t i o n (Elder et a l . , 1981; Kanter and G o l d f r i e d , 1979). In addition, close examination of a l l of these studies reveals many methodological problems, mainly in the area of measurement of each response system. For instance, several studies r e l i e d on s e l f report measures of autonomic arousal and did not c o n t r o l for extraneous s i t u a t i o n a l influences upon behavioural and verbal measures. C l e a r l y , further research i s required on t h i s issue. 34 Response System Generalization i n Reaction to Treatment The extent to which modification i n one response system w i l l influence the other response systems remains unclear, and hypotheses that have been postulated vary according to the extent to which the response systems are believed to i n t e r a c t . Borkovec (1976) emphasised the i n t e r a c t i v e q u a l i t y of the systems, and claimed that while autonomic responses are strong, simple manipulation of the behavioural and cognitive systems i s i n e f f e c t i v e , or i n e f f i c i e n t at lease. He a t t r i b u t e d t h i s to the d i s r u p t i v e e f f e c t of sym-pathetic arousal upon the remaining systems. Borkovec's statement c l o s e l y p a r a l l e l s the two-stage theory; arousal must be i n h i b i t e d before behavioural changes can be obtained. S i m i l a r l y , M a r z i l l i e r et a l . (1979) stated that more f e a r f u l subjects are more p h y s i o l o g i c a l l y reactive and require v e r i d i c a l changes in arousal before avoidance w i l l d ecline. However, as previously mentioned, empirical evidence has shown that behaviour can be a l t e r e d in s p i t e of autonomic arousal and, as a r e s u l t , arousal may subside (Leitenberg et a l . , 1971). While a s i n g l e response system has been shown to respond to a s p e c i f i c response focussed treatment, the extent to which the remaining response systems are influenced by that change i s unclear. As Rachman and Hodgson (1974) stated, each response system seems to react d i f f e r e n t l y to p a r t i c u l a r treatment regimes. Leitenberg et a l . (1971) did not f i n d a consistent r e l a t -ionship between changes in autonomic arousal and phobic behaviour during deconditioning. Others have shown that e l i m i n a t i o n of phobic behaviour can be preceded by either increases in autonomic arousal, decreases in autonomic arousal or no change (Bandura, 1978). Andrasik, Turner and Ollendick (1980) reported no s i g n i f i c a n t c o r r e l a t i o n s between pulse rate and s e l f report of discomfort at various stages during four out of eight in vivo exposure sessions with obsessive compulsive neurotics. Marks and Huson (1973) found l i t t l e c o r r e l a t i o n between s e l f report of arousal and p h y s i o l o g i c a l measures during imagination of phobic imagery (in Barlow et a l . , 1980). Barlow et a l . (1980) found varying patterns of synchrony and desynchrony during exposure sessions i n three agoraphobic women. Grey, Rachman and Sar ;tory (1981) also f a i l e d to f i n d a clear r e l a t i o n s h i p between heart rate and f e a r f u l behaviour during treatment. Conversely, Rabavalis et a l . (1976) found synchronous changes from baseline to post treatment periods between autonomic and subjective meas-ures under various t e s t i n g conditions. S i m i l a r l y , Connolly (1979) and Galassi et a l . (1981) observed covariance between autonomic and subjective systems. Obviously, cross study comparisons are confounded by various f a c t o r s , including the parameters hypothesised by Hodgson and Rachman (1974) to e f f e c t the degree of synchrony, such as i n t e n s i t y of the emotional response and the p a r t i c u l a r treatment employed, and other factors that may further complicate the r e l a t i o n s h i p between response p r o f i l e and treatment. For instance, covariance may occur d i f f e r e n t i a l l y across the eight d i f f e r e n t combinations of response p r o f i l e s i d e n t i f i e d by Rachman and Hodgson (1974). Also, the adequacy of assessment methods and the exact procedures used i n treatment implementation may confound cross study comparisons. Immediate and d i r e c t reduction of fear i n each response system i s preferred to the longer process involved i n applying a s p e c i f i c response focussed treatment to a p a r t i c u l a r response system, which may or may not be soon af t e r followed by changes in the other response systems. There i s growing evidence that desynchronous changes involving continued manifestation of fear i n at l e a s t one response system can lead to the return of pre-36 treatment l e v e l s of fear i n each response system (DeSilva and Rachman, 1981; Grey et a l . , 1979; Grey et a l . , 1981; Rachman, 1979a, 1980). Moreover, i f the subjective f e e l i n g of anxiety i s not removed, i n d i v i d u a l s remain d i s -s a t i s f i e d with treatment gains regardless of the changes made in autonomic and behavioural response systems (Lick and Katkin, 1976). Reduction of fear i n each system can be achieved i n two ways. The f i r s t i s Lazarus's m u l t i -modal therapy approach, which i s based on the premise that l a s t i n g therap-eu t i c change i s achieved in d i r e c t proportion to the number of s p e c i f i c parameters i d e n t i f i e d , monitored and treated within the BASIC ID model (Popler, 1977). The second, more e f f i c i e n t approach i s to i d e n t i f y treatments that are capable of influencing each response system d i r e c t l y . As Lang (1977) stated: A p r a c t i c a l approach would be to evaluate the multisystem e f f i c i e n c y of s p e c i f i c therapeutic regimes. Presuming that we are in a p o s i t i o n to assess behaviours pertinent to the three emotional response systems, we could not only compare treatments for th e i r effectiveness i n modifying their target response component, but also note th e i r ameliorative e f f e c t on the other response systems, (p. 182) Further research should be dire c t e d towards developing procedures which optimise a single treatment's influence on each response system. For i n s t -ance, there i s growing evidence that exposure therapy can d i r e c t l y influence the autonomic and subjective as well as the behavioural system Various explanations for the effectiveness of exposure therapy have been postulated, each assuming that exposure focuses upon one of the three response systems whose change i n d i r e c t l y e f f e c t s changes in the remaining response systems. Borkovec (1973) suggested that exposure acts upon conditioned autonomic resp-onses. Others a t t r i b u t e o v e r a l l e ffectiveness to the changes exposure prod-uces in s e l f statements; the i n d i v i d u a l r e a l i z e s that the anticipated d i s -astrous consequences do not occur (Emmelkamp et a l . , 1978; Emmelkamp, 1979), learns that there i s nothing to fear, and gains confidences i n both h i s own a b i l i t y to deal with the s i t u a t i o n , and i n the safety of the s i t u a t i o n (Bellack, 1980; E l l i s , 1979; Marshall et a l . , 1977; Mathews and Shaw, 1973) and, the decreased f e e l i n g s of discomfort mediate new approach behaviour (Foa and Steketee, 1979). Others a t t r i b u t e the e f f e c t s s o l e l y to behavioural changes (Marks, 1972). Hence, i t can be speculated that i f exposure therapy does d i r e c t l y influence each response system, then i t s effectiveness w i l l be maximised when the procedural aspects are t a i l o r e d to the response system(s) most f u n c t i o n a l l y relevant in any i n d i v i d u a l . For example, exposure duration has t r a d i t i o n a l l y been c o n t r o l l e d by predetermined time i n t e r v a l s or i n d i v i d -ual verbal reports of fear reduction (Andrasik et a l . , 1980). However, as shown, each system responds at d i f f e r e n t i a l rates of change to treatment so verbal reports may not be synchronous with autonomic and behavioural resp-onses. Therefore, i f the autonomic response i s c e n t r a l to an i n d i v i d u a l ' s o v e r a l l emotional response, termination of exposure should be based on red-uction i n autonomic arousal. Researchers have begun to investigate t h i s issue (Gauthier and Marshall, 1977; Thomas and Rapp. 1977). However, the r e l a t i o n s h i p between treatment and response p r o f i l e i s extremely complex and requires more extensive research. The main conclusions that can be drawn to date are that the three response systems frequently react d i f f e r e n t l y to treatment strategies., and that change i n one system does not always influence the remaining systems. 38 STATEMENT OF THE PROBLEM SELF EFFICACY THEORY VERSUS THE THREE-SYSTEMS MODEL The e s s e n t i a l d i f f e r e n c e between the three-systems model and s e l f e f f i c a c y theory i s that the former allows considerable response system independence, while the l a t t e r views the response systems as i n t e r l o c k i n g determinants of a more u n i f i e d construct. T h e o r e t i c a l Contrast In defending the proposed causal r o l e of s e l f e f f i c a c y , Bandura (1978) refuted the two-stage conditioning model which stated that autonomic arousal precedes and causes behavioural acts, and that s e l f e f f i c a c y i s a by-product of t h i s process. Bandura argues that as autonomic l a t e n c i e s are more pro-longed than the late n c i e s of s k e l e t a l response systems, autonomic arousal can not account for behaviour. Bandura also c i t e s evidence i n d i c a t i n g the presence of avoidance responses i n the absence of p h y s i o l o g i c a l arousal, and the persistence of avoidance after arousal i s extinguished. He concurs with Schwartz: A l l the data taken together shows that while fear (arousal), when present, may influence avoidance responding, i t i s not nec-essary for the occurrence of avoidance and i t s elimination does not r e s u l t i n the cessation of avoidance. (1978, p. 256) Therefore, Bandura argues that i f arousal cannot s a t i s f a c t o r i l y p r e d i c t avoidance behaviour, i t cannot serve as the causal antecedent for both beh-aviour and s e l f e f f i c a c y (as was suggested by Eysenck (1978) , Wolpe (1978) and Borkovec (1978)). This argument appears to be concordant with the three-systems model but Bandura's statement of the s o c i a l learning theory view of anxiety indicates that while he recognises some independence of the response systems, th e i r i n t e r a c t i v e nature i s emphasised to a much greater extent: 39 ... thought, a f f e c t , and action operate as r e c i p r o c a l l y i n t e r a c t i n g factors rather than as loosely linked components or as conjoint events ... The r e l a t i v e influence exerted by these three sets of i n t e r l o c k i n g factors w i l l vary i n i n d i v -iduals for d i f f e r e n t a c t i v i t i e s performed under d i f f e r e n t circumstances. (1978, p. 257) Fear reduction i n each system i s at t r i b u t e d to a superordinate cognitive factor (Newman and Brand, 1980). Bandura suggests throughout h i s research what while the three systems can be discordant and desynchronous, they are generally concordant and synchronous. He i s almost bound to present t h i s view i n order to uphold s e l f e f f i c a c y theory. That i s , i f s e l f e f f i c a c y mediates emotional responses, then the measures of emotional responding ( i . e . the behavioural, verbal and autonomic response systems) must a l l be responsive to changes i n s e l f e f f i c a c y . Several challenges have been made to Bandura's contention that s e l f e f f i c a c y mediates change i n each response system. Rachman noted that "Although there i s good reason to believe that improvements i n perceived s e l f e f f i c a c y are p a r t i c u l a r l y h e l p f u l , there i s no reason to suppose that a l l therapeutic changes are mediated by such improvements" (1978, p. 281). S i m i l a r l y , Kazdin (1978) stated that i t seems somewhat s i m p l i s t i c to a t t r i b u t e a l l therapeutic e f f e c t s to s e l f e f f i c a y , as i t i s u n l i k e l y to mediate a l l response components. In his review of Bandura's (1977) i n i t i a l statement of s e l f e f f i c a c y theory, Wolpe (1978) suggested that s e l f e f f i c a c y i s only relevant to the behavioural response system. In accordance, Kendrick (1979) found that s e l f e f f i c a c y was only cor r e l a t e d with behavioural outcome measures. S i m i l a r l y , Rachman (1978) implied that s e l f e f f i c a c y i s only d i r e c t l y relevant to the behavioural system. Rachman viewed s e l f e f f i c a c y as "an elaborated version of the sense of c o n t r o l l a b i l i t y " (1978, p. 8). C o n t r o l l a b i l i t y refers to a person's sense 40 of whether or not he i s in a p o s i t i o n to reduce the l i k e l i h o o d of an aversive event and/or i t s consequences occurring. U n c o n t r o l l a b i l i t y i s viewed as having a major e f f e c t on fear behaviour; confidence, or c o n t r o l l a b i l i t y , i s seen as a ma jor determinant of courageous behaviour, where "courageous" i s defined as the repeated approach and exposure to o b j e c t i v e l y dangerous s i t u a t i o n s in spite of high l e v e l s of subjective and autonomic arousal (Cox, Hallam, O'Connor and Rachman, 1980). Courage i s also dependent to a large extent upon competence, and to a lesser extent upon being.assigned tasks that involve s o c i a l r e s p o n s i b i l i t y (a phenomenon which Rachman (1979b) refe r s to "required helpfulness"), group support and role models. Rachman suggested that t r a i n i n g and p r a c t i c e lead to a growing sense of s e l f e f f i c a c y which i n turn reinforces courageous behaviour. However, reductions in autonomic and subjective fear that may follow repeated episodes of courageous behaviour are not a t t r i b u t e d to s e l f e f f i c a c y , but to habituation r e s u l t i n g from repeated exposure. Although Bandura (1978) said that exceptional cases of discordance between autonomic arousal and s e l f e f f i c a c y may occur in object-i v e l y dangerous s i t u a t i o n s , s e l f e f f i c a c y theory, as Lang (1978) noted, cannot explain why i n d i v i d u a l s who repeatedly approach s i t u a t i o n s that involve threat, and possess behavioural s k i l l s , continue to show arousal and to report fear ( i n i t i a l l y , u n t i l the fear and arousal habituate). S e l f e f f i c a c y theorising and research o r i g i n a l l y focussed upon the behavioural response system (or, performance attainments). However, the autonomic system i s c u r r e n t l y receiving greater emphasis. For instance: I t i s mainly perceived i n e f f i c a c y in coping with p o t e n t i a l l y aversive events that makes them fearsome. Experiences that increase perceived e f f i c a c y i n c o n t r o l l i n g p o t e n t i a l l y i n j u r i o u s events diminish fear arousal and increase commerce with what was previously avoided... People who believe that they can exercise some influence over the occurrence of aversive events d i s p l a y less autonomic arousal and impairment i n performance than those who do not believe they have personal c o n t r o l , even though both groups are subjected to the same aversive stimulation. (Bandura, 1978, p. 255) 41 That i s , low s e l f e f f i c a c y i s believed to r e s u l t i n preoccupation with perceived i n e f f i c a c y and p o t e n t i a l calamities which i n turn produce arousal. In contrast, perceived efficaciousness i s believed to reduce arousal (Bandura, 1980). In further s p e c i f y i n g the r e l a t i o n s h i p s between s e l f e f f i c a c y and arousal, Bandura (1980) predicted that at i d e n t i c a l l e v e l s of s e l f e f f i c a c y , decreasing strength of s e l f e f f i c a c y produces greater arousal. To t e s t t h i s p r e d i c t i o n , Bandura (1980) employed blood pressure and heart rate measures i n the assessment of spider phobics during a n t i c i p a t i o n and performance of tasks corresponding to varying strengths of perceived s e l f e f f i c a c i o u s n e s s . On tasks associated with strong percepts of s e l f e f f i c a c y , the subjects were v i s c e r a l l y unperturbed. On tasks assigned moderate l e v e l s of s e l f e f f i c a c y , heart rate and blood pressure increased above baseline l e v e l s . The a n t i c -i p a t i o n of tasks associated with weak percepts of s e l f e f f i c a c y (which could not usually be performed) resulted in decreases in heart rate and further increases in blood pressure. Several methodological inadequacies render t h i s data questionnable. F i r s t l y , the influence of d i f f e r e n t motor and cognitive demands inherent i n each task upon p h y s i o l o g i c a l response were not c o n t r o l l e d . Also, the reported increases i n heart rate were of questionnable c l i n i c a l s i g n i f i c a n c e . Moreover, the extent to which a spider phobic population i s representative of severe fear reactions i s equivocal. Most of Bandura's subject samples have been drawn from analogue populations, and the findings therefore may have l i m i t e d g e n e r a l i z a b i l i t y (Poser, 1978; Woolfolk and Laz-arus, 1979) . P h y s i o l o g i c a l measurements have not been obtained (except in the most recent reported study), and the e f f e c t of extraneous s i t u a t i o n a l influences upon verbal and behavioural measures have not been c o n t r o l l e d . Although, Bandura (1978) claimed that conditions which minimise the subject's incentive to misrepresent cognitive processes would be s u f f i c i e n t to obtain accurate verbal reports. However, h i s subjects' responses may not have r e f l e c t e d true fearfulness, and/or may have i n f l a t e d treatment effectiveness (Borkovec et a l . , 1977). Therefore, the r e l a t i o n s h i p between s e l f e f f i c a c y and each of the resp-onse systems should be examined under conditions which c o n t r o l extraneous s i t u a t i o n a l influences i n a population experiencing d e b i l i t a t i n g anxiety reactions. Treatment Choice Contrast Bandura contends that the higher the l e v e l of s e l f e f f i c a c y induced, the more e f f e c t i v e i s treatment. This contrasts to the treatment implic-ations derived from the three-systems model; greatest treatment e f f e c t i v e -ness r e s u l t s from implementing that treatment most appropriate for the i n d i v -idual's most f u n c t i o n a l l y relevant fear response(s) (Borkovec et a l . , 1977; Freedman et a l . , 1981). Bandura (1977) claimed that p a r t i c i p a n t modelling i s the most e f f e c t i v e fear reduction technique because enactive experiences are the strongest source of e f f i c a c y information. Symbolic modelling (based on v i c a r i o u s experience) and systematic d e s e n s i t i z a t i o n (based on arousal information) are considered equally l e s s e f f e c t i v e . However, Wolpe (1978) suggested, i n accordance with the three-systems model, that symbolic modelling and systematic d e s e n s i t i z a t i o n are probably d i f f e r e n t i a l l y e f f e c t i v e with d i f f e r e n t subjects possessing d i f f e r e n t response patterns. Hence, the negative and p o s i t i v e treatment e f f e c t s across a group of subjects probably cancel each other out. P a r t i c i p a n t modelling, however, focuses upon both arousal and behavioural response systems, i s therefore most appropriate for a wider range of subjects, and so appears to be more e f f e c t i v e . Bandura's theory f a i l s to consider the complexity of the i n t e r a c t i o n between i n d i v i d u a l response p r o f i l e and treatment. 43 Experimental Evidence Cox et a l ' s (1980) study did not support predictions from s e l f e f f i c a c y theory. They demonstrated that "courageous" (or, decorated) bomb disp o s a l operators performed laboratory stress-inducing tasks with s i g n i f i c a n t l y lower l e v e l s of p h y s i o l o g i c a l arousal then "noncourageous" (or, nondecorated) bomb disposal operators, though they reported the same degree of subjective fear. Aside from the causal issue of whether autonomic responding was red-uced by increased s e l f e f f i c a c y or by some other mechanism, t h i s appears to provide support for Bandura's p r e d i c t i o n that strong s e l f e f f i c a c y would be emp i r i c a l l y associated with reduced autonomic arousal. However, s e l f e f f i c -acy was measured i n r e l a t i o n to bomb disposal s k i l l s and not i n r e l a t i o n to the laboratory tasks. As s e l f e f f i c a c y i s supposedly very task s p e c i f i c , the degree of bomb disp o s a l related s e l f e f f i c a c y may not have been relevant to the laboratory task. More importantly, high s e l f e f f i c a c y for the task would have predicted concordance between the response systems, a l l i n d i c -ating r e l a t i v e l y low l e v e l s of anxiety. However, discordance was obtained. Self E f f i c a c y as an Hypothetical Construct Kozak and M i l l e r (in press) suggested that the three-systems model i s only a methodology for assessing fear, and lacks an hypothetical construct. Hypothetical constructs provide organizing p r i n c i p l e s for r e l a t i n g several s u p e r f i c i a l l y d i s s i m i l a r outcomes ( i . e . behavioural, verbal and autonomic), are not observed d i r e c t l y but are in f e r r e d from th e i r manifestations, and allow subsequent p r e d i c t i o n of r e l a t i o n s h i p s among the manifesting systems (Wiggins, 1973). Kozak and M i l l e r argued that Lang had equated fear as a construct with observable data, without s p e c i f y i n g p r i n c i p l e s for i d e n t i f y i n g from that data when an i n d i v i d u a l i s f e a r f u l , whereas, an hypothetical construct would enable one to account for the complexity of the data, to organize and integrate various response patterns across diverse i n d i v i d u a l s and s i t u a t i o n s . They claimed that even i f the systems do not covary, some-thing must be conceptually common across the systems as grouping them would be otherwise f r u i t l e s s . However, i t seems that the systems are grouped because of t h e i r r e l a t -ionships to stimulus conditions, and the nature of the stimulus conditions has served to guide research, for i t i s the basis upon which Hodgson and Rachman (1974) developed th e i r f i v e hypotheses. Therefore, the three-systems model could be viewed as a "response c l a s s e s " model (Wiggins, 1973) i n which the manifesting systems are viewed as signs of response classes rather than as signs of i n t e r n a l structures. Diverse a t t r i b u t e s can be considered mem-bers of a response c l a s s when i t can be demonstrated that they enter into some fun c t i o n a l r e l a t i o n s h i p with antecedent, concurrent and consequent stimulus conditons; the frequency, i n t e n s i t y or duration of a response c l a s s should be predictable from changes in those conditons (Wiggins, 1973). In spite of the previously c i t e d evidence and arguments regarding the c o n f l i c t s between predictions from s e l f e f f i c a c y theory and the three-systems model, Wilson (1978) has suggested that s e l f e f f i c a c y would be an appropriate hypothetical construct for the three-systems model. An examination of the r e l a t i o n s h i p between s e l f e f f i c a c y and the three response systems may be of some benefit i n t e s t i n g the usefulness of s e l f e f f i c a c y as an hypothetical construct. 45 PURPOSE OF THE STUDY S p e c i f i c predictions from s e l f e f f i c a c y theory and the three-systems model were analysed in t h i s study i n the context of solo piano performance anxiety. While both theories p r e d i c t more concordance among the response systems when intense emotional responses are e l i c i t e d , s e l f e f f i c a c y theory also predicts concordance under conditions associated with weak emotional responses. The three-systems model predicts more discordance under those conditions. Bandura has provided empirical support for a r e l a t i o n s h i p between s e l f e f f i c a c y and both behavioural and subjective indices of fear. However, the analogue nature of the research and the f a i l u r e to co n t r o l extraneous s i t -uational influences upon the behavioural and verbal response measures renders t h i s evidence questionnable. Also, further examination of the r e l a t i o n s h i p between autonomic responses and s e l f e f f i c a c y i s required. Therefore, t h i s study examined the r e l a t i o n s h i p of s e l f e f f i c a c y to each response system while attempting to c o n t r o l extraneous s i t u a t i o n a l influences. Useful i n f o r -mation regarding the s u i t a b i l i t y of s e l f e f f i c a c y as an hypothetical construct for the three-systems model was sought i n the process of testing their contrasting p r e d i c t i o n s . Further, the study provided a d d i t i o n a l data i n r e l a t i o n to Hodgson and Rachmans' (1974) hypotheses. Only two studies have provided empirical evidence for concordance amongst the response systems when intense emotional responses are e l i c i t e d . However, the Sartory et a l . (1977) study only examined subjective and autonomic responses, and also found that the concordance did not appreciably decrease as the emotional response weakened. The second study (Marks et a l . , 1971) only reported concordance among measures within the autonomic system. Therefore, Hodgson 46 and Rachmans' f i r s t hypothesis requires further supportive evidence. This study also attempted to assess concordance while attempting to minimise the e f f e c t s of measurement er r o r . S p e c i f i c a l l y , t h i s e n t a i l e d recording several p h y s i o l o g i c a l indices of the autonomic response, while attempting to c o n t r o l both the e f f e c t s of motor and cognitive demands of the performance task upon autonomic responding by including a neutral stimulus conditon, and extraneous influences upon behavioural and subjective measures. F i n a l l y , t h i s study provided a d d i t i o n a l data for the scales developed by Kendrick (1979) s p e c i f i c a l l y to measure musical performance anxiety. Experimental Hypotheses In accordance with the three-systems model, the following pattern of r e s u l t s was hypothesised; 1. (a) Evaluative conditions w i l l e l i c i t more intense emotional responses in the r e l a t i v e l y anxious group than nonevaluative conditions; (b) Evaluative conditions w i l l e l i c i t more intense emotional responses i n the r e l a t i v e l y anxious group than the r e l a t i v e l y nonanxious group; 2. (a) The behavioural, s e l f report and autonomic response systems w i l l demonstrate more synchronous changes in the r e l a t i v e l y anxious group than the r e l a t i v e l y nonanxious group; (b) In the r e l a t i v e l y anxious group, concordance w i l l be more evident under evaluative than under nonevaluative conditions; (c) Concordance w i l l be more evident in the r e l a t i v e l y anxious group under evaluative conditions than i n the r e l a t i v e l y nonanxious group under either evaluative or nonevaluative conditions; 3. (a) In the r e l a t i v e l y anxious group, s e l f e f f i c a c y w i l l be greater under nonevaluative than evaluative conditions; (b) S e l f e f f i c a c y w i l l be greater i n the r e l a t i v e l y nonanxious group under either evaluative or nonevaluative conditions, than i n the r e l a t i v e l y anxious group under evaluative conditions; (c) In the r e l a t i v e l y anxious group, c o r r e l a t i o n s between s e l f e f f i c a c y and each response system w i l l be greater under evaluative conditions than under nonevaluative conditions; (d) Correlations between s e l f e f f i c a c y and each response system w i l l be greater i n the r e l a t i v e l y anxious group under evaluative conditions, than in the r e l a t i v e l y nonanxious group under either evaluative or nonevaluative conditions. 4 8 METHOD Subjects Subjects were drawn from populations of piano students, contacted through either music programs at the University of B r i t i s h Columbia, Capilano College and King Edward College, or private music teachers around Vancouver. Students were judged by their respective teachers to have attained s u f f i c i e n t p r o f i c i e n c y for solo performance, and had performed solo p r i o r to contact. Students were approached either during cla s s e s , or by phone contact, at which point a basic r a t i o n a l e was given for the study, a screening questionnaire was completed, and those interested in p a r t i c i p a t i n g in the study were i d e n t i f i e d (see Appendix 1). The subject sample was selected on the basis of two c r i t e r i a . F i r s t l y , students were assigned either " r e l a t i v e l y anxious" or " r e l a t i v e l y non-anxious" status on the basis of a median s p l i t of screening questionnaire scores: scores of 11 or higher, and 10 or lower r e s p e c t i v e l y . Secondly, those a t t r i b u t e d " r e l a t i v e l y anxious" status must have also stated that anxiety during solo performances was a current source of concern, while " r e l a t i v e l y nonanxious" scorers stated that i t was not. It has been sug-gested by Bernstein (1973) , Borkovec et a l . (1974) and Borkovec and O'Brien (1976) that t h i s second c r i t e r i o n enhances the v a l i d i t y of subject s e l e c t i o n procedures. Measures r e l a t i n g to the aforementioned c r i t e r i a are presented in Appendix 2. Subjects were two groups of 20 piano students each. Ages ranged from 16 to 33 years; from 16 to 33 (M = 21.05, SD = 4.15) i n the r e l a t i v e l y anxious group (or group 1), and from 18 to 26 (M = 20.15, SD = 2.46) 49 in the r e l a t i v e l y nonanxious group (or group 2). There were 16 females and four males in group 1, and eight females and 12 males in group 2. Groups were matched on grade l e v e l (Toronto Conservatory standards), ranging from grade nine to A.R.C.T. Appendix 3 contains d e s c r i p t i v e s t a t i s t i c s of the above mentioned subject data. Experimental Setting The two performance assessments were conducted in a studio located in the Music Building at the Un i v e r s i t y of B r i t i s h Columbia. A l l subjects performed on a s i x - f o o t Bechstein grand piano. The studio also contained a video-camera, microphone, speaker and telemetry transmitters. A nearby room ("equipment room") contained a polygraph, FM telemetry receiver, videotape recording equipment and two o s c i l l o s c o p e s . Self Report Measures The Report of Confidence as a Performer Scale (RCP) - the screening questionnaire - consisted of 30 keyed t r u e / f a l s e statements describing autonomic, cognitive and behavioural manifestations of anxiety experienced during solo piano performances. Appel (1974) obtained a s p l i t h a l f r e l i a b i l i t y c o e f f i c i e n t of 0.94 from RCP scores. The S-R Inventory of General T r a i t Anxiousness (Endler and Okada, 1975) was adapted to the solo piano performance s i t u a t i o n to provide a concurrent v a l i d i t y measure for the RCP (see Appendix 4). Endler and Okada (1975) reported r e l i a b i l i t y c o e f f i c i e n t s ranging from 0.62 to 0.85. The inventory was administered immediately p r i o r to the f i r s t performance. Five s e l f report scales were administered in each performance session (Appendix 5). The State Scale (XI) of the S t a t e - T r a i t Anxiety Inventory (Speilberger, Gorsuch and Lushene, 1970) was administered approximately 15 minutes p r i o r to entering the studio (see Appendix 5A). The State 50 scale focuses upon q u a l i t i e s of tension, nervousness, worry and appre-hension e l i c i t e d by s p e c i f i c s i t u a t i o n s . Subjects respond to 20 s t a t e -ments to indicate their f e e l i n g s at that p a r t i c u l a r moment. Cronbach (1951) has reported an i n t e r n a l consistency range of 0.83 to 0.92 for t h i s scale, and empirical v a l i d i t y has been demonstrated (Lazarus and Opion, 1966). Also, Borkovec et a l . (1977) claim that t h i s scale i s not as susceptible to experimental demand influences as the frequently employed Fear Thermometer. Subjective Units of Distress Scales (SUDS) - zero to 100 - were used to guage the degree of discomfort experienced immediately p r i o r to and during performance (see Appendix 5B and 5C r e s p e c t i v e l y ) . The during performance SUDS were obtained for three periods; immediately after beginning, midway through, and j u s t p r i o r to completion of the performance. These ratings were provided r e t r o s p e c t i v e l y following completion of the performance. Two s e l f e f f i c a c y measures were administered approximately 10 minutes p r i o r to entering the studio. The Expectations of Personal E f f i c a c y for Musicians Scale (Kendrick, 1979) assesses p i a n i s t s ' expectations of being able to complete a number of tasks r e l a t i n g to piano performance (see Appendix 5D). Generalisation i s measured by s e l f e f f i c a c y ratings on f i v e d i f f e r e n t types of piano performance tasks. Two indices of s e l f e f f -icacy were derived from each subscale ( i . e . s p e c i f i c and general expectat-ions of personal e f f i c a c y ) . The f i r s t index was l e v e l of s e l f e f f i c a c y , or, the number of tasks subjects believed they could perform with some c e r t a i n t y . Certainty was the second index, r e f e r r e d to as the strength of s e l f e f f i c -acy. A strength score i s obtained for each l e v e l (task) from a seven point scale i n d i c a t i n g the degree of c e r t a i n t y with which the task could be 51 performed with anxiety under c o n t r o l . "With anxiety under c o n t r o l " was inc-luded as a c o n d i t i o n a l requirement i n consideration of the d i f f e r e n c e s be-ween Bandura's phobic populations and musicians. That i s , while most snake phobics would not touch a snake, p i a n i s t s have engaged i n most or a l l of the tasks l i s t e d i n the s e l f e f f i c a c y scale, a l b e i t with extreme l e v e l s of anxiety (Kendrick, 1 9 7 9 ) . Scale construction e n t a i l e d the establishment of consensual and concurrent v a l i d a t i o n , and a t e s t - r e t e s t r e l i a b i l i t y of . 8 8 . Kendrick's scale does not include items that represent the performance conditions in t h i s study. However, Bandura (1978) claims that i t i s e s s e n t i a l that s e l f e f f i c a c y measures be commensurate with the actual performance task. Therefore, a second s e l f e f f i c a c y measure was designed, and modified to s u i t the parameters of each performance condition (see Appendix 5E and 5F r e s p e c t i v e l y ) . The tasks (or s e l f e f f i c a c y level) were ranked upon two dimensions: degree df mastery and presence/absence of music sheets. These dimensions were derived from consultation with experienced p i a n i s t s . In both scales, s e l f e f f i c a c y l e v e l was calculated by summing the number of tasks subjects believed they could perform with a strength value of at l e a s t one. The Performance Anxiety S e l f Statement Scale (Kendrick, 1979) was administered immediately following each performance (see Appendix 5 H ) . This scale was s p e c i f i c a l l y developed to assess p i a n i s t s ' p o s i t i v e and negative thoughts before, during and after a performance. Scale construct-ion e n t a i l e d establishment of consensual, concurrent and construct v a l i d -a t i o n , and an i n t e r n a l consistency of 0 . 9 2 . As t h i s scale was designed to assess thoughts during performances before an audience, some items were s l i g h t l y modified to s u i t the nonaudience experimental condition i n t h i s study: items 7 , 1 5 , 26 and 28 (see Appendix 5 G ) . 52 Behavioural Measures Two behavioural measures were obtained for both performances. The f i r s t was a measure of performance effectiveness which e n t a i l e d 10 point sclaes of q u a l i t y of performance on each of the following dimensions: touch, phrasing, p i t c h and omission, rhythm, tempo, dynamics, memory and o v e r a l l effectiveness (see Appendix 6 for d e f i n i t i o n s of each term). This measure was a modification of the performance error count procedure used by Kendrick (1979). Performances were scored by two p i a n i s t s , both of whom held University degrees in music and A.R.C.T.s in piano perform-ance. They worked independently from copies of the music score and audiotape recordings of the music, and were b l i n d with respect to subjects' group membership and order of the performances. The performances were presented in a randomly counterbalanced order i n each group (the f i r s t performance was viewed f i r s t for ha l f the subjects in each group). Both performances of each subject were presented consecutively in order to allow a d i r e c t comparison of q u a l i t y to be made. The raters l i s t e n e d to each performance at l e a s t two times (ratings were checked during the second l i s t e n i n g ) . In addition the raters rated the amount of vigour with which each piece was played on a 10 point scale (from very l i t t l e to very much). Vigour was rated i n order to assess possible i n t e r a c t i o n s between auto-nomic measures and phy s i c a l movement. The performance effectiveness r a t i n g sheet i s presented i n Appendix 7. The second behavioural measure was a timed c h e c k l i s t of overt and observable i n d i c a t i o n s of performance anxiety. The c h e c k l i s t was developed by Kendrick (1979), and included the following items: knees tremble, hands tremble, l i f t s shoulders, s t i f f back and neck, s t i f f arms, face deadpan and moistens l i p s . Kendrick obtained i n t e r - r a t e r r e l i a b i l i t i e s between 0.8 and 1.0. The f i r s t two behaviours were omitted as they could not be c l e a r l y seen on the videotape. Two raters worked independently from video recordings of the performances, and were b l i n d with respect to the subjects' group membership and condition of performance, The performances in each group were presented i n a randomly counterbalanced order. Raters recorded the presence or absence of each target response within each of the following three 20 second i n t e r v a l s : immediately following the f i r s t note, midway through the performance, and immediately preceding the f i n a l note (see Appendix 8 for d e f i n i t i o n s of items and r a t i n g sheet). Raters were trained to a r e l i a b i l i t y l e v e l of 0.83. Autonomic Measures Three measures of autonomic arousal were c o l l e c t e d , using a telemetry device which allowed continuous recording before, during and after perform-ances. Skin resistance l e v e l and skin resistance response (amplitude, latency and r i s e time) were measured and converted to conductance units (Hassett, 1978). Amplitude was defined as — - ~ ~ , where SRT was skin S R 2 SR}_ resistance at the point before onset of the skin resistance response, and SR 2 was skin resistance at the peak of the response. Latency was defined as the i n t e r v a l between the onset of the tone and s t a r t of the response. Rise time was defined as the i n t e r v a l between the s t a r t of the response and the peak of the response (Venables and C h r i s t i e , 1980). Active eccrine gland s i t e s are most concentrated on palmar and plantar surfaces (Venables and C h r i s t i e , 1980), however, electrode placements i n these regions would i n t e r f e r e with musical performance. Therefore, the medial hypomall-eolar are - "the medial side of the foot over the abductor h a l l u c i s muscle adjacent to the plantar surface and midway between the f i r s t phalange and a point d i r e c t l y beneath the ankle" (Venables and C h r i s t i e , 1980) -was used. Rickles and Day (1968) found that t h i s s i t e provided comparable data to that of palmar placements except under low arousal conditions. A 54 two second tone (100 decibels, 1000 Hz) e l i c i t e d skin resistance responses. The tone was presented on three occasions, one minute p r i o r to the si g n a l to commence playing (tone 1), the s i g n a l to commence (tone 2), and one min-ute following the s i g n a l to commence (tone 3). Analyses were performed on responses to tones 1 and 3 only; responses to tone 2 would have been con-founded by movement a r t i f a c t r e s u l t i n g from commencement of playing. Tones 1 and 3 were chosen to provide measurements of skin resistance response both p r i o r to and during playing. Skin resistance l e v e l was recorded at three points; one minute p r i o r to the f i r s t note (just before tone 1), one minute following the f i r s t note (just before tone 3), and immediately p r i o r to the f i n a l note. An attempt to measure maximum and minimum responses was made i n order to ca l c u l a t e range-corrected heart rate and skin resistance l e v e l scores. Range cor r e c t i o n i s calculated by expressing each response as a r a t i o of a p a r t i c -ular subject's range of responsiveness: — X m i ? , where Xo = the observed Xmax - Xmin response (Hassett, 1978). Range cor r e c t i o n has been used to corre c t for i n d i v i d u a l differences i n response l a b i l i t y (Lykken, Rose, Luther and Maley, 1966; Lykken, 1972). The maximum value was assumed to be the subject's highest heart rate (over a 10 second sample) or highest l e v e l of skin conduc-tance (that did not represent the peak of a response) recorded throughout both assessment sessions. The minimum value was assumed to be those l e v e l s of skin conductance and heart rate recorded af t e r a period of r e l -axing, or any lower l e v e l s recorded throughout both assessment sessions. Twenty percent of the data points for each autonomic measure were randomly selected from each performance condition for the purposes of r e l -i a b i l i t y assessment. The r e l i a b i l i t y rater was b l i n d with respect to the experimental conditions and group membership of the subjects. 55 Heart rate was sampled over 10 second i n t e r v a l s . These began: 100 seconds p r i o r to the f i r s t note; 40 seconds p r i o r to the f i r s t note; 10 seconds p r i o r to the f i r s t note; 120 seconds past the f i r s t note; 180 seconds past the f i r s t note; 240 seconds past the f i r s t note; and 300 sec-onds past the f i r s t note. Measurements were not taken from the i n t e r v a l beginning 60 seconds past the f i r s t note to avoid confound with response to tone 3. In following the procedure used by Kendrick (1979), tachometer out-put ( i . e . beat to beat intervals) was averaged within each 10 second i n t e r v a l . Respiration was measured by d i v i d i n g the number of peaks occurring within an i n t e r v a l by the length of that i n t e r v a l , to provide a measure of breaths per minute. Two i n t e r v a l s were used: the duration from the i n i t -i a t i o n of recording to tone 1, and the duration of the musical piece. a I t was assumed that such extraneous factors as d i e t , a c t i v i t y l e v e l p r i o r to assessment sessions, current health, alcohol intake and so on, that may influence autonomic response, were c o n t r o l l e d by randomisation procedures. Apparatus and Equipment, An RCA black and white, closed c i r c u i t camera (model number TG2011/N), RCA video cassette recorder (model number VET 650) , and VHS tapes were used to record subjects' performances. In addition, an RCA Time-Date Generator (model number 1440B/02), attached to the video recording unit, displayed minutes, seconds and hundredths of a second within a 24 minute readout. Tapes were played back on an RCA 19 inch portable t e l e v i s i o n monitor (model number JD 975 WV) for v i s u a l r a t i n g purposes. Audio ratings were made from the video cassette u n i t . Sine wave tones (1 kHz) were generated v i a a tone generator, and amplified by a Bogen Acousta Master amplifier ( 100 W, model number CT 100). The tone was broadcast to the subjects v i a a Cerwin Vega speaker (model 56 number V-30), and was c a l i b r a t e d using a T r i p l e t t sound l e v e l meter (model number 375) at the point of the subject's head. An EKEG E l e c t r o n i c s Co. Ltd. FM telemetry system broadcast heart rate, r e s p i r a t i o n and skin resistance data from the subject i n the studio to a Grass Model 7 polygraph. Respiration and skin resistance were converted to d.c. voltages, which were amplified and used to frequency modulate 4000 Hz and 900 Hz c a r r i e r signals r e s p e c t i v e l y . The c a r r i e r signals were mixed to frequency modulate a 90 MHz transmitter, which was then demodulated i n the telemetry r e c e i v e r . The output si g n a l s to the polygraph were voltages prop-o r t i o n a l to the subject's r e s p i r a t i o n and skin resistance. Heart rate data was sent on a second transmitter (900 Hz c a r r i e r ) . Two Cathode Ray O s c i l l -oscopes were connected to each FM receiver to f a c i l i t a t e the accurate tuning of each re c e i v e r . Respiration was measured by a chest circumference s t r a i n guage contain-ing coconut carbon. The r e s p i r a t i o n telemetric system sent output to a 7P1 preamplifier on a Model 7 Grass Polygraph. Two Med Associates nonpolarizing s i l v e r - s i l v e r c hloride electrodes (model TDE-20, h a l f inch sensor diameter) were placed on the hypomalleolar area of that foot used l e a s t whilst p l a y i n g . The skin was f i r s t wiped clean with water. Electrode paste consisted of a mixture of one part p h y s i o l o g i c a l s a l i n e to two parts Unibase. This produces, a paste with a sodium chloride concentration comparable to that found i n sweat (0.05-0.75 M NaCl) thus minimising the p o s s i b i l i t y . o f sweat d i f f u s i n g into the e l e c t r o l y t e and chang-ing the c h l o r i d e concentration (Fowles, C h r i s t i e , Edelberg, Grings, Lykken and Venables, 1981). Double sided adhesive electrode c o l l a r s (Med Associates, TD-20) were used to ensure c o n t r o l of skin contact area; 92 mm2. A cotton sock was placed over the foot to maintain warmth. The skin resistance telem-57 e t r i c system sent output to a 7P1 preamplifier on a Grass Model 7 Polygraph. Two electrodes of the same d e s c r i p t i o n were attached to the l e f t shin and ri g h t c o l l a r bone, using Beckman electrode paste, a f t e r the skin had been f i r s t abraded with Hewlett-Packard Redux paste. P i l o t t r i a l s indicated that these placements were associated with the l e a s t movement a r t i f a c t . The E.K.E.G. telemetric system provided output to a 7P4A tachograph preamplifier on a Grass Model 7 Polygraph. Ganz Gehartet c a l i p e r s were used to measure i n t e r v a l s on the polygraph chart paper (accurate to 0.05 mm u n i t s ) . Assessment Conditions A l l subjects performed i n two conditions i n the following order: (a) Performance alone (nonevaluative) - subjects performed a musical piece in the studio with no-one else present. In an attempt to reduce anxiety, subjects were informed that t h i s f i r s t session was for purp-oses of pr a c t i c e and adaptation only, and were not made aware of the video camera. Subjects were debriefed following completion of the perf-ormance . (b) Performance before an audience (evaluative) - subjects performed the same musical piece before an audience of f i v e judges whom the subjects believed were evaluating t h e i r performance. The audience were seated i n a ha l f c i r c l e around the subject. Subjects were also made aware of the video camera. This condition occurred one week afte r the f i r s t . The audience was comprised mostly of graduate U n i v e r s i t y students, some of whom were musicians. The audience's age range was comparable to that of the subjects'. They were introduced as being "accomplished p i a n i s t s and experts i n behavioural assessment who have copies of your music and w i l l 58 be evaluating your performance". The audience was provided with music sheets and required to write notes on four occasions during each performance, maintain a neutral to f r i e n d l y f a c i a l expression, and to make no reaction to the tones. In addition, the audience was asked to check those behaviours on the Timed Behavioural C h e c k l i s t which were present at any time during the performance (see Appendix 9). This task was undertaken to j u s t i f y the audience's presence as behavioural assessors, and the data were not analysed. Procedure Sel e c t i o n of Subjects Approximately 120 p i a n i s t s completed the RCP. Each person s a t i s f y i n g the c r i t e r i a for i n c l u s i o n was phone contacted either by the researcher (R) or research a s s i s t a n t (RA), and asked i f they would be interested i n p a r t i c i p a t i n g i n the study (see Appendix 10). Subjects were t o l d they would be required to perform on two occasions; once alone and once before a small group of people. Subjects were also offered payment of f i v e d o l l a r s per session. Those who indicated i n t e r e s t were then t o l d to choose a piece of music, three to f i v e minutes i n length, which they had mastered and f e l t comfortable playing from memory. F i n a l l y , subjects were t o l d they would be c a l l e d again to arrange dates and times. Subjects were phone contacted again within two weeks and assigned times and dates. The t i t l e and grade l e v e l of chosen pieces were ascert-ained (the subject data sheet i s presented i n Appendix 11). Subjects were asked to bring a copy of the i r music to the f i r s t session with any d e l i b e r a t e changes i n tempo or dynamics marked on i t . Electrode placements were also described to subjects, and they were asked to r e f r a i n from wearing c l o t h i n g that would hinder electrode placement (e.g., high c o l l a r s etc.) Grade l e v e l s of the music pieces were then checked against the Toronto Conservatory of Music l i s t i n g to ensure that subjects had 59 chosen pieces at the i r grade l e v e l of p r o f i c i e n c y . In the week p r i o r to the f i r s t session, a l l subjects were reminded of dates, times, l o c a t i o n , music copies and cl o t h i n g by mail. F i f t y f i v e p i a n i s t s were selected for p a r t i c i p a t i o n i n the study. One person was randomly selected (within the constraints of maintaining a matched group design) to be dropped from the analysis i n order to produce equal group s i z e s . Four people completed only one assessment session. Ten people withdrew from the study p r i o r to the f i r s t session, a f t e r having v o l -unteered. Alone Condition (Condition 1) Performances were scheduled every 15 minutes (see Appendix 12). Subjects a r r i v e d 20 minutes p r i o r to t h e i r performance time and were greeted by RA^. RAi c o l l e c t e d the subject's music copy, p l a c -ing i t in that student's envelope, and read a set of in s t r u c t i o n s which described the procedure and emphasised the pr a c t i c e nature of t h i s session (these are presented i n Appendix 13). The subjects were informed that the purpose of t h i s session was to allow them to become f a m i l i a r with the s e t t i n g and procedures, and that evaluations would not be made. RAj_ then attached one heart rate electrode to the r i g h t c o l l a r bone. While RA^ attached remaining electrodes, subjects read and signed a consent form (see Appendix 14), and completed the S-R Inventory of General T r a i t Anxious-ness, the State Anxiety Scale, the Expectations of Personal E f f i c a c y Scales, the S e l f E f f i c a c y Scale and the Sud Scale. Subjects were instructed that " f i r s t impression" responses were most appropriate. RA^ then placed the completed questionnaires i n the subject's envelope, which was then handed to Subjects were taken to the piano (in studio) by R who attached the electrodes to the transmitters. R then t o l d subjects to: Announce your name and the t i t l e of your piece when I leave the room. Then, s i t very s t i l l u n t i l you hear the second tone, when you 60 should begin to play. I ' l l return when you f i n i s h to unhook you. After returning to the equipment room, R began the stopwatch, i n i t i a t e d polygraphic recording and started video equipment. Three minutes af t e r leaving the studio, R generated the f i r s t tone. A second tone was generated following a further 60 seconds, to s i g n a l subjects to begin playing. A t h i r d tone was generated following a further 60 seconds. The researcher recorded a time-date generator point on the polygraph chart paper, i n addit-ion to the onset of each tone and the beginning and end of the piece. Exact tone onsets were obtained by viewing the time-date generator from the video-tapes at a l a t e r date. Following completion of the performance, R returned to the studio, and informed the subject: Now I need to obtain an estimate of the range of your physiolog-i c a l responding, because everyone d i f f e r s i n the amount by which th e i r l e v e l of arousal can change. To do t h i s , I ' l l ask you to s i t q u i e t l y for a few minutes with your eyes closed and j u s t t r y to re l a x . I ' l l return i n a few moments. R then returned to the polygraph and monitored autonomic response for three to f i v e moments of re l a x a t i o n . R then returned to the studio, unhooked the subject, and walked the subject back to the equipment room where he/she was met by RA2. Subjects completed the SUD and Se l f Statement scales while RA2 removed and cleaned the electrodes. Completed scales were placed i n the subject's envelope. RA2 then"read the subject a set of i n s t -ructions which debriefed subjects about the nature of t h i s performance session (see Appendix 15). Subjects were informed that the i n i t i a l set of in s t r u c t i o n s was intended to reduce tension as much as possible but that, i n f a c t , t h e i r performances had been evaluated. They were also given the opt-to request that the videorecordings be erased. 61 The subject then read and signed a second consent form (see Appendix 16). RA2 Paid t n e subject and reminded him/her to r e f r a i n from perfecting t h e i r piece before the next session. Audience Condition (Condition 2) Subjects were again instructed to a r r i v e 20 minutes p r i o r to t h e i r performance, and were met by RA^ w n o read a set of i n s t r u c t i o n s , describing the evaluative role of the audience and the presence of a camera (see Appendix 17). The subjects were informed that the audience were experts who would be evaluating t h e i r performance which would al s o be videotaped. After attaching an electrode to the r i g h t c o l l a r bone, subjects were given a video questionnaire which asked them to state whether they had been "aware", "suspicious" or "unaware" of the video camera i n the previous session (see Appendix 18). Subjects then completed the State Anxiety Inventory, two s e l f e f f i c a c y scales and SUD scale whilst RA^ attached the remaining electrodes. U n t i l completion of the performance, the procedure used i n Condition 1 was followed except that the audience was present as subjects entered the studio. R then unhooked the subjects and walked them back to the equipment room where they were met by RA2. Subjects completed the Retrospective SUD and Self Statement scales while RA2 removed and cleaned the electrodes. RA2 then thanked the subjects for p a r t i c i p a t i n g , paid them f i v e d o l l a r s and informed them that they would be contacted, by mail, regarding the r e s u l t s . Debriefing Several months following the l a s t experimental session, subjects were debriefed by mail contact. The l e t t e r (presented in Appendix 20) con-tianed the following information: purpose of the research, nature of the audience and personal feedback. Design and S t a t i s t i c a l Analyses The experiment was a 2 x 2, mixed model design. The two independent groups formed the l e v e l s of a between f a c t o r , while the two conditions formed the l e v e l s of a within f a c t o r . Several s t a t i s t c i a l packages were used i n the analysis of the data. These included the S t a t i s t i c a l Package for the S o c i a l Sciences (Nie, H u l l , Jenkins, Steinbrenner and Bent, 1970) , BMDP (Dixon, 1981) and MIDAS ( Foa and Guire, 1976). RESULTS Summary of Findings The performance-conditions and the subject-groupings e f f e c t s were analysed through group mean comparisons. The hypotheses stated that r e l a t i v e l y anxious p i a n i s t s would be most affected by the presence of an audience. This p r e d i c t i o n was supported by the r e s u l t s . The audience condition e l i c i t e d s e l f report and autonomic changes i n that group on the following variables; state anxiety, mean s e l f e f f i c a c y strength, t o t a l SUDS and heart rate whilst playing. In contrast, the audience condition only produced heart rate changes in the r e l a t i v e l y nonanxious group. Moreover, the audience condition affected performance q u a l i t y to a greater extent in the r e l a t i v e l y anxious group. Therefore, as predicted, group mean diffe r e n c e s were generally larger in the audience condition. Analyses of extreme scorers on the i n i t i a l screening index (most anxious and l e a s t anxious p i a n i s t s ) enhanced those differences on the behavioural and s e l f report measures. It was also hypothesised that a l l changes occurring for r e l a t i v e l y anxious (and most anxious) p i a n i s t s would r e f l e c t increased anxiety. The r e s u l t s were also consistent with t h i s hypothesis: performance accomp-lishment diminished, subjective d i s t r e s s increased, state anxiety increased, s e l f e f f i c a c y decreased and heart rate elevated. Trends that did not reach s i g n i f i c a n c e were also suggestive of increased anxiety; more overt signs of anxiety tended to be observed, and more negative s e l f statements tended to be reported. Measures on which trends were not observed included skin conductance and r e s p i r a t i o n . That i s , of the 13 dependent va r i a b l e s , 64 f i v e r e f l e c t e d s i g n i f i c a n t l y increased anxiety, two r e f l e c t e d trends towards increased anxiety, and s i x f a i l e d to r e f l e c t any systematic resp-onse to the audience condition. In contrast, the dependent measures r e f l -ected either reduced anxiety (performance accomplishment increased), or increased anxiety (subjective d i s t r e s s and heart rate increased), or no change in r e l a t i v e l y nonanxious (and l e a s t anxious) p i a n i s t s . Concordance between the dependent measures recorded during performances was hypothesised to be more evident i n the r e l a t i v e l y anxious group when playing before an audience. That i s , the measures were expected to r e f l e c t comparable l e v e l s of anxiety in any given subject. Though few s i g n i f i c a n c t c o r r e l a t i o n s were obtained, those that did emerge supported that hypothesis. Concordance was obtained among the following r e l a t i o n s h i p s in the audience condition for r e l a t i v e l y anxious p i a n i s t s ; performance q u a l i t y and overt signs of anxiety (approached s i g n i f i c a n c e ) , s e l f statements and retrospective SUDS, r e s p i r a t i o n and skin conductance l e v e l , overt signs of anxiety and retrospecitve SUDS, heart rate and retrospective SUDS (approached s i g n i f -icance) , and heart rate and s e l f statements. That i s , concordance was obtained among dependent measures both within and between the response systems. Only two concordant r e l a t i o n s h i p s were observed in the r e l a t i v e l y nonanxious group, one i n each performance condition. F i n a l l y , s e l f e f f i c a c y theory predictions were not generally supported. R e l a t i v e l y anxious p i a n i s t s were expected to show reduced s e l f e f f i c a c y in the audience condition as compared to t h e i r ratings in the alone condition and to the ratings of r e l a t i v e l y nonanxious p i a n i s t s . These predictions were supported by the mean s e l f e f f i c a c y strength v a r i a b l e , which represented an average strength score for the three measures of s e l f e f f i c a c y judgements. However, only two of the s i x indices of s e l f e f f i c a c y demonstrated group d i f f e r e n c e s , and only one of those (strength of general expectations of 65 personal e f f i c a c y ) was influenced by the audience condition. Conviction ratings of a b i l i t y to perform the p a r t i c u l a r experimental tasks also f a i l e d to d i f f e r e n t i a t e the performance conditions in either group. Moreover, i n agreement with the three-systems model, s e l f e f f i c a c y only correlated with one of the three response systems (self r e p o r t ) . D e t a i l s of a l l of the analyses are described below. Analyses to evaluate the i n t e r a c t i o n between preliminary subject variables (age and sex) and dependent measures are presented f i r s t , follow-ed by examination of the v a l i d i t y of the i n i t i a l screening index. Interrater r e l i a b i l i t i e s and r e l a t i o n s h i p s among various dependent measures and the vigour with which the piece was played are then presented. The major anal-yses of the study include an evaluation of group means across and within performance conditions, and examination of c o r r e l a t i o n s among dependent measures. Due to instrument f a i l u r e , r e s p i r a t i o n data were not a v a i l a b l e for six subjects (three i n each group) i n the alone condition, and s i x d i f f e r e n t subjects (three in each group) i n the audience condition. Skin conductance data were also missing for one subject in each condition for s i m i l a r reasons. Missing data were replaced with unweighted means of the relevant group's performance in either the alone or audience condition. Range co r r e c t i o n of heart rate and skin conductance l e v e l data was deemed appropriate from the large amount of i n d i v i d u a l v a r i a b i l i t y on those measures. Range c o r r e c t i o n e f f e c t i v e l y replaces data that are confounded by i n d i v i d u a l p h y s i o l o g i c a l differences with scores that r e f l e c t the r e l a t i v e status of i n d i v i d u a l s with respect to the construct of i n t e r e s t ( i . e . anx-iety) (Lang, 1971). In accordance with Lang's suggestion, analyses were performed on both range corrected and absolute values of the autonomic 66 system. However, only range corrected values were presented i n t h i s section as the same pattern of r e s u l t s emerged from both types of a n a l y s i s . Absolute heart rate and skin conductance l e v e l means and standard deviations are presented i n Appendix 21. Preliminary Variables In order to assess the comparability of the r e l a t i v e l y anxious and r e l a t i v e l y nonanxious groups on uncontrolled subject v a r i a b l e s , age and sex d i s t r i b u t i o n s were examined. A one-way between groups analysis of variance showed that the mean age of subjects i n each group did not s i g n i f -i c a n t l y d i f f e r . However, a chi-square analysis showed that sex was unequally d i s t r i b u t e d across the two groups; %{2) = 6.8, p_ < .01. One-way between groups analyses of variance were performed to i d e n t i f y the r e l a t i o n s h i p s between sex and each dependent v a r i a b l e . Means, standard deviations and F values are presented i n Table I. From those variables i n d i c a t i n g s i g n i f i c a n t sex d i f f e r e n c e s , examination of means revealed that females reported lower strength of s p e c i f i c personal e f f i c a c y (SStrength), more frequent negative s e l f statements (SS) and lower skin conductance l e v e l at tone 1 (SCL1) when playing alone. When playing before an audience, females reported higher l e v e l s of state anxiety (State), lower strength values on a l l measures of s e l f e f f i c a c y , more frequent negative s e l f statements, higher l e v e l s of retrospective subjective units of d i s t r e s s (RSUDS) and lower skin conductance l e v e l at each recording period. However, sex could not be included as a grouping factor i n subsequent analyses, as some c e l l s would be too small for v a l i d s t a t i s t i c a l procedures. However, sex in t e r a c t i o n s were considered i n the o v e r a l l i n t e r p r e t a t i o n of the r e s u l t s as sex as a factor appeared to account for at l e a s t part of the variance i n several dependent v a r i a b l e s . 67 Measure Table 1 Means, Standard Deviations and F Values for Males and Females on each Dependent Variable Male (n = 16) M SD Female (n = 24) M SD PQ Alone 39.56 16.03 Audience 38.13 16.85 TC Alone 3.28 1.98 Audience 3.53 2.54 State Alone 37.00 11.90 Audience 36.56 11.22 SLevel Alone 4.69 0.70 Audience 4.88 0.34 SStrength Alone 21.44 5.02 Audience 21.63 5.70 GLevel Alone 4.94 0.25 Audience 4.94 0.25 GStrength Alone 20.56 6.46 Audience 20.56 6.72 Level Alone 4.50 0.73 Audience 4.56 0.63 Strength Alone 18.44 6.14 Audience 19.31 5.74 SUDS Alone 35.63 22.50 Audience 41.25 25.00 RSUDS Alone 106.81 71.09 Audience 120.44 93.48 SS Alone 91.63 43.54 Audience 100.31 61.63 SCL1 (^mhos) Alone 7.28 3.55 Audience 10.00 5.69 Amplitudel (yumhos) Alone 2.22 2.13 Audience 1.60 1.01 Latencyl (sees) Alone 3.75 Audience 2.51 Risetimel (sees) Alone 2.86 Audience 2.75 35.92 36.11 3.33 3.54 38.75 45.75 4.42 4.42 17.38 17.25 4.50 4.38 17.08 15.63 4.21 4.25 16.96 15.29 40.42 54.58 152.92 188.96 138.63 143.58 12.58 12.63 01 03 42 04-,80 ,46 ,71 ,24 ,96 ,87 ,74 .23 1. 1. 38 87 9.04 11.13 0.78 0.97 6.31 6.67 0.89 1.13 8.84 7.94 0.93 1.19 5.89 5.65 24.04 26.04 82.34 87.64 62.87 64.38 2.69 3.06 1.45 1.71 1.49 1.34 1.51 1.37 0.65 0.42 0.92 0.99 0.28 6.50 ,26 ,25 ,65 ,62 ,67 ,78 ,82 ,18 ,11 ,93 0.59 4.81 0.40 2.60 3.35 5.57 6.75 4.49 2.25 10.75 0.82 0.57 2.03 0.84 3.37 0.80 ns ns ns ns ns < .02 ns ns < .05 < .05 ns ns ns < .05 ns ns ns < .05 ns ns ns < .05 < .02 < .05 ns < .01 ns ns ns ns ns ns Table 1 (cont'd) Male (n = 16) Female (n = 24) Measure M SD M SD F E SCL3 (//mhos) Alone 10.63 5.47 6.84 3.49 7.17 < .02 Audience 15.04 10.50 7.02 4.90 10.64 < .01 Amplitude3 (/(mhos) Alone 0.82 0.79 0.78 0.90 0.03 ns Audience 1.57 1.55 0.78 1.08 3.58 ns Latency3 (sees) Alone 2.00 1.39 2.09 0.99 0.07 ns Audience 1.89 1.09 1.96 1.00 0.04 ns Risetime3 (sees) Alone 3.04 2.06 2.29 1.46 1.84 ns Audience 1.98 0.96 2.30 1.05 0.96 ns SCLE yumhos) Alone 11.19 10.77 6.59 3.28 3.88 ns Audience 15.53 11.96 6.72 4.15 11.14 < .01 Resp-pre Alone 16.77 2.34 16.45 2.65 0.15 ns Audience 15.78 2.45 17.98 2.31 8.34 < .01 Resp-during Alone 21.20 3.05 22.54 3.93 1.33 ns Audience 21.63 3.70 23.77 3.44 3.51 ns HR-pre Alone 94.85 12.74 92.13 16.98 0.03 ns Audience 109.50 20.83 102.31 20.68 1.16 ns HR-during Alone 108.15 17.11 100.62 18.23 1.72 ns Audience 126.31 22.76 121.04 21.84 0.54 ns PQ = Performance Quality; TC = Timed Che c k l i s t of Observable Signs of Anxiety; State = State Anxiety Scale; SLevel = Level of S p e c i f i c Expectations of Personal E f f i c a c y (Kendrick); SStrength = Strength of S p e c i f i c Expectations of Personal E f f i c a c y (Kendrick); GLevel = Level of General Expectations of Personal E f f i c a c y (Kendrick); GStrength = Strength of General Expectations of Personal E f f i c a c y (Kendrick); Level = Level of Self E f f i c a c y ; Strength = Strength of Self E f f i c a c y ; SUDS = Subjective Units of Dist r e s s ; RSUDS = Retrospective Subjective Units of Di s t r e s s ; SS = Self Statement Scale; SCL1/3 = Skin conductance l e v e l p r i o r to tone 1/3; Amplitudel/3 = Skin conductance response amplitude at tone 1/3; Latencyl/3 = Skin conductance response latency at tone 1/3; Risetimel/3 = Skin conductance response risetime at tone 1/3; SCLE = Skin conductance l e v e l at completion of the performance; Resp-pre = Respiration rate p r i o r to tone 1; Resp-during = Respiration rate during performance; HR-pre = Heart rate p r i o r to tone 1; HR-dur-ing = Heart rate during performance. 69 V a l i d i t y of the Screening Index The General T r a i t Anxiousness Scale adapted for piano performance was included as a backup measure for the Report of Confidence as a Performer Scale (RCP). The s i g n i f i c a n t Pearson product moment c o r r e l a t i o n , r_ = .71, between these two measures provided some concurrent empirical v a l i d a t i o n for the i d e n t i f i c a t i o n of r e l a t i v e l y anxious and r e l a t i v e l y nonanxious p i a n i s t s on the basis of RCP scores. A one-way between groups analysis of variance also indicated that r e l a t i v e l y anxious p i a n i s t s reported s i g n i f i c a n t l y higher l e v e l s of general t r a i t anxiousness (M = 33.1, SD = 7.15) than r e l a t i v e l y nonanxious p i a n i s t s (M = 23.7, SD = 5.45); F(l,38) = 21.86, £ < .001. Interrater R e l i a b i l i t i e s Interrater r e l i a b i l i t i e s were calculated separately for the alone and audience conditions for each of the nine rated v a r i a b l e s : performance q u a l i t y , vigour, observable signs of anxiety, skin conductance l e v e l , skin conductance response amplitude, latency and risetime, r e s p i r a t i o n (prior to and during performance) and heart rate (prior to and during performance). Observer agreement was calculated using Pearson product moment c o r r -e l a t i o n s , which are presented i n Table I I . R e l i a b i l i t i e s a l l surpassed the acceptable l e v e l of .80 (Kazdin, 1980). Phi c o r r e l a t i o n s were also calculated for each item on the Timed Ch e c k l i s t of Observable Signs of Anxiety to assess w i t h i n - i n t e r v a l rater agreements. These are presented i n Table I I I . The analysis showed that, on some v a r i a b l e s , the w i t h i n - i n t e r v a l agreement c o e f f i c i e n t s were not as high as the o v e r a l l i n t e r r a t e r c o e f f i c i e n t . Table II Interrater R e l i a b i l i t i e s Assessed by Pearson Correlations Variable Alone (r) Audience (r) PQ 0.84 0.89 Vigour 0.86 0.89 TC 0.92 0.95 SCL 0.99 0.99 Amplitude 0.95 0.95 Latency 0.99 0.96 Risetime 0.93 0.92 Respiration-pre 0.99 0.98 Respiration-during 0.90 0.88 HR-pre 0.99 0.99 HR-during 0.99 0.99 Table III Phi-Correlations for the Timed Checklist of Observable Signs of Anxiety Alone Audience Item L i f t s Shoulders S t i f f Back and Neck S t i f f Arms Face Deadpan Moistens Lips Interval 1 Interval 2 Interval 3 Interval 1 Int e r v a l 2 Interval 3 0.69 0.65 0.72 0.86 1.00 0.92 0.72 0.69 0.75 1.00 1.00 1.00 1.00 0.95 1.00 0.67 1.00 1.00 0.92 1.00 0.78 0.88 0.90 0.92 1.00 0.69 0.64 0.79 0.83 1.00 72 'Vigour', 'Video' and 'Respiration' Design-Control Measures In order to assess for possible confound between the amount of vigour with which each piece was played, and autonomic responding, separate Pearson product moment c o r r e l a t i o n s were calculated between vigour and r e s p i r a t i o n , heart rate and skin conductance l e v e l s that were obtained whilst subjects were performing. These are presented i n Table IV. The c o e f f i c i e n t s were generally small, and therefore vigour was not included as a covariate i n the subsequent analyses. A chi-square analysis f a i l e d to reveal s i g n i f i c a n t group dif f e r e n c e s regarding the degree of awareness of the videocamera when playing alone; X (1) = 2.35, ns. Frequencies are presented i n Table V. Respiration was measured to assess the degree to which i t might confound heart rate data. However, the obtained Pearson product moment co r r e l a t i o n s (presented in Table VI) were generally weak. Therefore, r e s p i r a t i o n was not included as a covariate in subsequent analyses. Group Means Analyses In order to reduce the p r o b a b i l i t y of making Type 1 error while assessing group mean di f f e r e n c e s across and within conditions, separate multivariate analyses over repeated measures were performed on four sets of dependent measures: behavioural; s e l f report; skin conductance; and heart rate and r e s p i r a t i o n . As mentioned, range corrected skin conductance l e v e l and heart rate analyses are presented (absolute values are presented i n Appendix 21). Subjects demonstrated considerable v a r i a b i l i t y i n maximum and minimum values of skin conductance l e v e l , p a r t i c u l a r l y within the r e l a t i v e l y nonanxious group. For maximum values; r e l a t i v e l y anxious group, M = 9.42, SD = 4.67; r e l a t i v e l y nonanxious group, M =15.52, SD = 12.87. For minimum values; r e l a t i v e l y anxious group, M = 5.30, Table IV Pearson C o r r e l a t i o n C o e f f i c i e n t s Between Vigour and Heart Rate, Respiration and Skin Conductance Variables R e l a t i v e l y Anxious R e l a t i v e l y Nonanxious (£) (r) Resp-during Alone 0.09 -0.31 Audience 0.24 0.05 HR-during Alone 0.28 0.17 Audience 0.09 0.20 SCL3 Alone -0.12 0.04 Audience 0.21 0.29 SCLE Alone -0.03 -0.05 Audience 0.17 0.21 *p < .05 Table V Group Frequencies of Reported Awareness, Suspicion and Unawareness of the Camera R e l a t i v e l y Anxious R e l a t i v e l y Nonanxious (n = 2 0 ) (n = 2 0 ) Aware Suspicious Unaware 7 1 2 Table VI Pearson Product Moment C o e f f i c i e n t s Between Respiration Rate and Heart Rate R e l a t i v e l y Nonanxious <£> -0.10 -0.03 Resp-dur ing,HR-dur ing Alone 0.15 0.40 Audience 0.12 0.07 Variables R e l a t i v e l y Anxious (r) Resp-pre,HR-pre Alone -0.46* Audience -0.28 SD = 2.53; r e l a t i v e l y nonanxious group, M = 4.95, SD = 2.90. Within group v a r i a b i l i t y was also high i n heart rate data. For maximum values; r e l a t i v e -l y anxious group, M = 130.41, SD = 15.73; r e l a t i v e l y nonanxious group, M = 126.02, SD = 26.62. For minimum values; r e l a t i v e l y anxious group, M = 82.26, SD = 13.23; r e l a t i v e l y nonanxious group, M = 75.79, SD = 12.63. Means and standard deviations for each dependent measure are presented i n Table VII. Behavioural Measures These included performance q u a l i t y (on which scores could range from 0 to 80) and overt signs of anxiety (which could range from 0 to 15). No s i g n i f i c a n t main e f f e c t s emerged from the multivariate analysis of behavioural measures. However, the obtained F value for the group x condition i n t e r a c t i o n was s i g n i f i c a n t ; F(2,37) = 3.45, £ < .05. The behaviour of r e l a t i v e l y anxious p i a n i s t s d i f f e r e d from that of r e l -a t i v e l y nonanxious p i a n i s t s when performing before an audience. In order to discover i f t h i s e f f e c t was observed on d i s c r e t e behavioural measures, two-way between-within univariate analyses of variance were performed on each dependent v a r i a b l e . A summary of the between-within univariate analyses i s presented i n Appendix 22. The group x condition F value for performance q u a l i t y (PQ) was s i g n i f i c a n t ; F(l,38) = 4.64, £ < .04. Simple e f f e c t s analyses were performed to further examine the nature of t h i s i n t e r a c t i o n . The groups' PQ scores did not s i g n i f i c a n t l y d i f f e r when playing alone, while the F value was almost s i g n i f i c a n t when playing before an audience; F(l,38) = 3.89, £ < .056. Although dependent samples t - t e s t s indicated that neither groups' q u a l i t y of performing s i g n i f i c a n t l y changed from playing alone to before an audience, an analysis of diffe r e n c e scores reached s i g n i f i c a n c e ; F(l,38) = 4.64, £ < .04. Difference score means indicated that the pres-ence of an audience influenced the performances of r e l a t i v e l y anxious p i a n i s t s to a s i g n i f i c a n t l y greater extent than the performances of Table VII Means and Standard Deviations for each Dependent Measure Measures R e l a t i v e l y Anxious R e l a t i v e l y Nonanxious (n = 20) M SD (n = 20) M SD PQ Alone Audience TC Alone Audience State Alone Audience Level Mean Alone Audience Strength Mean Alone Audience SUDS Alone Audience RSUDS Alone Audience SUDS Tot a l Alone Audience SS Alone Audience RSCL1 Alone Audience RSCL3 Alone Audience RSCLE Alone Audience Amplitudel (yUmhos) Alone Audience Latencyl (sees) Alone Audience 35.32 33.00 3.00 3.70 41.25 48.00 4.22 4.18 15.27 13.48 40.50 55.50 135.00 182.70 175.50 238.20 144.90 155.30 0.30 0.33 0.58 0.58 0.54 0.53 1.77 1.39 3.08 2.82 12.24 12.28 1.89 2.00 11.13 11.58 0.52 0.73 3.75 4.23 22.59 23.50 72.10 85.82 82.30 97.75 63.02 70.55 0.26 0.35 0.33 0.34 0.23 0.27 1.27 1.84 1.32 1.48 39.42 41.62 3.50 3.35 34.85 36.15 4.80 4.87 21.42 22.18 36.50 43.00 133.90 140.40 170.40 183.40 94.75 97.20 0.22 0.30 0.50 0.54 0.38 0.54 2.05 1.38 3.47 2.64 15.58 15.23 1.47 1.93 8.18 9.15 0.37 0.2'3 4.54 3.89 24.34 27.74 89.85 101.30 108.60 118.40 45.85 46.95 0.21 0.34 0.40 0.34 0.33 0.33 2.14 1.01 2.09 0.92 78 Measure Latency3 (sees) Alone Audience Ristetimel (sees) Alone Audience Risetime3 (sees) Alone Audience Resp-pre Alone Audience Resp-during Alone Audience RHR-pre Alone Audience RHR-during Alone Audience Table VII (cont'd) R e l a t i v e l y Anxious (n = 20) M SD 2.31 1.26 1.93 1.09 3.55 1.52 3.11 1.33 2.75 1.85 2.43 1.05 16.40 2.05 17.13 3.08 23.13 2.84 23.14 3.83 0.33 0.22 0.53 0.22 0.45 0.20 0.88 0.13 Re l a t i v e l y Nonanxious (n = 20) M SD 2.44 1.65 1.94 0.97 3.23 1.53 3.00 2.00 2.44 1.65 1.91 0.93 16.76 2.93 17.06 2.05 20.88 4.03 22.69 3.56 0.30 0.15 0.49 0.24 0''.56 0.23 0.85 0.16 Level Mean = average of SLevel, GLevel and Level; Strength Mean = average of SStrength, GStrength and Strength; SUDS Total = SUDS + RSUDS; RSCL1/3/E = range corrected SCL1/3/E; RHR-pre = range corrected HR-pre; RHR-during = range corrected HR-during 79 r e l a t i v e l y nonanxious p i a n i s t s . Thus, while the performance q u a l i t y of r e l a t i v e l y anxious p i a n i s t s tended to decrease before an audience, the perf-ormance q u a l i t y of r e l a t i v e l y nonanxious p i a n i s t s tended to improve under that condition. No s i g n i f i c a n t d ifferences were found on the Timed Check-l i s t of observable signs of anxiety. S e l f Report Measures M u l t i v a r i a t e s t a t i s t i c s were applied to the following s e l f report v a r i a b l e s ; state anxiety, mean l e v e l of s e l f e f f i c a c y , mean strength of s e l f e f f i c a c y , t o t a l SUDS and s e l f statements. Level of s e l f e f f i c a c y r eferred to the number of tasks subjects believed they could perform with some c e r t a i n t y (scores could range from 0 to 5), and strength of s e l f e f f i c a c y referred to the degree of c e r t a i n t y with which subjects believed they could perform the tasks ( scores could range from 0 to 30). Means of the l e v e l and strength scores from the three scales of s e l f e f f i c a c y ( i . e . S p e c i f i c Expectations of Personal E f f i c a c y , General Expectations of Personal E f f i c a c y and Self E f f i c a c y ) were analysed to provide global meas-ures. Separate analyses of each scale are presented l a t e r . Both group and condition main e f f e c t s were s i g n i f i c a n t ; F(5.34) = 7.39, £ < .001, and F(5,34) = 3.56, £ < .02, r e s p e c t i v e l y . Univariate analyses (see Appendix 22) indicated that each s e l f report measure except t o t a l SUDS contributed to the o v e r a l l group e f f e c t . Examination of the means suggested that r e l a t i v e l y anxious p i a n i s t s generally reported higher l e v e l s of state anxiety, lower l e v e l and strength of s e l f e f f i c a c y , and more frequent negative s e l f statements. The condition main e f f e c t can be a t t r i b u t e d to state anxiety and t o t a l SUDS (see Appendix 22). Examination of means suggested that playing before an audience generally e l i c i t e d higher reports of both state anxiety and SUDS. More importantly, the multivariate analysis produced a s i g n i f i c a n t group x condition i n t e r a c t i o n ; F(5,34) = 3.03, £ < .03. Univariate analyses i d e n t i f i e d mean s e l f e f f i c a c y strength and t o t a l SUDS as c o n t r i b -uting to t h i s i n t e r a c t i v e s i g n i f i c a n c e , with state anxiety approaching s i g n i f i c a n c e (see Appendix 22). To examine the nature of these univariate i n t e r a c t i o n s further, simple e f f e c t s analyses were performed. The groups s i g n i f i c a n t l y d i f f e r e d on mean s e l f e f f i c a c y strength both when playing alone and before an audience: F(l,38) = 21.81, £ < .001, and F(l,38) = 45.86, £ < .001, re s p e c t i v e l y . The means indicated that, i n both cond-i t i o n s , r e l a t i v e l y anxious p i a n i s t s reported lower l e v e l s of mean s e l f e f f i c a c y strength than r e l a t i v e l y nonanxious p i a n i s t s . Dependent samples t - t e s t s indicated that the r e l a t i v e l y anxious group reported s i g n i f i c a n t reductions i n mean s e l f e f f i c a c y strength when playing before an audience, t(19) = 2.61, £ < .02, while no changes occurred across conditions for the r e l a t i v e l y nonanxious group. Also, analyses of variance using d i f f -erence scores indicated that the presence of an audience influenced the s e l f e f f i c a c y strength of r e l a t i v e l y anxious p i a n i s t s to a s i g n i f i c a n t l y greater extent than that of r e l a t i v e l y nonanxious p i a n i s t s ; F(l,38) = 9.41, 2 < .01. The groups did not s i g n i f i c a n t l y d i f f e r on t o t a l SUDS either when playing alone or before an audience. However, dependent samples t - t e s t s indicated that the presence of an audience e l i c i t e d s i g n i f i c a n t l y higher l e v e l s of SUDS i n the r e l a t i v e l y anxious group; t(19) = -2.86, £ < .01. No dif f e r e n c e across conditions was observed for r e l a t i v e l y nonanxious p i a n i s t s . Moreover, the extent to which t o t a l SUDS was influenced by audience presence was s i g n i f i c a n t l y greater for r e l a t i v e l y anxious p i a n i s t s than for r e l a t i v e l y nonanxious p i a n i s t s ; F(l,38) = 4.21, £ < .05. R e l a t i v e l y anxious p i a n i s t s reported s i g n i f i c a n t l y higher l e v e l s of state anxiety than r e l a t i v e l y nonanxious p i a n i s t s both when playing alone and before an audience; F(l,38) = 4.30, £ < .05, and F(l,38) = 12.89, 81 £ < .001, re s p e c t i v e l y . Also, dependent samples t - t e s t s indicated the ^•presence of an audience s i g n i f i c a n t l y increased the r e l a t i v e l y anxious group's l e v e l of state anxiety, t(19) = -2.89, £ < .01, while no di f f e r e n c e s were observed for r e l a t i v e l y nonanxious p i a n i s t s . However, state anxiety was not influenced by audience presence to a s i g n i f i c a n t l y greater extent i n either group. Univariate analyses of variance on s e l f report measures further indicated that groups' SS scores d i f f e r e d s i g n i f i c a n t l y i n both performance conditions; F(l,38) = 8.28, £ < .01, and F(l,38) = 9.42, £ < .01, resp-e c i t v e l y . In both conditions, r e l a t i v e l y anxious p i a n i s t s reported a s i g n i f i c a n t l y greater number of negative s e l f statements than r e l a t i v e l y nonanxious p i a n i s t s . No i n t e r a c t i o n e f f e c t was observed on t h i s v a r i a b l e . Skin Conductance Measures These included range corrected skin conductance l e v e l (prior to tone 1, tone 3 and completion of the performance), and skin conductance response, or, amplitude, latency and r i s e time (at tone 1 and tone 3). Neither the main e f f e c t s nor i n t e r a c t i o n e f f e c t were s i g n i f i c a n t i n the multivariate a n a l y s i s . Even the more l i b e r a l univariate approach f a i l e d to reveal s i g n i f i c a n t differences on any v a r i a b l e . Nor were s i g n i f -i c a n t i n t e r a c t i v e e f f e c t s apparent when time of measurement ( i . e . , tone 1, tone 3, or end of performance) was analysed (see Appendix 23). Heart Rate and Respiration Measures These included range corrected heart rate and r e s p i r a t i o n before and during performances. Neither the group main e f f e c t nor i n t e r a c t i o n e f f e c t reached s i g n i f i c a n c e i n the multivariate a n a l y s i s . However, the condition main e f f e c t was s i g n i f i c a n t ; F(4,35) = 15.59, £ < .001. Univariate analyses of variance indicated that both heart rate p r i o r to playing and during playing s i g n i f i c a n t l y d i f f e r e d across conditions; F(l,38) = 17.45, £ < .001, and F(l,38) = 63.19, £ < .001, resp-82 e c t i v e l y . For both measures, the means indicated that the presence of an audience was associated with considerable increases i n heart rate. Simple e f f e c t s analyses of variance indicated that heart rate l e v e l s (prior to and during) did not s i g n i f i c a n t l y d i f f e r across the groups either when playing alone or before an audience. Nor did the groups d i f f e r i n the extent to which heart rate changed across conditions. Analyses of the i n t e r a c t i o n between time of measurement ( i . e . , p r i o r to or during) and heart rate and r e s p i r a t i o n l e v e l s did not produce s i g n i f i c a n t r e s u l t s (see Appendix 23). Extreme Group Means Analyses The preceding set of analyses suggested that while experimental man-ip u l a t i o n s were generally e f f e c t i v e , the groups were not as d i s t i n c t as intended. To t e s t t h i s hypothesis, post hoc analyses were performed on the 10 highest and 10 lowest scorers on the RCP. Repeated measures multivariate analyses followed by univariate analyses were performed as before. Range corrected heart rate and skin conductance l e v e l were also c a l c u l a t e d for these analyses (absolute values are presented i n Appendix 21). Maximum values for skin conductance l e v e l were; M = 8.95, SD = 3.36 for the most anxious group and M = 12.49, SD = 13.50 for the l e a s t anxious group. Skin conductance minimum l e v e l s were; M = 5.39, SD = 2.48 for the most anxious group, and M = 3.53, SD = 1.31 for the l e a s t anxious group. In heart rate data, maximum values were; M = 126.47, SD = 15.81 for the most anxious group and M = 125.52 , SD = 23.68 for the 'least anxious group. Heart rate minimum values were; M = 77.98, SD = 10.10 for the most anxious group, and M = 75.16, SD = 11.95 for the l e a s t anxious group. Means and standard deviations for for each dependent measure are presented in Table VIII. Behavioural Measures These included performance q u a l i t y and observable signs of anxiety. M u l t i v a r i a t e analyses of these measures produced a s i g n i f i c a n t Table VIII Means and Standard Deviations for each Dependent Measure for the Most Anxious and Least Anxious Groups Most Anxious Least Anxious (n = 10) (n = 10) Measures M SD M SD PQ Alone 30.05 10.95 37.10 13.82 Audience 25.80 2.44 41.40 11.21 TC Alone 3.00 1.70 4.10 1.73 Audience 4.20 2.40 3.40 2.55 State Alone 42.10 7.89 30.80 4.56 Audience 51.60 8.72 34.40 10.17 Level Alone 3.80 1.03 4.70 0.68 Audience 3.60 1.51 4.80 0.42 Strength Alone 15.50 6.90 20.70 5.03 Audience 11.40 5.34 21.70 4.24 Level Mean Alone 3.87 0.45 4.88 0.32 Audience 3.73 0.77 4.90 0.23 Strength Mean Alone 13.77 3.41 22.97 4.68 Audience 11.23 3.68 23.27 4.04 SUDS Alone . 39.90 16.40 31.00 29.23 Audience 62.00 19.89 36.00 31.69 RSUDS Alone 155.10 86.20 104.50 96.74 Audience 213.50 22.29 127.00 116.00 SUDS Tot a l Alone 197.10 91.64 135.50 121.80 Audience 275.50 70.49 163.00 139.10 SS Alone 161.00 67.35 79.10 10.50 Audience 178.20 68.12 92.00 47.94 RSCL1 Alone 0.39 0.30 0.18 0.22 Audience 0.47 0.42 0.18 0.32 RSCL3 Alone 0.43 0.31 0.63 0.37 Audience 0.59 0.35 0.45 0.37 RSCLE Alone 0.46 0.19 0.44 0.37 Audience 0.49 0.30 0.45 0.33 Table VIII (cont'd) Measures Most Anxious (n = 10) M SD Least Anxious (n = 10) M SD Amplitudel (//mhos) Alone Audience Amplitude3 (/tmhos) Alone Audience Latencyl (sees) Alone Audience Latency3 (sees) Alone Audience Risetimel (sees) Alone Audience Risetime3 (sees) Alone Audience Resp-pre Alone Audience Resp-during Alone Audience RHR-pre Alone Audience RHR-during Alone Audience 1.60 0.69 0.71 0.36 2.87 2.79 2.35 2.21 3.47 2.73 2.69 2.25 16.15 16.69 23.16 22.14 0.33 0.54 0.42 0.87 0.82 0.45 0.43 0.28 1.30 1.19 1.24 1.25 1.15 1.28 2.38 1.01 1.22 3.54 2.61 3.09 0.24 0.26 0.23 0.14 1.32 1.17 0.70 0.74 4.09 2.74 1.62 1.66 2.99 3.30 2.31 1.79 16.52 17.43 20.60 23.36 0.28 0.44 0.56 0.85 1.26 0.85 0.90 0.97 2.18 0.84 1.14 1.09 1.56 2.45 1.07 0.94 3.26 2.12 5.21 3.91 0.17 0.25 0.24 0.16 85 group e f f e c t ; F(2,17) = 6.75, £ < .02. Subsequent univariate analyses a t t r i b u t e d t h i s significance to the PQ va r i a b l e ; F(l,18) = 5.24, £ < .04. A summary of the s i g n i f i c a n t univariate analyses i s presented i n Appendix 24. Examination of the means indicated that the most anxious p i a n i s t s generally performed l e s s well than the l e a s t anxious p i a n i s t s . The multivariate analysis also resulted i n a s i g n i f i c a n t group x condition i n t e r a c t i o n ; F(2,17) = 7.59, £ < .01. Univariate analyses again only reached s i g n i f i c a n c e on the PQ va r i a b l e ; F(l,18) = 11.02, £ < .01. Simple e f f e c t s analyses of variance were performed to further analyse t h i s f i n d i n g . While the groups' performance q u a l i t y did not s i g n i f i c a n t l y d i f f e r in the alone condition, l e a s t anxious p i a n i s t s were rated as playing s i g n i f -i c a n t l y better before an audience than most anxious p i a n i s t s ; F(l,18) = 11,48, £ < .01. Dependent samples t - t e s t s indicated that the q u a l i t y of most anxious p i a n i s t s ' playing significantly diminished before an audience (t(9) = 2.39), £ < .025), while l e a s t anxious p i a n i s t s ' playing s i g n i f i c a n t l y improved (t(9) = -2.30, £ < .025). Therefore, t h i s post hoc analysis enhanced the PQ group e f f e c t . A comparison of the experimental group PQ means and the post hoc group PQ means i s depicted i n Figure 1. Despite trends suggestive of group differences on the TC v a r i a b l e , no s i g n i f i c a n t e f f e c t s were obtained. Experimental and post hoc group means are presented i n Figure 2. Se l f Report Measures These included state anxiety, mean s e l f e f f i c a c y l e v e l , mean s e l f e f f i c a c y strength, t o t a l SUDS and s e l f statements. The multivar-i a t e analysis produced s i g n i f i c a n t group and condition main e f f e c t s ; F(5,14) = 10.00, £ < .001, and F(5.14) = 4.93, £ < .01, re s p e c t i v e l y . Univariate analyses of variance indicated that the most anxious group generally reported higher l e v e l s of state anxiety, lower l e v e l and strength of s e l f e f f i c a c y and more negative s e l f statements than the l e a s t anxious group. 86 A l o n e A u d i e n c e F i g u r e 1 : M e a n P e r f o r m a n c e Q u a l i t y f o r R e l a t i v e l y A n x i o u s ( n = 2 0 ) , R e l a t i v e l y N o n a n x i o u s ( n = 2 0 ) , M o s t A n x i o u s ( n = 1 0 ) a n d L e a s t A n x i o u s ( n = 1 0 ) P i a n i s t s . 87 Most Anxious 0 -O R e l a t i v e l y Anxious Q O R e l a t i v e l y Nonanxious 1 1 Least Anxious Alone Audience Figure 2: Mean Timed Che c k l i s t Scores for R e l a t i v e l y Anxious (n = 20), R e l a t i v e l y Nonanxious (n = 20), Most Anxious (n = 10) and Least Anxious (n = 10) P i a n i s t s . Also, the presence of an audience generally e l i c i t e d higher l e v e l s of both state anxiety and t o t a l SUDS. However, the group x condition i n t e r a c t i o n was not s i g n i f i c a n t . Between-within univariate analyses i d e n t i f i e d one s i g n i f i c a n t i n t e r a c t i o n e f f e c t (mean s e l f e f f i c a c y strength). Simple e f f e c t s analyses showed that most anxious subjects reported lower l e v e l s of s e l f e f f i c a c y strength both when playing alone (F(l,18) = 25.27, p_ < .001) and before an audience (F(l,18) = 48.55, £ < .001). Dependent samples t - t e s t s indicated that while the most anxious group's s e l f e f f i c a c y strength almost s i g n i f i c a n t l y decreased across conditions (t(9) = 2.18, £ < .057), no e f f e c t was observed for the l e a s t anxious group. Moreover, univariate analyses using diffe r e n c e scores showed that the presence of an audience influenced s e l f e f f i c a c y strength to a s i g n i f i c a n t l y greater extent i n the most anxious group than the l e a s t anxious group; F{1,18) = 5.15, £ < .04. Se l f e f f i c a c y strength means for experimental and post hoc groups are presented i n Figure 3. Simple e f f e c t s analyses of variance further showed that while the groups' SUDS scores did not s i g n i f i c a n t l y d i f f e r i n the alone condition, the most anxious p i a n i s t s reported s i g n i f i c a n t l y higher SUDS than the le a s t anxious p i a n i s t s before an audience; F(1,18) = 5.21, £ < .025. Moreover, dependent samples t - t e s t s showed that both groups' SUDS l e v e l s s i g n i f i c a n t l y increased when playing before an audience; t(9) = -2.75, £ < .025, and t(9) = -2.70, £ < .025, res p e c t i v e l y (see Figure 4). Also, state anxiety d i f f e r e n t i a t e d the groups i n both performance conditions; F(l,18) = 15.35, £ < .001, and F(l,18) = 16.49, £ < .001, re s p e c t i v e l y . Dependent samples t - t e s t s indicated that the most anxious p i a n i s t s ' l e v e l of state anxiety s i g n i f i c a n t l y increased before an audience, while no changes were observed for l e a s t anxious p i a n i s t s (see Figure 5). 89 sz c QJ 4J co > i u (0 o •H M-1 4-1 w Qi W C (0 s 24 22 20 18 16 14 12 10 Q £ Most Anxious O - O R e l a t i v e l y Anxious 0 D R e l a t i v e l y Nonanxious 1 • Least Anxious D -• — o Alone Audience Figure 3: Mean Self E f f i c a c y Strength for R e l a t i v e l y Anxious (n = 20), R e l a t i v e l y Nonanxious (n = 20), Most Anxious (n = 10) and Least Anxious (n = 10) P i a n i s t s . 90 Q D (0 -U O E-t C (0 (1) S Most Anxious Q— —-Q R e l a t i v e l y Anxious (__) —D R e l a t i v e l y Nonanxious fl fl Least Anxious 280 260 240 220 200 180 c r " u — 160 — 140 — 120 < £ L Alone Audience Figure 4: Mean Total SUDS for R e l a t i v e l y Anxious (n = 20), R e l a t i v e l y Nonanxious (n = 20), Most Anxious (n = 10) and Least Anxious (n = 10) P i a n i s t s . 91 % £ Most Anxious 0 ~ ~ — ~ " 0 R e l a t i v e l y Anxious D D R e l a t i v e l y Nonanxious Alone Audience Figure 5 : Mean State Anxiety for R e l a t i v e l y Anxious (n = 20), R e l a t i v e l y Nonanxious (n = 20), Most Anxious (n = 10) and Least Anxious (n = 10) P i a n i s t s . The groups' SS scores also d i f f e r e d i n both performance conditions; F(l,18) = 11.90, £ < .01, and F(l,18) = 10.71, £ < .01, r e s p e c t i v e l y . However, dependent samples t - t e s t s f a i l e d to reveal s i g n i f i c a n t d ifferences across conditions for either group. Experimental and post hoc group means are presented i n Figure 6. Therefore, the post hoc analyses of s e l f report measures produced very s i m i l a r r e s u l t s to the analysis of the experimental groups, although the former enhanced the group e f f e c t of the SUDS measure i n the audience cond-i t i o n , and generally increased the differences between group means. Skin Conductance Measures These included range corrected skin conductance l e v e l (prior to tone 1, tone 3 and completion of the performance), and skin conductance response, or, amplitude, latency and risetime (at tone 1 and tone 3). Neither main e f f e c t s nor the i n t e r a c t i o n e f f e c t were s i g n i f -icant in the multivariate a n a l y s i s . Univariate analyses, however, ident-i f i e d two e f f e c t s . F i r s t , RSCL at tone 1 was generally s i g n i f i c a n t l y higher for most anxious p i a n i s t s . Second, amplitude at tone 1 was generally s i g n i f i c a n t l y larger when playing alone (see Appendix 24). Heart Rate and Respiration Measures These included range corrected heart rate and r e s p i r a t i o n p r i o r to and during performances. Neither the group main e f f e c t nor the i n t e r a c t i o n e f f e c t reached s i g n i f i c a n c e in the multivar-i a t e a n a l y s i s . However, the condition main e f f e c t was s i g n i f i c a n t ; F(4,15) = 8.61, £ < .001. Univariate analyses i d e n t i f i e d heart rate p r i o r to and during playing as the variables contributing to t h i s s i g n i f i c a n e . Examina-t i o n of the means indicated that the presence of an audience e l i c i t e d considerable increases in both heart rate measures for both groups of sub-j e c t s . Heart rate-pre and heart rate-during means are presented i n Figure 7 and Figure 8 r e s p e c t i v e l y . 93 c QJ e QJ 4-1 03 4J-co QJ co QJ > •H -U <B CT> QJ O QJ X! e 3 190 170 150 130 110 90 70 0 % Most Anxious O O R e l a t i v e l y Anxious • —Q R e l a t i v e l y Nonanxious H — — B Least Anxious a — — — — — ° Alone Audience Figure 6: Mean Number of Negative Self Statements for R e l a t i v e l y Anxious (n = 20), R e l a t i v e l y Nonanxious (n = 20), Most Anxious (n = 10) and Least Anxious (n = 10) P i a n i s t s . 0 0 Most Anxious O - " — ~"~O R e l a t i v e l y Anxious — —£_] R e l a t i v e l y Nonanxious | H Least Anxious a i « 0 . 6 0 . 5 0 . 4 0 . 3 0 . 2 Alone Audience Figure 7 : Mean Heart Rate (range corrected) p r i o r to playing for R e l a t i v e l y Anxious (n = 2 0 ), R e l a t i v e l y Nonanxious (n = 2 0 ), Most Anxious (n = 10) and Least Anxious (n = 10) P i a n i s t s . 95 Figure 8: Mean Heart Rate (range corrected) during performances for R e l a t i v e l y Anxious (n = 20), R e l a t i v e l y Nonanxious (n = 20), Most Anxious (n = 10) and Least Anxious (n = 10) P i a n i s t s . 96 In conclusion, the group means analysis of r e l a t i v e l y anxious and r e l a t i v e l y nonanxious p i a n i s t s indicated that the groups d i f f e r e d on performance q u a l i t y , state anxiety, mean s e l f e f f i c a c y strength, t o t a l SUDS and s e l f statements. No group differences were apparent on the autonomic measures. The e f f e c t of the audience condition was to enhance or create group differences on performance q u a l i t y , state anxiety and mean s e l f e f f i c a c y strength. In addition, the audience condition elevated heart rate measures for both groups of p i a n i s t s . When extreme groups were selected, the group e f f e c t was enhanced on performance q u a l i t y , although the Timed Ch e c k l i s t data f a i l e d to produce s i g n i f i c a n t e f f e c t s . Furthermore, the group e f f e c t s were generally enhanced within s e l f report data. However, t h i s analysis had l i t t l e e f f e c t on the autonomic data. Correlations Among Dependent Measures In order to tes t the hypotheses regarding the degree of concordance between the three response systems, separate Pearson product moment c o r r -e l a t i o n s were generated between each dependent variable (except those measuring skin conductance response) recorded during performances. These included performance q u a l i t y , observable signs of anxiety, retrospective SUDS, mean skin conductance l e v e l (averaged at tone 3 and at completion of the performance), heart-rate during and resp i r a t i o n - d u r i n g . The sign of the c o e f f i c i e n t indicated whether the r e l a t i o n s h i p was concordant or discordant. The size of the c o e f f i c i e n t indicated the strength of the concordance or discordance. The s i g n i f i c a n c e l e v e l of each c o r r e l a t i o n c o e f f i c i e n t was assessed using Student's t tests with N-2 degrees of freedom (°c= .05). Fisher's Z-transformation tests of s i g n i f -icant d i f f e r e n c e s between c o r r e l a t i o n s were used to compare the degree of concordance across conditions and groups. An error rate of .10 was chosen as a Bonferonni significance l e v e l c o r r e c t i o n for testing differences between the c o r r e l a t i o n s ; .10 was deemed appropriate for reducing the p r o b a b i l i t y of Type 1 e r r o r . Correlations were generated for r e l a t i v e l y anxious and r e l a t i v e l y nonanxious p i a n i s t s . I t had been hypothesised that concordance amongst the response systems would occur more often, or with greater strength, for r e l a t i v e l y anxious p i a n i s t s when playing before an audience. Behavioural Measures A concordant r e l a t i o n s h i p between PQ and TC was indicated by a negative c o e f f i c i e n t . For both groups, these variables were discordant when playing alone and concordant when playing before an audience (see Table IX). However, the Fisher's Z-transformation test (OC = .05) showed that the audience condition did not e l i c i t s i g n i f i c a n t l y greater concordance for r e l a t i v e l y anxious p i a n i s t s than for r e l a t i v e l y nonanxious p i a n i s t s . Moreover, none of the c o e f f i c i e n t s reached s i g n i f -icance. However, the c o r r e l a t i o n for r e l a t i v e l y anxious p i a n i s t s i n the audience condition approached s i g n i f i c a n c e . Self Report Measures A concordant r e l a t i o n s h i p between SS and RSUDS was indicated by a p o s i t i v e c o e f f i c i e n t (see Table X). For both groups, these va r i a b l e s were concordant both when playing alone and before an audience. The only c o r r e l a t i o n c o e f f i c i e n t that did not reach s i g n i f i c a n c e occurred for the r e l a t i v e l y nonanxious group i n the audience condition. However, the presence of an audience did not e l i c i t s i g n i f i c a n t l y stronger concordance for r e l a t i v e l y anxious pi a n i s t s . Autonomic Measures As three s i g n i f i c a n c e t e s t s were performed among these measures, the Bonferonni c o r r e c t i o n procedure produced anoC= .033. Concord ance among the three autonomic variables were a l l indicated by p o s i t i v e Table IX Pearson Correlations Between Behavioural Measures TC Alone Audience R e l a t i v e l y Anxious ° ' 3 4 - ° ' 3 6 R e l a t i v e l y Nonanxious 0 ' 0 1 " ° - 2 6 99 Table X Pearson Correlations Between S e l f Report Measures RSUDS  Alone Audience SS R e l a t i v e l y Anxious R e l a t i v e l y Nonanxious 0.70* 0.47* 0.50* 0.16 * p < 0.05 100 c o e f f i c i e n t s (see Table XI). None of the c o r r e l a t i o n s between heart rate and skin conductance l e v e l reached significance. This r e l a t i o n s h i p was discordant for r e l a t i v e l y anxious p i a n i s t s i n both performance conditions. No s i g n i f i c a n t differences emerged between the c o e f f i c i e n t s . Respiration and skin conductance l e v e l were s i g n f i c a n t l y concordant i n the audience condition for r e l a t i v e l y anxious p i a n i s t s , while they were almost s i g n i f i c a n t l y discordant for r e l a t i v e l y nonanxious p i a n i s t s i n the same condition. No s i g n i f i c a n t differences emerged between the coe f f -i c i e n t s . Respiration and heart rate were concordant for both goups i n both performance conditions. However, none of the c o r r e l a t i o n s reached s i g -n i f i c a n c e , and no s i g n i f i c a n t differences emerged between the c o e f f i c i e n t s . Behavioural and Se l f Report Measures As four s i g n i f i c a n c e tests were performed among behavioural and s e l f report measures, the Bonferonni c o r r e c t i o n procedure produced an ot= .025. Concordance between PQ and RSUDS was indicated by a negative c o e f f i c i e n t (see Table XII). The r e l a t i o n -ship was concordant for both groups i n both performance conditions. However, none of the c o r r e l a t i o n s reached s i g n i f i c a n c e , and Fisher's Z-transformation tests f a i l e d to reveal any s i g n i f i c a n t differences between the c o e f f i c i e n t s . Concordance between PQ and SS was also indicated by a negative c o e f f i c i e n t . The r e l a t i o n s h i p was concordant for both groups i n both performance conditions. However, none of the c o r r e l a t i o n s reached s i g -n i f i c a n c e , although the r e l a t i o n s h i p for r e l a t i v e l y nonanxious p i a n i s t s in the alone condition was almost s i g n i f i c a n t l y concordant. No s i g n i f -icant differences emerged between the c o e f f i c i e n t s . Concordance between TC and RSUDS was indicated by a p o s i t i v e c o e f f -101 RHR-during Resp-during Table XI Pearson Correlations Between Autonomic Measures Mean SCL RHR-during Alone Audience Alone Audience R e l a t i v e l y Anxious R e l a t i v e l y Nonanxious R e l a t i v e l y Anxious R e l a t i v e l y Nonanxious -0.24 -0.26 -0.07 0.19 0.07 0.40* 0.21 0.07 0.32 -0.33 0.29 0.11 * p < 0.05 102 Table XII Pearson Correlations Between Behavioural and Self Report Measures RSUDS Alone Audience SS Alone Audience PQ R e l a t i v e l y Anxious R e l a t i v e l y Nonanxious -0.21 -0.11 -0.25 -0.18 -0.18 -0.18 -0.37 -0.27 TC R e l a t i v e l y Anxious R e l a t i v e l y Nonanxious -0.22 0.41* -0.56* 0.25 0.04 -0.07 -0.29 0.16 * p < 0.05 i c i e n t . The r e l a t i o n s h i p was discordant for both groups when playing alone and concordant when playing before an audience. These c o r r e l a t i o n s were only s i g n i f i c a n t for r e l a t i v e l y anxious p i a n i s t s i n the audience condition, and for r e l a t i v e l y nonanxious p i a n i s t s i n the alone c o n d i t i o n . No s i g n i f i c a n t differences emerged between the c o e f f i c i e n t s . Concordance between TC and SS was also indicated by a p o s i t i v e c o e f f -i c i e n t . However, the c o r r e l a t i o n s were very weak, and no trends emerged. Behavioural and Autonomic Measures As s i x s i g n i f i c a n c e tests were performed among behavioural and autonomic measures, the Bonferonni c o r r e c t i o n proced-ure produced an oc = .017. Concordance between skin conductance l e v e l and PQ was indicated by a negative c o e f f i c i e n t (see Table XIII). No trends were apparent, and^no s i g n i f i c a n t d i f f e r e n c e s emerged from Fisher's Z-transformation t e s t s . The c o r r e l a t i o n s were generally very weak. Concordance between heart rate and PQ was also indicated by a negative c o e f f i c i e n t . The r e l a t i o n s h i p was concordant for both groups i n the alone condition, and discordant before an audience. However, the c o r r e l a t i o n s were generally very weak and no s i g n i f i c a n t d i f f e r e n c e s emerged between the c o e f f i c i e n t s . Concordance between r e s p i r a t i o n and PQ was indicated by a negative c o e f f i c i e n t . No trends were apparent, and no s i g n i f i c a n t d i f f e r e n c e s emerged between the c o e f f i c i e n t s . The c o r r e l a t i o n s were generally very weak. Concordance between skin conductance l e v e l and TC was indicated by a p o s i t i v e c o e f f i c i e n t . No trends were apparent, and no s i g n i f i c a n t d i f f e r e n c e s were found between the c o e f f i c i e n t s . The c o r r e l a t i o n s were generally very weak. Concordance between heart rate and TC was indicated by a p o s i t i v e c o e f f i c i e n t . The r e l a t i o n s h i p was discordant for r e l a t i v e l y anxious 104 Table XIII Pearson Correlations Between Behavioural and Autonomic Measures Mean SCL RHR-during Resp-during Alone Audience Alone Audience Alone Audience R e l a t i v e l y Anxious PQ Re l a t i v e l y Nonanxious R e l a t i v e l y Anxious TC R e l a t i v e l y Nonanxious -0.09 0.00 -0.01 0.05 -0.06 0.25 -0.05 -0.25 -0.31 0.13 -0.17 0.07 -0.06 0.01 -0.35 -0.18 -0.22 0.05 0.07 -0.03 -0.11 0.17 0.06 0.52* * p < 0.05 105 p i a n i s t s both when playing alone and before an audience. However, none of the c o r r e l a t i o n s reached s i g n i f i c a n c e , and no s i g n i f i c a n t differences were found between the c o e f f i c i e n t s . Concordance between r e s p i r a t i o n and TC was also indicated by a p o s i t i v e c o e f f i c i e n t . The only c o r r e l a t i o n to reach s i g n i f i c a n c e occurred for r e l a t i v e l y nonanxious p i a n i s t s i n the audience condition. No s i g n i f -icant differences were found using Fisher's Z-transformation t e s t s . Self Report and Autonomic Measures As six si g n i f i c a n c e tests were performed among s e l f report and autonomic measures, the Bonferonni c o r r e c t i o n proced-ure produced an oc = .017. Concordance between RSUDS and each autonomic measure was indicated by p o s i t i v e c o e f f i c i e n t s (see Table XIV). No trends were apparent i n the r e l a t i o n s h i p between RSUDS and skin conductance l e v e l . The c o e f f i c i e n t s were very small, and no s i g n i f i c a n t differences were found between them. RSUDS and heart rate were s i g n i f i c a n t l y concordant i n the alone condition, and almost s i g n i f i c a n t l y concordant i n the audience condit-ion for r e l a t i v e l y anxious p i a n i s t s . The c o r r e l a t i o n s for r e l a t i v e l y nonanxious p i a n i s t s were generally weaker. No s i g n i f i c a n t d i f f e r e n c e s were found between the c o e f f i c i e n t s . The r e l a t i o n s h i p between RSUDS and r e s p i r a t i o n was concordant for both groups i n both performance conditions. However, none of the c o r r e l -ations reached s i g n i f i c a n c e , and no s i g n i f i c a n t differences were found between the c o e f f i c i e n t s . SS and skin conductance l e v e l were discordant i n the alone condition and concordant i n the audience condition for r e l a t i v e l y anxious p i a n i s t s . However, none of the c o r r e l a t i o n s reached s i g n i f i c a n c e , and no s i g n i f -icant differences were found between the c o e f f i c i e n t s . 106 Table XIV Pearson Correlations Between S e l f Report and Autonomic Measures RSUDS Alone Audience SS Alone Audience Mean SCL R e l a t i v e l y Anxious R e l a t i v e l y Nonanxious -0.18 0.06 -0.00 0.06 -0.14 0.19 0.22 0.12 RHR-during R e l a t i v e l y Anxious R e l a t i v e l y Nonanxious 0.41* 0.36* -0.08 0.10 0.08 0.38* 0.04 -0.40* Resp-during R e l a t i v e l y Anxious R e l a t i v e l y Nonanxious 0.30 0.09 0.06 0.12 -0.16 -0.28 0.07 0.23 * p < 0.05 SS and heart rate were s i g n i f i c a n t l y concordant for r e l a t i v e l y anxious p i a n i s t s i n the audience condition. However, the presence of an audience did not e l i c i t s i g n i f i c a n t l y greater concordance than the alone condition. The r e l a t i o n s h i p was s i g n i f i c a n t l y discordant for r e l a t i v e l y nonanxious p i a n i s t s when playing before an audience. No trends emerged from the c o r r e l a t i o n s between SS and r e s p i r a t i o n . None of the c o r r e l a t i o n s reached s i g n i f i c a n c e , and no s i g n i f i c a n t differences were found using Fisher's Z-transformation t e s t s . In conclusion, few s i g n i f i c a n t r e l a t i o n s h i p s emerged from the c o r r e l a t i o n a l analyses, and no s i g n i f i c a n t differences were found between the c o r r e l a t i o n s . However, a l l but one of the s i g n i f i c a n t concordant r e l -ationships observed i n the audience condition occurred for the r e l a t i v e l y anxious group. This was found i n the following r e l a t i o n s h i p s ; PQ and TC (approached s i g n i f i c a n c e ) , SS and RSUDS, r e s p i r a t i o n and skin conduct-ance l e v e l , TC and RSUDS, heart rate and RSUDS (approached s i g n i f i c a n c e ) , and heart rate and SS. The other concordant r e l a t i o n s h i p observed i n the audience condition occurred between TC and r e s p i r a t i o n i n the r e l a t i v e -l y nonanxious group. Three a d d i t i o n a l s i g n i f i c a n t concordant r e l a t i o n s h i p s were obtained i n t h i s a n a l y s i s ; SS and RSUDS for both groups of p i a n i s t s in the alone condition, and heart rate and RSUDS for r e l a t i v e l y anxious p i a n i s t s i n the alone condition. In contrast, the only s i g n i f i c a n t discordance that was observed occurred for the r e l a t i v e l y nonanxious group i n either the alone or performance conditions. This was found i n the following r e l a t i o n s h i p s ; r e s p i r a t i o n and skin conductance l e v e l (approached s i g n i f i c a n c e ) , TC and RSUDS, and heart rate and SS. Sel f E f f i c a c y Between-within univariate analyses of variance were performed on the s i x d i f f e r e n t indices of s e l f e f f i c a c y ; SLevel, SStrength, GLevel, GStrength, Level and Strength. Means and standard deviations are presented i n Table XV. S i g n i f i c a n t group x condition in t e r a c t i o n s were only obtained on GStrength (F(l,38) = 5.17, £ < .03) and Strength (F(l,38) = 5.40, £ < .03). Simple e f f e c t s analyses of variance were performed to examine the nature of these i n t e r a c t i o n s further. R e l a t i v e l y anxious p i a n i s t s reported a s i g n i f i c a n t l y lower strength of general expectations of personal e f f i c a c y (GStrength) both when playing alone and before an audience; F(l,38) = 14.39, £ < .001, and F(l,38) = 44.73, £ < .001, r e s p e c t i v e l y . Dependent samples t - t e s t s indicated that an audience presence e l i c i t e d s i g n i f i c a n t l y lower l e v e l s of GStrength in the r e l a t i v e l y anxious group; t(19) = 2.24, £ < .04. No s i g n i f i c a n t d i f f e r e n c e occurred across conditions for r e l a t i v e l y nonanxious p i a n i s t s . Also, analyses of variance using d i f f e r e n c e scores indicated that the extent to which GStrength was influenced by the audience condition was s i g n i f i c a n t l y greater for r e l a t i v e l y anxious p i a n i s t s than for r e l a t i v e l y nonanxious p i a n i s t s ; F(l,38) = 5.17, £ < .03. R e l a t i v e l y anxious p i a n i s t s also reported s i g n i f i c a n t l y lower strengths of s e l f e f f i c a c y (Strength) than r e l a t i v e l y nonanxious p i a n i s t s in the audience condition; F(l,38) = 15.32, £ < .001. Neither groups' scores were s i g n i f i c a n t l y influenced by the presence of an audience. However, an analysis using difference scores indicated that the extent to which the audience condition influenced Strength was s i g n i f i c a n t l y greater for r e l a t i v e l y anxious p i a n i s t s than for r e l a t i v e l y nonanxious p i a n i s t s ; F(l,38) = 5.40, £ < .03. Item 2 of the Self E f f i c a c y scale d i r e c t l y represented the exper-imental performing conditions of t h i s study. Therefore group d i f f e r -ences on the strength values (or, conviction ratings) given to that 109 Table XV Means and Standard Deviations of Self E f f i c a c y Measures R e l a t i v e l y Anxious R e l a t i v e l y Nonanxious (n = 20) (n = 20) Measures M SD M SD SLevel Alone 4.20 0.83 4.85 0.49 Audience 4.25 1.02 4.95 0.22 SStrength Alone 15.25 5.12 22.75 4.54 Audience 14.75 5.69 23.25 4.34 GLevel Alone 4.40 0.94 4.95 0.22 Audience 4.20 1.20 5.00 0.00 GStrength Alone 14.30 7.63 22.65 6.22 Audience 11.95 5.89 23.25 4.73 Level Alone 4.05 0.95 4.60 0.68 Audience 4.10 1.25 4.65 0.59 Strength Alone 16.25 6.48 18.85 5.22 Audience 13.75 5.49 20.05 4.66 SLevel = l e v e l of s p e c i f i c expectations of personal e f f i c a c y (Kendrick); SStrength = strength of s p e c i f i c expectations of personal e f f i c a c y (Kendrick); GLevel = l e v e l of general expectations of personal e f f i c a c y (Kendrick); GStrength = strength of general expectations of personal e f f i c a c y (Kendrick); Level = l e v e l of s e l f e f f i c a c y ; Strength = strength of s e l f e f f i c a c y . item were examined. Means and standard deviations are presented in Table XVI. A between-within univariate analysis of variance indicated a s i g n i f i c a n t group main e f f e c t ; F(l,38) = 6.07, £ < .02. Examination of means showed that the r e l a t i v e l y anxious group generally reported lower conviction i n t h e i r a b i l i t y to perform the task at hand with t h e i r anxiety under c o n t r o l than the r e l a t i v e l y nonanxious group. The group x condition i n t e r a c t i o n also reached s i g n i f i c a n c e ; F(l,38) = 7.93,' £ < .01. Simple e f f e c t s analyses of variance were performed to further examine the nature of t h i s i n t e r a c t i o n . While the groups' l e v e l of conviction did not d i f f e r i n the alone condition, r e l a t i v e l y anxious p i a n i s t s reported s i g n i f i c a n t l y lower convictions in their a b i l i t y than r e l a t i v e l y non-anxious p i a n i s t s when playing before an audience; F(l,38) = 10.28, p_ < .01. Dependent samples t - t e s t s indicated that while the audience condition did not s i g n i f i c a n t l y influence the conviction ratings of r e l a t i v e l y anxious p i a n i s t s , the conviction ratings of r e l a t i v e l y non-anxious p i a n i s t s s i g n i f i c a n t l y increased; t(19) = -2.60, £ < .02. An analysis using d i f f e r e n c e scores, however, f a i l e d to reach s i g n i f i c a n c e . The analysis of conviction ratings was.repeated for the l e a s t anxious and most anxious groups. A between-within univariate analysis of variance produced a s i g n i f i c a n t group e f f e c t ; F(l,18) = 41.64, £ < .001 Least anxious p i a n i s t s generally reported higher conviction ratings than most anxious p i a n i s t s . Simple e f f e c t s analyses of variance further showed that t h i s was the case i n both performance conditions; F(l,18) = 39.88, £ < .001, and F(l,18) = 37.03, £ < .001, r e s p e c t i v e l y . A s i g n i f i c a n t condition e f f e c t was a l s o obtained; F(l,18) = 7.71, £ < .02. The presence of an audience was generally associated with higher convict-ion ratings than the alone condition. However, neither groups' l e v e l of c o n v i c t i o n s i g n i f i c a n t l y changed over conditions and no i n t e r a c t i o n e f f e c t I l l Table XVI Means and Standard Deviations for Item 2 of the S e l f E f f i c a c y Scale Alone Audience R e l a t i v e l y Anxious (n = 20) M SD 4.55 4.10 1.40 1.29 R e l a t i v e l y Nonanxious (n = 20) M SD 4.85 5.40 1.14 0.75 Alone Audience Most Anxious (n = 10) M SD 2.90 3.20 0.88 1.03 Least Anxious (n = 10) M SD 5.30 5.60 0.82 0.70 112 was observed. To t e s t the hypotheses concerning c o r r e l a t i o n s between s e l f e f f i c a c y and the three response systems, Pearson product moment c o r r e l a t i o n s were generated between conviction ratings on item 2 of the Se l f E f f i c a c y scale and the following dependent v a r i a b l e s ; performance q u a l i t y , scores on the Timed C h e c k l i s t , state anxiety, retrospective SUDS, mean range corrected skin conductance l e v e l (averaged at tone 3 and at the end of performance recording periods), range corrected heart rate-during and r e s p i r a t i o n -during. These variables were recorded during the performances. C o r r e l -ations were generated for r e l a t i v e l y anxious and r e l a t i v e l y nonanxious groups, and are presented i n Table XVII. It had been hypothesised that s e l f e f f i c a c y would c o r r e l a t e with the dependent variables most often for the r e l a t i v e l y anxious group i n the audience condition. No consistent trends emerged from the c o r r e l a t i o n a l analyses. A p r e d i c t i v e c o r r e l a t i o n between s e l f e f f i c a c y and PQ was indicated by a p o s i t i v e c o e f f i c i e n t , and t h i s occurred for r e l a t i v e l y nonanxious p i a n i s t s i n both performance conditions. However, none of the c o r r e l a t i o n s reached s i g n i f i c a n c e . As with the remaining v a r i a b l e s , a p r e d i c t i v e c o r r e l a t i o n between s e l f e f f i c a c y and TC was indicated by a negative c o e f f i c i e n t . However, none of the TC c o r r e l a t i o n s either reached s i g n i f i c a n c e , or occurred i n a p r e d i c t i v e d i r e c t i o n . The c o r r e l a t i o n s with state anxiety a l l occurred i n p r e d i c t i v e d i r e c t -ions. C o e f f i c i e n t s for r e l a t i v e l y anxious p i a n i s t s i n the audience cond-i t i o n and for r e l a t i v e l y nonanxious p i a n i s t s i n the alone condition reached s i g n i f i c a n c e . 113 Table XVII Pearson Correlations Between Self E f f i c a c y and Dependent Variables R e l a t i v e l y Anxious R e l a t i v e l y Nonanxious Alone Audience Alone Audience PQ -0.18 -0.02 0.15 0.26 TC 0.02 0.28 0.11 0.08 State -0.29 -0.56* -0.77* -0.24 RSUDS -0.43* -0.40* -0.18 0.07 SS -0.10 -0.38* -0.60* -0.50* MRSCL 0.15 0.17 0.16 -0.11 RHR-during -0.43* -0.21 -0.15 0.18 Resp-during -0.12 0.18 -0.05 0.26 * p < 0.05 114 S e l f e f f i c a c y c o r r e l a t i o n s with RSUDS were s i g n i f i c a n t and p r e d i c t i v e for both performance conditions i n the r e l a t i v e l y anxious group. The c o r r e l -ations were weaker in the r e l a t i v e l y nonanxious group. Se l f e f f i c a c y c o r r e l a t i o n s with SS were s i g n i f i c a n t and p r e d i c t i v e for both performance conditions i n the r e l a t i v e l y nonanxious group and for the audience condition i n the r e l a t i v e l y anxious group. None of the c o r r e l a t i o n s with skin conductance l e v e l reached s i g n i f -icance and no trends were apparent. The only c o e f f i c i e n t to reach s i g n i f i c a n c e i n the c o r r e l a t i o n s between s e l f e f f i c a c y and heart rate occurred for the r e l a t i v e l y anxious group i n the alone condition. F i n a l l y , none of the c o r r e l a t i o n s with r e s p i r a t i o n reached s i g n i f i c a n c e and no trends were apparent. In conclusion, few s i g n i f i c a n t c o r r e l a t i o n s were obtained, most of those occurring with the s e l f report system. The audience condition e l i c i t e d four s i g n i f i c a n t l y p r e d i c t i v e c o r r e l a t i o n s , three of those occurring i n the r e l a t i v e l y anxious group. The alone condition e l i c i t e d three s i g n i f i c a n t l y p r e d i c t i v e c o r r e l a t i o n s , two of those occurring i n the r e l a t i v e l y nonanxious group. In order to te s t Bandura's s p e c i f i c hypothesis concerning the a b i l i t y of s e l f e f f i c a c y strength to p r e d i c t arousal l e v e l s , Pearson product moment c o r r e l a t i o n s were generated between strength values at l e v e l s 3 , 4 and 5 r e s p e c t i v e l y of the Se l f E f f i c a c y scale and both heart rate p r i o r to and during playing. These c o r r e l a t i o n s are presented i n Table XVIII. None of the c o r r e l a t i o n s reached s i g n i f i c a n c e . At l e v e l s 3 and 4 , the co r r e l a t i o n s occurred i n a p r e d i c t i v e d i r e c t i o n ; as s e l f e f f i c a c y strength 115 Table XVIII Pearson Correlations Between Self E f f i c a c y Strength at Three Levels and Heart Rate HR-pre HR-during Level 3 (n = 12) -0.14 -0.20 Level 4 (n = 20) -0.35 -0.24 Level 5 (n = 45) 0.21 0.04 * p < 0.05 decreased, heart rate tended to increase. However, the c o r r e l a t i o n s were nonpredictive at l e v e l 5. 117 DISCUSSION The r e s u l t s of t h i s study showed that the audience condition e l i c i t e d more intense emotional responses in r e l a t i v e l y anxious p i a n i s t s , and that, in c e r t a i n measures, the groups d i f f e r e d i n the extent to which anxiety was observed. Group dif f e r e n c e s were generally enhanced when extreme scorers on the i n i t i a l screening index were analysed. As hypothesised, the more anxious p i a n i s t s experienced synchronous changes amongst the response systems as the i n t e n s i t y of the emotional response increased, consistent with the three-systems model. Desynchronous changes were observed in the le s s anxious p i a n i s t s . This was also consistent with the three-systems model, as i t em p i r i c a l l y i l l u s t r a t e d that the response systems can vary independently. This desynchrony was a t t r i b u t e d to the high demand nature of the audience condition. Despite the generally weak corr -e l a t i o n s between the measures of each response system, concordance was most evident when intense emotional responses were e l i c i t e d . This fi n d i n g also supported the three-systems model. In t h i s study, data related to the s e l f e f f i c a c y measures did not provide support for s e l f e f f i c a c y theory. Self e f f i c a c y ratings were neither consistent with changes in the l e v e l s of anxiety obtained in each response system, nor correlated with more than one response system. Therefore, the o v e r a l l r e s u l t s were interpreted as being more consistent with the three-systems model than with s e l f e f f i c a c y theory. This chapter describes the e f f e c t s of both the group and condition independent v a r i a b l e s . This i s followed by discussion of synchrony or desynchrony and concordance or discordance amongst the response systems. F i n a l l y , s e l f e f f i c a c y data are commented upon. Also, various related issues 118 are raised throughout the discussion, including the influence of high demand conditions, the l i m i t a t i o n s of various dependent measures including problems i n the measurement of two autonomic channels, and p o t e n t i a l problems with the s e l f e f f i c a c y theory. Experimental Manipulation It had been hypothesised that the presence of an audience would pro-mote the greatest anxiety in the r e l a t i v e l y anxious group. This hypothesis was supported by the behavioural and s e l f report dependent measures. On one of the two behavioural measures (PQ), the r e l a t i v e l y anxious group were influenced to a s i g n i f i c a n t l y greater extent by the audience condition than the r e l a t i v e l y nonanxious group. Of the f i v e s e l f report measures, three demonstrated s i g n i f i c a n t condition e f f e c t s only in the r e l a t i v e l y anxious group. These measures were state anxiety, t o t a l SUDS and mean s e l f e f f i c a c y strength. While skin conductance data did not r e f l e c t a condition e f f e c t , heart rate both p r i o r to playing and during playing considerably increased when an audience was present. However, the heart rate e f f e c t occurred for both groups of p i a n i s t s . Therefore, the chosen method for the experimental manipulation appeared to be mostly e f f e c t i v e . This suggests than an evaluative audience, v i s i b l e videocamera and an i n s t r u c t i o n a l set emphasising performance evaluation, together served as quite powerful sources of evaluative anxiety for p i a n i s t s who s e l f reported performance anxiety. As such, the r e l a t i v e l y anxious p i a n i s t s seemed to experience more intense emotional responses in the audience condition and le s s intense emotional responses in the alone condition, thus allowing the hypotheses regarding synchrony and concordance rates amongst the three response systems to be l e g i t i m a t e l y tested. I t i s possible that the conditions' e f f e c t was moderated by apprehen-sion surrounding the i n i t i a l exposure to experimental procedures and to the large d i s p l a y of sophisticated, and perhaps intimidating equipment. As such, the f i r s t performance condition (alone) may have e l i c i t e d more anxiety than intended. In addition, habituation to nonspecific factors may have occurred by the second performance condition (audience). These considerations would mitigate the impact of the audience condition. Stronger group by condition i n t e r a c t i o n s may have resulted had the order of conditions been counter-balanced. However, i t was deemed inappropriate to expose the more anxious subjects to s t r e s s f u l conditions without an e a r l i e r i n t e r a c t i o n . The r e s u l t s also provide confirmation of the legitimacy of grouping subjects as r e l a t i v e l y anxious or r e l a t i v e l y nonanxious p i a n i s t s . The RCP received empirical concurrent v a l i d a t i o n from the modified version of the General T r a i t Anxiousness scale that was adapted to the solo piano perform-ance s i t u a t i o n . I t had been hypothesised that the groups would d i f f e r most during the audience condition. This hypothesis was supported in the analysis of performance q u a l i t y , where the group e f f e c t was very close to s i g n i f i c a n c e . Group differences were also obtained for three of the f i v e s e l f report measures (state anxiety, mean s e l f e f f i c a c y strength and s e l f statements) i n both performance conditions. However, the group differences were apprec-i a b l y larger i n the audience condition for two of those measures (state anxiety and mean s e l f e f f i c a c y strength). Group e f f e c t s were not observed on any autonomic measure. Of the 13 dependent measures, f i v e demonstrated s i g n i f i c a n t predicted e f f e c t s and two demonstrated trends towards predicted e f f e c t s i n the anxious p i a n i s t s . Six measures did not support the hypothes-es, and these mostly represented the autonomic response system. 120 Examination of group means suggested that the groups were not as d i s t -i n c t as intended. The median-split of the RCP scores may have resulted i n an overlapping of actual anxiety l e v e l s between the groups. To t e s t t h i s p o s s i b i l i t y , post hoc analyses were performed on the 10 highest and 10 lowest RCP scorers. These analyses enhanced group differences i n the audience condition for the PQ behavioural measure. In addition, s e l f report group mean differences were enhanced by the post hoc analyses, and the group e f f e c t was appreciably larger i n the audience condition for three s e l f report measures; state anxiety, mean s e l f e f f i c a c y l e v e l and mean s e l f e f f i c a c y strength. Moreover, the groups also only d i f f e r e d i n the audience condition on a fourth s e l f report measure ( t o t a l SUDS). While group dif f e r e n c e s were obtained on a skin conductance measure (RSCL1) by the post hoc a n a l y s i s , t h i s e f f e c t was present in both conditions. Therefore, the post hoc analyses of extreme groups produced somewhat larger group dif f e r e n c e s i n the audience condition within the behavioural, and s e l f report measures. This finding suggests that while the RCP i s an adequ-ate measure for i d e n t i f y i n g anxious and nonanxious p i a n i s t s , more extreme scores more r e l i a b l y i d e n t i f y groups with substantial differences in perf-ormance anxiety. I t was apparent in the analysis of the experimental groups that some r e l a t i v e l y nonanxious group members reported comparatively high l e v -e l s of subjective d i s t r e s s and/or experienced high l e v e l s of autonomic arous-a l . In addition, some r e l a t i v e l y anxious group members reported comparat-i v e l y low l e v e l s of subjective d i s t r e s s and/or experienced l i t t l e arousal. This pattern has been observed by others, who accordingly express concern for the v a l i d i t y of findings based on subjects selected only on s e l f report c r i t e r i a (Borkovec, 1973a, 1973b, 1976; Hicks and Steinber, 1976; Bernstein and Paul, 1971). Such observations usually r e s u l t from using s e l f report screening measures because they are susceptible to demand influences. This 121 study attempted to overcome or minimise demand influences f i r s t l y by emphas-i s i n g the a c c e p t a b i l i t y of both anxiousness and nonanxiousness for the study in the recruitment i n s t r u c t i o n s . This strategy was suggested by Bernstein and Paul (1973). Secondly, an a d d i t i o n a l subject s e l e c t i o n c r i t -erion was included to exclude those subjects for whom demand influences were most l i k e l y to be operative; high RCP scorers were also required to i d e n t i f y performance anxiety as a current source of concern while low RCP scorers were required to state that i t was not. This strategy was suggested by Bernstein (1973). It i s also possible that those who volunteered to p a r t i c i p a t e i n the study were generally l e s s anxious than those who avoided p a r t i c i p a t i o n due to excessive performance anxiety. (Although, only four of the 10 v o l -unteers who dropped out of the study were r e l a t i v e l y anxious performers). Had t h i s been an important consideration, however, the mean RCP score for the r e l a t i v e l y anxious group would have been lower than that of the popul-ati o n of anxious p i a n i s t s . However, Appel's (1972) group of 30 graduate music students who experienced performance anxiety reported RCP scores ranging from six to 25, averaging at 15.53. R e l a t i v e l y anxious p i a n i s t s i n th i s study reported RCP scores ranging from 11 to 27, averaging at 15.5. R e l a t i v e l y nonanxious p i a n s i s t s reported RCP scores ranging from one to 10, averaging at 6.5. Therefore , the r e l a t i v e l y anxious group in t h i s study appeared to be comparable to the anxious p i a n i s t s in Appel's study, whereas, several subjects i d e n t i f i e d as r e l a t i v e l y nonanxious i n t h i s study reported RCP scores higher than those of a proportion of Appel's anxious subjects. This again suggests that the groups did not r e f l e c t as s u b s t a n t i a l l y d i f f -erent anxiety l e v e l s as intended. While the apparent overlap between the experimental groups reduced the l i k e l i h o o d of obtaining expected trends within the analyses, the r e s u l t s have greater g e n e r a l i z a b i l i t y than would 122 those obtained from extreme scorers. The gener ality of the r e s u l t s can be examined by two means. F i r s t , the extent to which scores obtained by the r e l a t i v e l y anxious and most anxious groups on various dependent measures compared to those obtained by other populations of anxious i n d i v i d u a l s can be examined. Second, the extent to which differences (across conditions and groups) that were s t a t -i s t i c a l l y s i g n i f i c a n t represented differences of r e a l magnitude can be examined. Comparative data are not a v a i l a b l e for the performance q u a l i t y measure. The changes i n performance q u a l i t y across conditions experienced by both groups were r e l a t i v e l y small, which may suggest that t h i s measure was not s e n s i t i v e enough to detect the e f f e c t s of d i f f e r e n t i a l anxiety l e v e l s within groups. However, the differences between groups i n performance q u a l i t y were r e l a t i v e l y large i n the audience condition, e s p e c i a l l y when comparing l e a s t anxious and most anxious subjects; most anxious subjects generally performed well below the l e v e l of average performance q u a l i t y whereas l e a s t anxious subjects generally performed at average l e v e l s . State anxiety scores obtained by r e l a t i v e l y anxious and p a r t i c u l a r l y by most anxious p i a n i s t s i n the audience condition were well above the mean obtained by general medical and s u r g i c a l populations (M = 42.38) (Speilberger et a l . , 1970) and were comparable to those obtained from anxious p i a n i s t s i n Kendrick's (1979) study (M = 51.42, SD = 12.24). In contrast, means obtained by r e l a t i v e l y nonanxious and l e a s t anxious p i a n i s t s were well below those obtained by general medical and s u r g i c a l p a t i e n t s . In addition, the changes experienced across conditions i n state anxiety scores were large for both anxious groups ( a 17% increase for the most anxious group). While the changes experienced by the anxious groups i n strength of s e l f e f f i c a c y were not s u b s t a n t i a l , group differences on t h i s v ariable were of s i g n i f i c a n t magnitude, e s p e c i a l l y i n the post hoc analyses where 123 most anxious p i a n i s t s ' scores were 40% lower than those of l e a s t anxious p i a n i s t s . SUDS l e v e l s were s l i g h t l y above moderate l e v e l s ( i . e . , 50) i n the audience condition for r e l a t i v e l y anxious p i a n i s t s , and well above moderate l e v e l s for most anxious p i a n i s t s . The RSUDS measure r e f l e c t e d further elevation above moderate l e v e l s of anxiety i n those groups. In contrast, r e l a t i v e l y nonanxious and p a r t i c u l a r l y l e a s t anxious p i a n i s t s reported SUDS and RSUDS l e v e l s well below moderate l e v e l s i n the audience condition. In addition, the anxious groups experienced changes of s i g n i f -icant magnitude i n both the SUDS measure (a 23% increase for the most anxious group) and the RSUDS measure (a 19% increase for the most anxious groups) across conditions. Mean negative s e l f statement scores of the anxious groups were comparable to those obtained by Kendrick (1979) (M = 157.15). Group differences on t h i s v a r i a b l e were quite large i n both conditions ( 18% and 27% r e s p e c t i v e l y ) . F i n a l l y , the magnitude of heart rate elevations for the anxious and nonanxious groups were s u b s t a n t i a l . I t therefore appears that psychometric scores obtained under s t r e s s f u l conditions by the anxious groups i n t h i s study were comparable to those obtained by other anxious populations. While changes across conditions i n performance q u a l i t y may not have attained ' c l i n i c a l ' s i g n i f i c a n c e , the increases i n SUDS, RSUDS, and state anxiety i n the anxious groups appeared s u b s t a n t i a l , as did the elevations i n heart rate experienced by the anxious and nonanxious groups. In addition, group differences i n the audience condition on performance q u a l i t y , state anxiety, s e l f e f f i c a c y strength, SUDS, RSUDS, and negative s e l f statements appeared to r e f l e c t s u b s t a n t i a l e f f e c t s . The i n t e r a c t i o n r e s u l t i n g from sex diff e r e n c e s was another factor to consider i n i n t e r p r e t i n g the major analyses. The r e l a t i v e l y anxious group 124 was comprised of s i g n i f i c a n t l y more females than males. Sex interacted with several dependent va r i a b l e s , moreso i n the audience condition. However, sex could not be included as a grouping factor as some c e l l s would have been too small for v a l i d s t a t i s t i c a l analyses. Skin conduct-ance l e v e l at tone one was s i g n i f i c a n t l y lower for females in both perf-ormance conditions. Also, skin conductance l e v e l s at tone 3 and at complet-ion of the performance were s i g n i f i c a n t l y lower for females i n the audience condition. S i m i l a r l y , Buck, M i l l e r and Caul (1974) found that females demonstrated less skin conductance responsiveness than males. Therefore, group differences in the audience condition on skin conductance l e v e l may have been mitigated by the large proportion of females i n the r e l a t i v e l y anxious group. On the other hand, group differences on s e l f report measures may have been enhanced by the sex i n t e r a c t i o n . The mean values obtained by both males and females on two s e l f report measures (strength of s p e c i f i c s e l f e f f i c a c y and s e l f statements) did not appreciably change across the performance conditions. However, the mean value obtained by the r e l a t i v e l y anxious group, on the s e l f e f f i c a c y measure at l e a s t , did s i g n i f i c a n t l y change across the performance conditions. This suggests that the group e f f e c t ( i . e . l e v e l of anxiety) was greater than the sex e f f e c t on that var-i a b l e . As a r e s u l t , the only s e l f report variables in which sex differences i n f l a t e d group dif f e r e n c e s in the audience condition were state anxiety, strength of general expectations of personal e f f i c a c y and strength of s e l f e f f i c a c y . I t seems p l a u s i b l e to postulate that, considering the large s i z e of the group differences on those v a r i a b l e s , the group e f f e c t may have remained had the variance a t t r i b u t a b l e to sex been removed. Three Response Systems I t had been hypothesised that synchronous changes would be observed amongst the behavioural, s e l f report and autonomic response systems in the r e l a t i v e l y anxious group. That i s , each response system should have r e f l e c t e d increases i n anxiety as a r e s u l t of i n t e n s i f i c a t i o n of the emotional response i n the audience condition. In contrast, r e l a t i v e l y nonanxious p i a n i s t s were expected to demonstrate l i t t l e change in resp-onse to the audience condition. In general, synchrony was observed among the response systems i n the r e l a t i v e l y anxious group, while desynchrony was observed i n the r e l a t i v e l y nonanxious group. The following changes occurred across performance conditions for r e l a t i v e l y anxious p i a n i s t s ; performance q u a l i t y tended to decrease (this trend reached s i g n i f i c a n c e i n the most anxious group), state anxiety increased, t o t a l subjective units of d i s t r e s s increased, mean s e l f e f f i c a c y strength decreased, and heart rate increased. Trends were also present i n d i c a t i n g more frequent signs of overt anxiety (e s p e c i a l l y i n the most anxious group) and more frequent negative s e l f statements. A l l of these changes were i n d i c a t i v e of higher l e v e l s of anxiety. S i m i l a r l y , Sartory et a l . (1977) found that heart rate and subjective units of d i s t r e s s synchronously increased i n a l i n e a r fashion as the i n t e n s i t y of the emotional response increased. The following changes occurred across performance conditions for r e l a t i v e l y nonanxious p i a n i s t s ; performance q u a l i t y tended to increase (this trend reached s i g n i f i c a n c e i n the l e a s t anxious group), and heart rate increased. Trends were also present i n d i c a t i n g fewer signs of overt anxiety, more frequent negative s e l f statements and increased t o t a l subjective units of d i s t r e s s (this trend reached s i g n i f i c a n c e i n the l e a s t anxious group). That i s , while behavioural measures r e f l e c t e d reduced anxiety, some s e l f report measures remained stable, others r e f l e c t e d trends towards increased anxiety and heart rate l e v e l s r e f l e c t e d heightened auton-omic arousal. These findings support the three-systems model by showing that the three response systems can vary independently, p a r t i c u l a r l y when only weak emotional responses are assessed. Furthermore, the r e s u l t s pertaining to r e l a t i v e l y anxious p i a n i s t s support Hodgson and Rachmans' (1974) p r e d i c t i o n that synchrony would be more evident .when emotional responses are i n t e n s i f i e d . Most of the v a r i a b l e s r e f l e c t i n g increased anxiety in the r e l a t i v e l y nonanxious and l e a s t anxious groups belonged to the s e l f report system. Accordingly, Lang (1971) has stated that the s e l f report system i s the response system most l i k e l y to r e f l e c t anxiety in mildly f e a r f u l subjects, as the s e l f report system i s believed to be the most s e n s i t i v e system (Agras and Jacobs, 1981), and the most vulnerable to demand influences (Borkovec and O'Brien, 1976). However, heart rate l e v e l s also r e f l e c t e d increased anxiety i n the r e l a t i v e l y anxious and l e a s t anxious groups. This was somewhat unexpected as the autonomic system i s believed to be the l e a s t s e n s i t i v e response system (Agras and Jacobs, 1981). Heart rate i s generally believed to be a d i r e c t and r e l i a b l e measure of anxiety (Borkovec et a l . , 1977). However, group differences were not obtained on t h i s measure. Neither r e s p i r a t i o n nor vigour correlated s i g n i f -i c a n c t l y with heart rate, and therefore, probably did not confound heart rate data. The a n t i c i p a t o r y , and e s p e c i a l l y the performance, heart rate l e v e l s were generally very high. However, Borkovec et a l . (1974) recorded heart rates peaking at 115 beats per minute i n a group of s o c i a l l y anxious i n d i v i d u a l s , and Kendrick's (1979) anxious p i a n i s t s attained a mean heart rate of 116 beats per minute p r i o r to treatment. Also, the p h y s i c a l e f f o r t involved i n playing the piano may have contributed to the elevation observed during performances i n t h i s study. In addition, the method of analysing averages of tachometer output (as used i n both t h i s and Kendrick's studies) may have i n f l a t e d mean heart rate l e v e l s as a r e s u l t of the i n f l -uence of extreme scores. Nevertheless, analyses of d i f f e r e n c e scores 127 should have c o n t r o l l e d for the cognitive and motor demands of the task, thereby revealing the influence of d i f f e r e n t i a l l e v e l s of anxiety. This influence was not observed. The unpredicted elevation of heart rate i n the audience condition for the r e l a t i v e l y nonanxious group p a r a l l e l s the findings of Borkovec et a l . (1974) . They found that high demand conditions resulted i n s i g n i f i c a n t l y increased heart rate over low demand conditions i n a group of 'low s o c i a l l y anxious' i n d i v i d u a l s . The high demand condition of the i r study e n t a i l e d an expectation that subjects were to i n t e r a c t i n a s o c i a l s i t u a t i o n while demonstrating l i t t l e or no anxiety. Performing a musical piece before an audience may reasonably be considered to be a more demanding condition than performing alone. Hodgson and Rachman (1974) hypothesised that high demand conditions would r e s u l t i n greater discordance among the response systems than low demand conditions. That i s , demanding s i t u a t i o n s are believed to f a c i l i t a t e behavioural approach to a feared stimulus, or performance accomp-lishment, while maintaining or enhancing arousal l e v e l s (subjective and autonomic).' This hypothesis has received empirical support (Grey et a l . , 1979) . S i m i l a r l y , r e l a t i v e l y nonanxious and l e a s t anxious subjects i n t h i s study demonstrated increased autonomic arousal and subjective d i s t r e s s (though the l a t t e r only reached s i g n i f i c a n c e i n the l e a s t anxious group) and greater performance accomplishment when playing before an audience. These r e s u l t s can therefore be parsimoniously explained by the influence of increased demand. An a l t e r n a t i v e explanation of the performance improvement evidenced, p a r t i c u l a r l y i n the l e a s t anxious group, would be a c u r v i l i n e a r r e l a t i o n s h i p between autonomic arousal and performance. Lomas (1937) found that public speakers who reported severe performance anxiety gave speeches judged l e a s t s a t i s f a c t o r y , those who reported moderate performance anxiety gave 128 speeches judges superior, and those who reported l i t t l e or no performance anxiety gave speeches judged mediocre (in Kendrick, 1979). In following t h i s argument, i t would be hypothesised that r e l a t i v e l y nonanxious and l e a s t anxious p i a n i s t s experienced increased arousal i n the audience condition s u f f i c i e n t to motivate superior performance q u a l i t y , In contrast, r e l a t -i v e l y anxious and most anxious p i a n i s t s would have experienced severely high l e v e l s of arousal i n the audience condition s u f f i c i e n t to di s r u p t performance. Borkovec (1976) has also postulated that very high l e v e l s of arousal have d i s r u p t i v e e f f e c t s upon the remaining response systems. How-ever, as previously mentioned, that notion assumes that arousal mediates behaviour, and thus i s based upon a more i n t e r a c t i v e model of the response systems than that o r i g i n a l l y formulated by Lang (1968). Also, that notion cannot explain why the same l e v e l of autonomic arousal was accompanied by improved performance in one group of subjects, and diminished performance i n the other group of subjects. I t i s also important to consider that the actual improvement demonstrated by the l e a s t anxious group was not very s u b s t a n t i a l , i n which case any attempt to explain the s t a t i s t i c a l s i g n i f -icance may be u n j u s t i f i e d without r e p l i c a t i o n of the f i n d i n g . Skin conductance l e v e l and response measures did not r e f l e c t any condition or group e f f e c t s i n t h i s study. Skin conductance l e v e l data appeared to be r e l i a b l y recorded. However, more error was apparent i n the recording of skin conductance response, which may have been due to the use of telemetry which tended to reduce the r e s o l u t i o n of polygraph output. In addition, measurement of skin conductance response appeared to be con-founded by at l e a s t two f a c t o r s . F i r s t , i t i s possible that spontaneous (or, nonspecific) f l u c t u a t i o n s were recorded when a tone-reaction did not occur. This may account for the extreme latency times recorded for some subjects, as response latency time r a r e l y exceeds one to two seconds (Hass-e t t , 1978) . A serious design l i m i t a t i o n was the f a i l u r e to spec i f y the 129 time i n which responses should occur before being considered nonspecific. Second, responses to tone 3 (which was ^broadcast one minute following commencement of playing) may have interacted with the loudness with which the piece was played. Such an i n t e r a c t i o n could have been systematically investigated had the loudness been rated as a dependent measure. Skin conductance l e v e l also f a i l e d to produce systematic e f f e c t s . Skin conductance may be le s s s e n s i t i v e than other measures of anxiety, requiring more intense s t i m u l i to e l i c i t responsiveness. In general, the autonomic system i s believed to be the l e a s t s e n s i t i v e response system (Agras and Jacobs, 1981). Also, the large amount of i n d i v i d u a l v a r i a t i o n i n the skin conductance data may have contributed to the unsystematic nature of the f i n d i n g s . The f a c t that skin conductance i s influenced by many factors a d d i t i o n a l to anxiety l e v e l s may have p a r t l y explained such v a r i a b i l i t y . I t i s for t h i s reason that Hodgson and Rachman (1974) hypothesised that skin conductance would show greater desynchrony with behavioural and s e l f report measures than heart r a t e . Craig and Wood (1971) suggested that skin conductance i s p a r t i c u l a r l y responsive to a t t e n t i o n a l processes. S i m i l a r l y , Paul (1969) noted: Higher conductance l e v e l s have recently been found under conditions of attention, novelty or u n f a m i l i a r i t y than under actual stress (Flanagan, 1964; Roessler et a l . , 1966), i n some changing i n the opposite d i r e c t i o n to other p h y s i o l o g i c a l measures (MacNeilage, 1966) thus suggesting greater independence from sympathetic c o n t r o l , (in Hodgson and Rachman, 1974) <j Therefore, skin conductance may not have been an appropriate measure for t h i s study, as piano performance ( p a r t i c u l a r l y when playing from memory), requires intense concentration on the task at hand (Kendrick, 1979). A t t e n t i o n a l factors may have masked or confounded the e f f e c t s of d i f f e r e n t -i a l l e v e l s of anxiety upon skin conductance l e v e l . Accordingly, examination of i n d i v i d u a l data indicated that approximately h a l f of the r e l a t i v e l y anxious group experienced increased skin conductance l e v e l s i n the audience 130 condition, while the rest experienced either decreased l e v e l s or no change. A s i m i l a r pattern emerged from the r e l a t i v e l y nonanxious group. As such, group dif f e r e n c e s would have been more l i k e l y p r i o r to playing at the tone 1 period of measurement. Although skin conductance l e v e l of most anxious p i a n i s t s ' was s i g n i f i c a n t l y higher than that of l e a s t anxious p i a n i s t s ' at t h i s point, the d i f f e r e n c e was apparent i n both performance conditions. I t was apparent from v i s u a l inspection of the recordings and from the r e l a t i v e l y low r e l i a b i l i t i e s for r e s p i r a t i o n recordings whilst playing, that respiration-during measures were confounded by movement a r t i f a c t . This may have obscured the often observed r e l a t i o n s h i p between r e s p i r a t i o n rate and heart rate that was not found to be s i g n i f i c a n t in t h i s study. Though the observed trends i n the Timed Che c k l i s t of Observable Signs of Anxiety data occurred in predicted d i r e c t i o n s , they did not reach s i g n i f -icance. Most of the c h e c k l i s t items were observed very r a r e l y , accounting for the somewhat low w i t h i n - i n t e r v a l rater agreements on several items. Also, the l o c a t i o n of the videocamera meant that several of the f i n e motor behav-iours were obscured (e.g. hands tremble, knees tremble). If the camera had been located closer to the subjects, t h i s behavioural measure may have yielded more information. However, Borkovec et a l . (1974) also f a i l e d to f i n d group dif f e r e n c e s using a more complete version of a Timed C h e c k l i s t . They a t t r i b u t e d t h e i r unsystematic findings to low i n t e r r a t e r r e l i a b i l i t y (r = .71). Also, Kendrick (1979) suggested that some of the more subtle behaviours observed during actual performances do not appear on the video-recording. She questionned the usefulness of the Timed Che c k l i s t for video r a t i n g purposes. In t h i s respect, i t i s notable that mean scores on t h i s v a r i a b l e were extremely low, which further suggests that t h i s measure i s not s u i t a b l e for video ratings of signs of anxiety in p i a n i s t s . Although the r e l a t i v e l y anxious group reported s i g n i f i c a n t l y higher l e v e l s of subjective units of d i s t r e s s when playing before an audience than when playing alone, only the extreme groups d i f f e r e d s i g n i f i c a n t l y on t h i s v a r i a b l e . The large amount of i n t e r - i n d i v i d u a l v a r i a b i l i t y may have con-tributed to the f a i l u r e to obtain a differ e n c e between r e l a t i v e l y anxious and r e l a t i v e l y nonanxious groups. More subject t r a i n i n g i n the standard-ised use of the SUD scales may have yielded more systematic fi n d i n g s . I t was also hypothesised, i n accordance with Hodgson and Rachman(1974), that concordance would be most evident when intense emotional responses were e l i c i t e d ; that i s , i n the r e l a t i v e l y anxious group when performing before an audience. I n t e r c o r r e l a t i o n s were examined among the dependent -a measures recorded during performances. In general, the c o r r e l a t i o n s were weak and few attained s t a t i s t i c a l s i g n i f i c a n c e . Past research has found that the i n t e r c o r r e l a t i o n s among measures within a response system are generally stronger than i n t e r c o r r e l a t i o n s between response systems (Lang, 1971). Within behavioural measures, performance q u a l i t y and scores on the Timed C h e c k l i s t were discordant when playing alone and concordant when playing before an audience. That i s , as performance q u a l i t y decreased, the number of overt signs of anxiety tended to increase. The only c o r r e l a t i o n , however, that approached s i g n i f i c a n c e occurred for r e l a t i v e l y anxious p i a n i s t s in the audience condition. The r e l a t i o n s h i p s between s e l f statements and retrospective SUDS were s i g n i f i c a n t l y concordant for r e l a t i v e l y anxious performers i n both conditions. Despite s i g n i f i c a n t concordance i n the alone condition, these measures cor r e l a t e d very poorly for r e l a t i v e l y nonanxious p i a n i s t s i n the audience condition. Only one r e l a t i o n s h i p reached s i g n i f -icance within the autonomic measures. This occurred between skin conduct-ance and r e s p i r a t i o n for r e l a t i v e l y anxious p i a n i s t s when playing before an audience; as r e s p i r a t i o n quickened, the l e v e l of skin conductance tended to 132 increase. In contrast, an almost s i g n i f i c a n t uncoupling occurred between those measures for r e l a t i v e l y nonanxious p i a n i s t s i n the same condition. Low i n t e r c o r r e l a t i o n s within the autonomic system have been previously rep-orted (Lang, 1971; Hassett, 1978), and i s one reason for recording more than one autonomic measure. As previously mentioned, skin conductance data were probably confounded by a t t e n t i o n a l f a c t o r s , thereby disturbing the i n t e r c o r r e l a t i o n s . In summary, the within-system i n t e r c o r r e l a t i o n s were generally strongest in the s e l f report system, followed by the behavioural system, and f i n a l l y the autonomic system. Also, i n accordance with p r e d i c t i o n s , a l l but one of the s i g n i f i c a n t concordances obtained occurred for r e l a t i v e l y anxious p i a n i s t s when playing before an audience. In contrast, discordant trends were most apparent in the r e l a t i v e l y nonanxious group. This pattern of r e s u l t s lends support to the hypothesis that concordance would be more evid-ent when more intense emotional responses are evoked. However, these e f f e c t s for within systems c o r r e l a t i o n s were not s u f f i c i e n t l y large to produce d i f f -erences between the c o r r e l a t i o n s . That i s , the audience condition did not e l i c i t s i g n i f i c a n t l y greater concordances i n the r e l a t i v e l y anxious group. Between-systems c o r r e l a t i o n s were then analysed. None of the c o r r e l -ations between performance q u a l i t y and the s e l f report measures reached l e v e l s beyond that a t t r i b u t a b l e to chance. A general tendency was observed, however, for performance q u a l i t y to decrease as more SUDS and more frequent negative s e l f statements were reported. While assessing anxious p i a n i s t s i n condit-ions s i m i l a r to the alone condition i n t h i s study, Kendrick (1979) obtained a s i g n i f i c a n t c o r r e l a t i o n (r_ = .57) between performance and s e l f statements. The only s e l f report measure i n t h i s study with which Timed Ch e c k l i s t data were s i g n i f i c a n t l y concordant was retrospecitve SUDS. This occurred for the r e l a t i v e l y anxious group when playing before an audience; more signs of 133 overt anxiety were recorded as subjects reported more subjective d i s t r e s s . This r e l a t i o n s h i p was s i g n i f i c a n t l y discordant for r e l a t i v e l y nonanxious p i a n i s t s i n the alone condition. This pattern of r e s u l t s also lent support to the hypothesis. However, s i g n i f i c a n t d ifferences were not obtained between the c o r r e l a t i o n s , and the r e l a t i o n s h i p s between the Timed C h e c k l i s t data and s e l f statements were generally weak. Therefore, only one of the four subsets of c o r r e l a t i o n a l analyses between the behavioural and s e l f report systems provided some support for the hypotheses. Yet, stronger r e l a t i o n s h i p s were expected, as the i n t e r c o r r e l a t i o n s between the behavioural and s e l f report systems are reportedly the strongest (Rach-man, 1978). This may only be the case for behavioural avoidance. Unlike behavioural approach t e s t s , where subjects have the choice to avoid the feared stimulus, f e a r f u l subjects i n t h i s study were required to confront the feared s i t u a t i o n . As a r e s u l t , concordance rates may have been weakened by enforcing behavioural accomplishment in the anxious subjects i n spite of subjective and autonomic arousal. That i s , the high demand nature of the audience condition may have counteracted the e f f e c t s on concordance rates of intense emotional responses. Correlations between behavioural and autonomic measures were generally weak. Performance q u a l i t y did not s i g n i f i c a n t l y c o r r e l a t e with any auton-omic measure; nor did any trends emerge i n the data. Kendrick (1979) also f a i l e d to f i n d s i g n f i c a n t c o r r e l a t i o n s between performance q u a l i t y and heart rate. S i m i l a r l y , the Timed C h e c k l i s t data c o r r e l a t e d only weakly with the autonomic measures. The only r e l a t i o n s h i p that reached s i g n i f i c a n c e occurred i n the r e l a t i v e l y nonanxious group when playing before an audience; subjects tended to appear more anxious as r e s p i r a t i o n quickened. This r e s u l t did not support the hypothesis. Rachman (1978) has previously commented upon the generally low c o r r e l a t i o n s found between behavioural and autonomic 134 measures. This has been atributed to the unequal s e n s i t i v i t y of the sys-tems (Lang, 1971). Error v a r i a t i o n i n measures also reduces i n t e r c o r r e l -ations (Hodgson and Rachman, 1974) which may p a r t l y explain the lack of r e l a t i o n s h i p with skin conductance l e v e l s . Two of the s i x subsets of i n t e r c o r r e l a t i o n s between the s e l f report and autonomic responses lent support to the hypothesis. Heart rate s i g n i f -i c a n t l y correlated with subjective units of d i s t r e s s i n a concordant d i r e c t i o n . This occurred for r e l a t i v e l y anxious p i a n i s t s i n both perform-ance conditions. However, greater concordance was not e l i c i t e d by the audience condition. Those variables were very poorly r e l a t e d i n the r e l a t -i v e l y nonanxious qroup. More support for the hypothesis was evident in the c o r r e l a t i o n s between heart rate and s e l f statements; the audience condition e l i c i t e d s i g n i f i c a n t concordance i n the r e l a t i v e l y anxious group which was not e l i c i t e d by the alone condition. In contrast, r e l a t i v e l y nonanxious p i a n i s t s experienced a s i g n i f i c a n t uncoupling of heart rate and s e l f statements. The only autonomic measure that related at a l l systematically to s e l f r e D o r t measures was heart rate. Skin conductance did not s i g n i f i c a n t l y c o r r e l a t e with either behavioural or s e l f report measures. These findings support Hodgson and Rachmans' (1974) f i f t h hypothesis that heart rate would be more concordant or synchronous with the remaining response systems than skin conductance. In general, the i n t e r c o r r e l a t i o n s amongst the response systems were weak. In ad d i t i o n , s i g n i f i c a n t differences were not obtained between the c o r r e l a t i o n s when using a conservative Bonferonni c o r r e c t i o n procedure. That i s . the audience condition did not e l i c i t s i g n i f i c a n t l y stronger concordances for the r e l a t i v e l y anxious group than the alone condition, 135 or the r e l a t i v e l y nonanxious group. The case for concordance among the d i f f e r e n t measures of the hypothetical construct of i n t e r e s t , namely anxiety, may be complicated by other factors as yet unspecified. Again, one i s struck by the complexity of the phenomena. Problems measuring the r e l a t i o n s h i p s may have derived from the low s t a t i s t i c a l power associated with c o r r e l a t i o n a l analyses. R e l a t i v e l y small numbers of subjects would also appear to have minimised the number of s i g n i f i c a n t c o r r e l a t i o n s found. Also, Hodgson and Rachman (1974) noted that the range of scores a f f e c t s the c o r r e l a t i o n c o e f f i c i e n t s ; lower c o r r e l a t i o n s are to be expected when the range of scores i s small. However, a l l but one of the s i g n i f i c a n t c o r r e l a t i o n s obtained l e n t support to the hypothesis that concordance would be more evident amongst the response systems of r e l a t i v e l y anxious p i a n i s t s when playing before an audience. In contast, r e l a t i v e l y nonanxious p i a n i s t s demonstrated either nondescript or discordant r e l a t i o n s h i p s . The c o r r e l a t i o n a l r e s u l t s also support the notion of the three response systems, by i l l u s t r a t i n g that they are not n e c e s s a r i l y well c o r r e l a t e d at any given point. This set of data, therefore, does not support the unitary model of fear. This study also revealed some q u a l i t a t i v e differences i n the response patterns of more f e a r f u l and l e s s f e a r f u l subjects. Therefore, unless precautions are taken to ensure that mildly f e a r f u l subjects are excluded, the v a l i d i t y of generalizing from analogue to c l i n i c a l populations i s questionnable. As t h i s study indicated, and as previously suggested (Bernstein and Paul, 1971; S a l l i s e t a l . , 1980), severely f e a r f u l i n d i v -iduals are more l i k l e y to respond to the feared object in each response system whereas les s f e a r f u l i n d i v i d u a l s seem to respond to varying degrees i n each system. 136 This study also demonstrated the importance of using multiple mode measurement i n fear and anxiety related research. Under c e r t a i n conditions, l e v e l s of subjective anxiety are u n l i k e l y to p a r a l l e l l e v e l s of either behavioural or p h y s i o l o g i c a l arousal. Several measures were obtained for each response system i n t h i s study. Despite p o t e n t i a l biases operating within s e l f report measures (Borkovec et a l . , 1977; Lick, 1977; Lick and Katkin, 1976), these measures were generally able to discriminate the two groups and were responsive to the experimental manipulation. Within the autonomic system, skin conductance and r e s p i r a t i o n (being confounded by measurement problems) did not d i f f e r e n t i a t e the groups or conditions, while heart rate d i f f e r e n t i a t e d the conditions but not the groups. Of the two behavioural measures, the performance q u a l i t y variable provided the most information. Self E f f i c a c y Group by condition i n t e r a c t i o n s only occurred on two of the six ind-ices of s e l f e f f i c a c y ; strength of general expectations of personal e f f i c a c y and strength of s e l f e f f i c a c y . As predicted, the audience condition e l i c i t e d greater group dif f e r e n c e s on those variables than the alone condition. However, i t should be noted that sex in t e r a c t i o n s were also present on those v a r i a b l e s . The extent to which group e f f e c t s would have remained had the variance a t t r i b u t a b l e to sex been removed i s uncertain. Sex differences may therefore be an important consideration i n any assessment of s e l f e f f i c a c y . It should also be noted that predicted i n t e r a c t i o n e f f e c t s were apparent i n the analysis of mean s e l f e f f i c a c y strength ( i . e . the average strength scores for the three measures). However, in accordance with Bandura's recommendation for microanalytic analyses, that variable i s not appropriate for determining the r e l a t i o n s h i p s between s e l f e f f i c a c y judgements and the response systems. 137 C e i l i n g e f f e c t s were apparent on the l e v e l of s e l f e f f i c a c y i n d i c e s , which reduce the power to detect s i g n i f i c a n t d ifferences on those v a r i a b l e s . The r e s u l t s suggest that those indices were inappropriate for t h i s sample of subjects to the extent that they were i n s e n s i t i v e to the d i f f e r e n t l e v e l s of the independent variables because the range of scores was so r e s t r i c t e d . I t i s not possible to compare t h i s study's r e s u l t s with those obtained by Kendrick (1979) from the Expectations of Personal E f f i c a c y scale where only composite scores (combined l e v e l and strength) were presented. I t would appear that the l e v e l of s e l f e f f i c a c y correlated very l i t t l e with the strength of s e l f e f f i c a c y i n t h i s study, and the usefulness of the l e v e l index was equivocal. Self e f f i c a c y ratings not only distinguished the groups i n the audience condition, but also i n the alone condition. S e l f e f f i c a c y theory would p r e d i c t , that, as s e l f e f f i c a c y mediates a l l of the response systems, group differences should also have been observed i n each response system when playing alone. However, only two measures, both within the s e l f report system and both involving sex in t e r a c t i o n s (state anxiety and s e l f s t a t e -ments) distinguished the groups i n the alone condition. The behavioural and autonomic response systems did not d i f f e r e n t i a t e the groups when play-ing alone, and therefore, the r e s u l t s from the alone condition do not support s e l f e f f i c a c y theory p r e d i c t i o n s . The only index of s e l f e f f i c a c y that r e f l e c t e d reduced confidence i n the r e l a t i v e l y anxious p i a n i s t s ' a b i l i t y to perform when playing before an audience was the strength of general expectations of personal e f f i c a c y . Yet, i t was expected that the Se l f E f f i c a c y scale that was s p e c i f i c a l l y designed to be commensurate with the performance conditions i n t h i s study, would be most l i k e l y to r e f l e c t that reduction i n confidence. In contrast, the General Expectations of Personal E f f i c a c y scale was comprised of items 138 l e a s t representative of the performance conditions. However, the fi n d i n g that a s i g n i f i c a n t decrease i n at l e a s t one measure of s e l f e f f i c a c y was accompanied by increased l e v e l s of anxiety i n each response system i s consistent with s e l f e f f i c a c y theory. That i s , a reduced confidence i n one's a b i l i t y to perform c e r t a i n tasks i s believed to produce more s e l f defeating thoughts and d i s t r e s s , diminished behavioural mastery and heightened arousal (Bandura, 1982). Therefore, such findings could be interpreted to suggest that s e l f e f f i c a c y mediated a l l of those increases i n anxiety experienced by the r e l a t i v e l y anxious group. However, the pattern of r e s u l t s exhibited by the r e l a t i v e l y nonanxious group was not consistent with s e l f e f f i c a c y theory. S e l f e f f i c a c y theory p r e d i c t s that s t a b i l i t y of s e l f e f f i c a c y judgements across conditions should be accompanied by s t a b i l i t y of anxiety l e v e l s i n each response system. However, while s e l f e f f i c a c y indices did not a l t e r across conditions for the r e l a t i v e l y nonanxious group, the i r heart rates increased. Therefore, i t would be more parsimonious to explain these two sets of findings for each group i n terms of a single model (such as the three-systems model) than to postulate the mediating role of s e l f e f f i c a c y for only one set of r e s u l t s . Predictions from s e l f e f f i c a c y theory were more reasonably tested when conviction ratings of a b i l i t y to perform the s p e c i f i c experimental task i n each performance condition (as represented by item 2 of the Self E f f i c a c y scale) were examined. In accordance with the experimental hypotheses, the groups' l e v e l s of conviction did not s i g n i f i c a n t l y d i f f e r i n the alone condition. Whereas, r e l a t i v e l y anxious p i a n i s t s reported s i g n i f i c a n t l y lower conviction than r e l a t i v e l y nonanxious p i a n i s t s i n the audience condition. However, t h i s d i f f e r e n c e was not due to an expected decrease i n conviction for r e l a t i v e l y anxious p i a n i s t s , but to a s i g n i f i c a n t increase i n conviction 139 for r e l a t i v e l y nonanxious p i a n i s t s . The r e s u l t s pertaining to the f i r s t mentioned group were u n l i k e l y to be due to inadequate group d i s t i n c t i v e n e s s , as s i m i l i a r r e s u l t s were obtained when examining most anxious and l e a s t anxious p i a n i s t s . However, habituation of the anxious subjects to the experimental nature of the s e t t i n g following the i n i t i a l exposure in the alone condition may have mitigated the e f f e c t of the audience condition on t h e i r conviction ratings. Yet, the three response systems did show increased l e v e l s of anxiety in response to the audience condition. Therefore, the lack of consistency between the conviction ratings for anxious p i a n i s t s and the i r response system measures does not lend support to s e l f e f f i c a c y theory. Moreover, while conviction ratings increased for r e l a t i v e l y anxious p i a n i s t s , they did not experience s i g n i f i c a n t anxiety reductions i n any response system, and even experienced heightened autonomic arousal. As such, t h i s more "reasonable" t e s t of Bandura's predictions s t i l l f a i l e d to lend support to his theory. Bandura's research has mostly focused upon fears of spiders and snakes in subjects who probably had few enduring encounters with th e i r feared objects due to avoidance behaviour. Therefore, those subjects were u n l i k e l y to have developed s k i l l s related to feared objects. In contrast, p i a n i s t s are required to play before audiences, and are more l i k e l y to have developed s k i l l r elated to th e i r feared s i t u a t i o n s . Rachman (1978) claimed that s e l f e f f i c a c y (or, c o n t r o l l a b i l i t y ) i s very dependent on competence or s k i l l s . P i a n i s t s i n t h i s study had a l l attained quite developed s k i l l bases, as t h e i r grades ranged from nine to A.R.C.T. l e v e l s . As a r e s u l t , judgements of competence may have remained stable across the conditions, thereby accounting for the s t a b i l i t y of the s e l f e f f i c a c y ratings in the anxious groups. Indeed, t h i s observation i s consistent with the commonplace rep-140 o r t from expert concert p i a n i s t s that they are able to play, and do so most e f f e c t i v e l y , i n front of very demanding audiences, despite consid-erable apprehension and d i s t r e s s (Kendrick, 1979). I t had been hypothesised that s e l f e f f i c a c y would be most cor r e l a t e d with the response systems i n the r e l a t i v e l y anxious group when playing before an audience. This hypothesis was based on the three-systems model p r e d i c t i o n that the response systems would also be most concordant under those conditions. However, s e l f e f f i c a c y ratings i n t h i s study s i g n i f -i c a n t l y correlated with s e l f report measures only. This p a r a l l e l s Kendrick's (1979) pretreatment c o r r e l a t i o n a l f i n d i n g s . Also, the c o r r -e l a t i o n s were neither more evident nor stronger for the audience condition or for the r e l a t i v e l y anxious group. A second analysis was performed on the r e l a t i o n s h i p between s e l f e f f -icacy and autonomic arousal on the basis of Bandura's (1980) p r e d i c t i o n that s e l f e f f i c a c y strength i s most p r e d i c t i v e of a n t i c i p a t o r y and perf-ormance arousal when the l e v e l of s e l f e f f i c a c y i s held constant. However, neither a n t i c i p a t o r y nor performance heart rate l e v e l s s i g n i f i c a n t l y c orrelated with s e l f e f f i c a c y strength at either l e v e l 3, 4 or 5. Also, i n contrast to predicted trends, heart rate tended to increase as s e l f e f f i c a c y strength increased at l e v e l 5. Therefore, the c o r r e l a t i o n a l analyses also f a i l e d to lend support to s e l f e f f i c a c y theory p r e d i c t i o n s . Conclusion The r e s u l t s were more consistent with the three-systems model than with s e l f e f f i c a c y theory. Consistency with the three-systems model was based on two major fin d i n g s . F i r s t , the response systems demonstrated independent change over time ( i . e . desynchrony) and independent i n t e n s i t i e s at any given time ( i . e . discordance) under c e r t a i n conditions. Second, 141 synchrony and concordance were more evident when intense emotional responses were e l i c i t e d , although the concordance rates were not as frequent or as strong as expected. Inconsistency with the s e l f e f f i c a c y theory was based on three major f i n d i n g s . F i r s t , increases or decreases in s e l f e f f i c a c y were not accompanied by decreases or increases, r e s p e c t i v e l y , i n anxiety l e v e l s of each response system. Second, s t a b i l i t y of s e l f e f f i c a c y was not accompanied by s t a b i l i t y of anxiety l e v e l s . Third, s e l f e f f i c a c y did not c o r r e l a t e with more than one response system. Together, these findings suggest that s e l f e f f i c a c y did not mediate changes in each response system. I t would seem from comments following t h e i r performances that some p i a n i s t s were surprised by t h e i r reactions to the experimental conditions, or by the q u a l i t y of th e i r performances. That i s , th e i r pre-performance predictions were not substantiated by th e i r actual performance. Bandura's theory i s based on the assumption that people are good predictors of their own reactions, but t h i s may not necess a r i l y be the case. The counter arg-ument offered by s e l f e f f i c a c y theory i s that, unless m i c r o a n a l y t i c a l analyses are performed, r e l a t i o n s h i p s between s e l f e f f i c a c y judgements and actions w i l l become incongruent through the influence of such intervening variables as reappraisals and f a u l t y s e l f knowledge (Bandura, 1982). However, the s e l f e f f i c a c y scale designed for t h i s study was very s p e c i f i c to the experimental tasks, and was completed very soon before the subjects performed. I f s e l f e f f i c a c y analyses must be more s p e c i f i c than the proc-edure used i n t h i s study in order to obtain predicted r e s u l t s , the u s e f u l -ness and a p p l i c a b i l i t y of s e l f e f f i c a c y theory are severely l i m i t e d . In conclusion, s e l f e f f i c a c y theory does not seem able to accommodate the predictions from the three-systems model, and thus would not appear to be s u i t a b l e hypothetical construct for that model as was suggested by Wilson (1978). On the other hand, the three-systems model also requires further empirical i n v e s t i g a t i o n i n order to c l a r i f y the nature of those conditions that r e l i a b l y p r e d i c t more or l e s s concordance and/or synchrony. 143 REFERENCES Agras, W., & Jacob, R. Phobia: Nature and Measurements. In M. Maviss-a k a l i n & D. Barlow (Eds.), Phobia: Psychological and Pharmacological Treatment. New York: G u i l f o r d Press, 1981 Anderson, M., & Borkovec, T. Imagery Processing and Fear Reduction during Repeated Exposures to Two Types of Phobic Imagery. Behaviour Research  and Therapy, 1980, 18, 537-540. Andrasik, F., Turner, S., & Ollendick, T. S e l f Report and Physiologic Responding During In Vivo Flooding. Behaviour Research and Therapy, 1980, 18, 593-594. Appel, S. Modifying Solo Performance'Anxiety i n Adult P i a n i s t s (Doctoral D i s s e r t a t i o n , 1974). D i s s e r t a t i o n Abstracts International, 1974, 35, 3503A. Appel, S. Modifying Solo Performance Anxiety i n Adult P i a n i s t s . Journal  of Music Therapy, 1976, 13_, 2-16. Bandura, A. S e l f E f f i c a c y : Towards a Unifying Theory of Behavior Therapy. Psychological Review, 1977, 84, 191-215. Bandura, A. The Se l f System i n Reciprocal Determinism. American Psycholog- i s t , 1978, 33(1), 344-58. Bandura, A. Reflections on S e l f E f f i c a c y . Advances i n Behaviour Research  and Therapy, 1978, 1, 237-269. Bandura, A. In Search of Pure U n i d i r e c t i o n a l Determinants. Behavior Ther- apy, 1981, 12, 30-40. Bandura, A. S e l f - E f f i c a c y Mechanism i n Human Agency. American Psychologist, 1982, 37, 122-147. Bandura, A., & Adams, N. Analysis of Self E f f i c a c y Theory of Behavioural Change. Cognitive Therapy and Research, 1977, _1J4) , 287-310. Bandura, A., Adams, N., & Beyer, J . Cognitive Processes Mediating Behavioral Change. Journal of Personality and S o c i a l Psychology, 1977, 35_(3), 125-139. Bandura, A., Adams, N., Hardy, A., & Howells, A. Tests of the Generality of Se l f E f f i c a c y Theory. Cognitive Therapy and Research, 1980, 4_(1) , 39-66. Bandura, A., Reese, L., & Adams, N. Microanalysis of Action and Fear Arousal as a Function of D i f f e r e n t i a l Levels of Perceived S e l f E f f i c a c y . Journal of Personality and S o c i a l Psychology, 1982, _4_3 (1) , 5-21. Barlow, D., & Mavissakalin, M. Directions i n the Assessment and Treatment of Phobia: The Next Decade. In M. Mavissakalin & D. Barlow (Eds.) Phobia: Psychological and Pharmacological Treatment. New York: G u i l f o r d Press, 1981. 144 Barlow, D., Mavissakalin, M., & S c h o l f i e l d , L. Patterns of Desynchrony i n Agoraphobia: A Preliminary Report. Behaviour Research and Therapy, 1980, 18.(5) , 441-448. Beiman, I., O'Neil, P., Wachtel, D., Finge, E., Johnson, S., & Feuerstein, M. V a l i d a t i o n of a Self Report/Behavioural Subject S e l e c t i o n Proc-edure for Analog Fear Research. Behavior Therapy, 1978, _9_, 169-177. Bellack, A. Anxiety and Neurotic Disorders. In A. Kazdin, A. Bellack & M. Hersen (Eds.) New Perspectives i n Abnormal Psychology. New York: Oxford University Press, 1980 Benjamin, S., Marks, I., & Huson, J . Active Muscular Relaxation i n Desens-i t i z a t i o n of Phobic Patients. Psychological Medicine, 1972, _2_, 381-390. Bernstein, D. Behavioural Fear Assessment: Anxiety or A r t i f a c t ? In N. Adams & B. Unikel (Eds.) Issues and Trends i n Behavior Therapy. I l l i n o i s : Charles C. Thomas, 1973 Bernstein, D., & Paul, G. Some Comments on Therapy Analogue Research with Small Animal Phobias. Journal of Behaviour Therapy and Experimental  Psychiatry, 1971, 2j 225-237. Bersch, P. Eysenck's Theory of Incubation: A C r i t i c a l A n a l y s i s . Behaviour  Research and Therapy, 1980, 18, 11-17. Borkovec, T. The E f f e c t s of I n s t r u c t i o n a l Suggestion and P h y s i o l o g i c a l Cues on Analogue Fear. Behavior Therapy, 1973(a), 4_, 185-192. Borkovec, T. The Role of Expectancy and P h y s i o l o g i c a l Feedback i n Fear Research: A Review with Special Reference to Subject C h a r a c t e r i s t i c s . Behavior Therapy, 1973 (b), 4_, 491-505. Borkovec, T. P h y s i o l o g i c a l and Cognitive Processes i n the Regulation of Anx-i e t y . In G. Schwartz & S. Shapiro (Eds.) Consciousness and S e l f  Regulation: Advances i n Research, V o l . 1. New York: Plenum Press, 1976 Borkovec, T. Self E f f i c a c y : Cause or R e f l e c t i o n of Behaviour Change? Advances i n Behaviour Research and Therapy, 1977/8, 1_, 163-170. Borkovec, T., Grayson, J . , & Hrnnings, B. M i t i g a t i o n of False P h y s i o l o g i c a l Feedback E f f e c t s on Anxiety Via Cognitive Appraisal. Cognitive Therapy  and Research, 1979, 3.(4), 381-387. Borkovec, T., & O'Brien, G. Methodological and Target Behaviour Issues in Analogue Therapy Outcome Research. Progress in Behaviour Modification , 1976, 2_, 33-72. Borkovec, T., & Rachman, S. The U t i l i t y of Analogue Research, Behaviour  Research and Therapy, 1979, 17, 253-261. Borkovec, T., & Sides, J . C r i t i c a l Procedural Variables Related to the P h y s i o l o g i c a l E f f e c t s of Progressive Relaxation: A Review. Behaviour  Research and Therapy, 1979, _17_, 119-125. 145 Borkovec, T., Stone, N., O'Brien, G., & Kaloupek, D. Evaluation of a C l i n i c a l l y Relevant Target Behaviour for Outcome Research. Behavior Therapy, 1974, 5., 503-513. Borkovec, T., Weerts, T., & Bernstein, D. Assessment of Anxiety. In A. Ciminero, K. Calhoun & H. Adams (Eds.) Handbook of Behavioural  Assessment. New York: John Wiley & Sons, Inc., 1977 Boulougouris, J . , Rabavalis, A., & Stefa n i s , C. Psychophysiological Resp-onses i n Obsessive-Compulsive Patients. Behaviour Research and Therapy, 1979, 15, 221-230. Buck, R., M i l l e r , R., & Caul, W. Sex, Personality and P h y s i o l o g i c a l Variables in the Communication of Emotion v i a F a c i a l Expression. Journal of  Personality and S o c i a l Psychology, 1974, J33, 587-596. Chambless, D., Foa, E., Groves, G., & Goldstein, A. Flooding with B r e v i t a l i n the Treatment of Agoraphobia: Countereffective? Behaviour Res- earch and Therapy, 1979, 17, 243-251. Cohen, D. Comparison of Self Report and Overt Behavioural Procedures for Assessing Acrophobia. Behavior Therapy, 1977, 8_, 17-23. Connolly, J . Tonic P h y s i o l o g i c a l Responses to Repeated Presentations of Phobic S t i m u l i . Behaviour Research and Therapy, 1979, 17_, 189-196. Co s t e l l o , C. Symptoms of Psychopathology. New York: Wiley & Sons, 1970 Cox, D., Hallam, R., O'Connor, K., & Rachman, S. An Experimental Analysis of Fearlessness and Courage, i n press. Coyne, J . A C r i t i q u e of Cognitions as Causal E n t i t i e s with P a r t i c u l a r Ref-erence to Depression. Cognitive Therapy and Research, 1982, _6_(1) , 3-14. Craig, K., & Wood, K. Autonomic Components of Observers Responses to P i c t -ures of Homicide Victims and Nude Females. Journal of Experimental  Research i n Personality, 1971, _5_(4) , 304-309. Davidson, R. S p e c i a l i t y and Patterning of Biobehavioural Systems. American  Psychologist, 1978, _3_3, 430-436. De S i l v a , P., & Rachman, S. Is Exposure a Necessary Condition for Fear-Reduction?. Behaviour Research and Therapy, 1981, _19_, 227-232. Elder, J . , E d e l s t e i n , B., & Fremow, W. C l i e n t by Treatment Interactions i n Response A c q u i s i t i o n and Cognitive Restructuring Approaches. Cognitive  Therapy and Research, 1981, _5_, 202-210. E l l i s , A. A Note on the Treatment of Agoraphobics with Cognitive M o d i f i c a t i o n Versus Prolonged Exposure i n Vivo. Behaviour Research and Therapy, 1979, 1979, 17, 162-164. Emmelkamp, P. Self Observation Versus Flooding i n the Treatment of Agora-phobia. Behaviour Research and Therapy, 1974, 12, 229-237. Emmelkamp, P. Phobias: Theoretical and Behavioural Th e o r e t i c a l Considerat-ions. In J . Boulougouris & A. Rabavalis (Eds.) The Treatment of 146 Phobic and Obsessive Compulsive Disorders. Oxford: Pergamon Press, 1977 Emmelkmap, P. The Behavioural Study of C l i n i c a l Phobias. In M. Hersen, R. E i s l e r and P. M i l l e r (Eds. ). Progress i n Behaviour Modification . New York: Academic Press, 1979 Emmelkamp, P., & Emmelkamp-Benner, A. E f f e c r s of H i s t o r i c a l l y Portrayed Modelling and Group Treatment on Sekf Observation: A Comparison with Agoraphobics. Behaviour Research and Therapy, 1975, 13, 135-139. Emmelkamp, P., Kuipers, A., & Eggeraat, J . Cognitive M o d i f i c a t i o n versus Prolonged Exposure i n Vivo: A Comparison with Agoraphobics as Subjects. Behaviour Research and Therapy, 1978, 16., 33-41. Endler, N., & Okada, M. A Multidimensional Measure of T r a i t Anxiety: The S-R Inventory of General T r a i t Anxiousness. Journal of Consulting and  C l i n i c a l Psychology, 1975, 43_(3) , 319-329. Eysenck, H. The Learning Theory Model of Neurosis - A New Approach. Beh- aviour Research and Therapy, 1976, _14_, 251-267. Eysenck, H. Expectations As Causal Elements in Behavioural Change. Advances  i n Behaviour Research and Therapy, 1978, _1, 171-175. F e l t z , D. Path Analysis of the Causal Elements i n Bandura's Theory of Self E f f i c a c y and an Anxiety Based Model of Avoidance Behaviour. Journal  of Personality and S o c i a l Psychology, 1982, 42_(4) , 764-781. Foa, E., Jameson, J . , Turner, R., & Paynes, L. Massed versus Spaced Exposure Sessions i n the Treatment of Agoraphobia. Behaviour Research and  Therapy, 1980, 18., 333-338. ,Foa, E., & Steketee, G. Obsessive-Compulsives: Conceptual Issues and Treat-ment Interventions. In M. Hersen, R. E i s l e r and P. M i l l e r (Eds.) Progress i n Behaviour Mo d i f i c a t i o n . New York: Academic Press, 1979 Fowles, D., C h r i s t i e , M., Edelberg, R., Grings, W., Lykken, D., & Venables, P. Publi c a t i o n Recommendations for Electrodermal Measurements. The Society  for Psychophysiological Research, 1981, 18.(3), 232-239. Freedman, A., Dornbush, R., & Shapiro, B. Anxiety: Here Today and Here Tomorrow. Comprehensive Psychiatry, 1981, 2_2 (1) , 44-53. G a l a s s i , J . , F r i e r s o n , H., & Sharer, R. Behaviour of High, Moderate and Low Test Anxious Subjects During an Actual Test S i t u a t i o n . Journal of  Consulting and C l i n i c a l Psychology, 1981, 49_(1) , 51-62. Gauthier, J . , & Marshall, W. The Determination of Optimal Exposure to Phobic Stimuli i n Flooding Therapy. Behaviour Research and Therapy, 1977, 15, 403-410. Glass, G., & Stanley, J . S t a t i s t i c a l Methods i n Education and Psychology. New Jersey: P r e n t i c e - H a l l , Inc, 1970 Grey, S., Rachman, S., & Sartory, G. Return of Fear: The Role of Inhib-i t i o n . Behaviour Research and Therapy, 1981, _L9, 135-143. 147 Grey, S., Sartory, G., & Rachman, S. Synchronous and Desynchronous Changes during Fear Reduction. Behaviour Research and Therapy, 1979, 17, 137-147 Hafner, J . , & Marks, I. Exposure i n Vivo with Agoraphobics: Contributions of Diazapem, Group Exposure and Anxiety Evocation. Psychological Med- i c i n e , 1976, 6_, 71-88. Hallam, R., & Rachman, S. Courageous Actos ro Courageous Actors. Personal-i t y and Individual Differences, 1980, 1_(14) , 341-346. Hasset, J . A Primer of Psychophysiology. San Fransisco: W. H. Freeman & Co., 1978 Hersen, M. Self-Assessment of Fear. Behavior Therapy, 1973, _4_, 241-247. Hersen, M. Complex Problems Require Complex Solutions. Behavior Therapy, 1981, 12, 15-29. Hicks, D., & Sheinberg, R. Therapeutic Rationale, Monetary Incentive and Fear Behaviour: A Methodological Note on Analogue Research. Behaviour  Research and Therapy, 1976, _14, 83-84. Hodgson, R., & Rachman, S. Desynchrony i n Measures of Fear. Behaviour  Research and Therapy, 1974, 12, 319-326. Hollandsworth, J . , Glazeski, R., Kirkland, K., Hones, G., & Van Norman, L. An Analysis of the Nature and E f f e c t s of Test Anxiety: Cognitive , Behavioural and P h y s i o l o g i c a l Components. Cognitive Therapy and  Research, 1979,_3_(2), 165-180. Holroyd, K., Westbrook, T., Wolf, M., & Badhorn, E. Performance, Cognition, and P h y s i o l o g i c a l Responding i n Test Anxiety. Journal of Abnormal  Psychology, 1978, 87(4) , 442-451. Hugdahl, K. The Three-Systems Model of Fear and Emotion - A C r i t i c a l Exam-in a t i o n . Behaviour Research and Therapy, 1981, _19, 75-85. Hugdahl, K., & Karker, A. B i o l o g i c a l versus E x p e r i e n t i a l Factors i n Phobic Conditioning. Behaviour Research and Therapy, 1981, 19, 109-115. Hutchings, D., Denney, D., Basgall, J . , & Houston, B. Anxiety Management and Applied Relaxation in Reducing General Anxiety. Behaviour Research  and Therapy, 1980, 18, 181-190. Kanter, N., & G o l d f r i e d , M. Relative Effectiveness of Rational Restructuring and S e l f Control Desensitization i n the Reduction of Interpersonal Anx-i e t y . Behavior Therapy, 1979, 10, 472-190. Kaplan, R., & Litrownik, A. Some S t a t i s t i c a l Methods for the Assessment of Multiple Outcome i n Behavioural Research. Behavior Therapy, 1977, 8_, 383-392. Kazdin, A. Conceptual and Assessment Issues Raised by S e l f - E f f i c a c y Theory. Advances i n Behaviour Research and Therapy, 1978, _1, 177-185. 148 Kazdin, A. Evaluating the Generality of Findings i n Analogue Therapy Res-earch. Journal of Consulting and C l i n i c a l Psychology, 1978, 46(4), 673-686. Kazdin, A. Research Design in C l i n i c a l Psychology. New York: Harper and Row Pub., 1980 Kendrick, M. Reduction of Musical Performance Anxiety by A t t e n t i o n a l T r a i n -ing and Behavioural Rehearsal: An Exploration of Cognitive Mediational Processes (Doctoral D i s s e r t a t i o n , University of B r i t i s h Columbia, 1979) Kendrick, M., Craig, K., Lawson, D., & Davidson, P. Cognitive and Behav-i o r a l Therapy for Musical Performance Anxiety. Journal of Consulting  and C l i n i c a l Psychology, 1982, 50.(3), 333-362. Kirsch, I. 'Microanalytic' Analyses of E f f i c a c y Expectations as Predictors of Performance. Cognitive Therapy and Research, 1980, 4_(2), 259-262. Kozak, M., & M i l l e r , G. Hypothetical Constructs versus Intervening Variables: A Reappraisal of the Three-Systems Model of Fear. Behavioural Assess-ment, in press. Lande, S. P h y s i o l o g i c a l and Subjective Measures of Anxiety During Flooding. Behaviour Research and Therapy, 1982, 2Q_(1) , 81-88. Lang, P. The A p p l i c a t i o n of P s y c h o p h y s i o l o g i c a l Methods to the Study of Psychotherapy and Behaviour Mo d i f i c a t i o n . In A. Bergin and S. G a r f i e l d (Eds.) Handbook of Psychotherapy and Behaviour Change. New York: John Wiley & Sons, Inc., 1971 Lang, P. P h y s i o l o g i c a l Assessment of Anxiety and Fear. In J . Cone and R. Hawkins (Eds.) Behavioural Assessment: New Directions in C l i n i c a l  Psychology. New York: Brunner-Mazel Pub., 1977 Lang,P., Kozak, M., M i l l e r , G., Levin, D., & McLean, A. Emotional Imagery: Conceptual Structures and Pattern of Somato-Visceral Response, psycho- physiology, 1980, _17_(2) , 179-192. Lehrer, P., Schoicket, S., Carrington, P., & Woolfolk, R. Psychophysiological and Cognitive Responses to S t r e s s f u l S t imuli in Subjects P r a c t i c i n g Progressive Relaxation and C l i n i c a l l y Standardized Meditation. Behaviour  Research and Therapy, 1980, 18, 293-303. Leitenberg, H., Agras, S., Butz, R., & Wincze, J . Relationships between Heart Rate and Behavioural Change during the Treatment of Phobias. Journal of Abnormal Psychology, 1971, 78(1) , 59-68. Levine, B., Gorman, S., & Sherry, G. Reinforcement Expectancy E f f e c t s i n Phobia Research Volunteers. Psychological Record, 1978, 28, 585-588. Lick, J . The E f f e c t s of Pretreatment Demand C h a r a c t e r i s t c i s on Verbally Reported Fear. Behavior Therapy, 1977, Q_, 727-730. Lick, J . , & Katkin, E. Assessment of Anxiety and Fear. In M. Hersen and A. Bellack (Eds.) Behavioural Assessment: A P r a c t i c a l Handbook. Oxford: Pergamon Press, 1976 149 Marks, I. Perspectives on Flooding. Seminars i n Psychiatry, 1972, £, 129-138 Marks, I. Exposure: Conceptual Issues. In S. Agras (Ed.) Behaviour Mod-i f i c a t i o n : P r i n c i p l e s and C l i n i c a l A p p l i c a t i o n s. Boston: L i t t l e , Brown and Co., 1978 Marks, I., Marset, P., Boulougouris, J . , & Huson, J . P h y s i o l o g i c a l Accomp-animents of Neutral and Phobic Imagery. Psychological Medicine, 1971, 1, 299-307. Marshall, W., Gauthier, J . , C h r i s t i e , M., Cur r i e , D., Gordon, A. Flooding Therapy: Eff e c t i v e n e s s , Stimulus C h a r a c t e r i s t c i s , and the Value of B r i e f i n Vivo Exposure. Behaviour Research and Therapy, 1977, .15, 79-87. M a r z i l l i e r , J . Cognitive Therapy and Behavioural P r a c t i c e . Behaviour  Research and Therapy, 1980, 18, 249-258. M a r z i l l i e r , J . , C a r r o l l , D., & Newland, J . Self-Report and P h y s i o l o g i c a l Changes Accompanying Repeated Imagining of a Phobic Scene. Behaviour  Research and Therapy, 1979, 17_, 71-77. Mathews, A. Fear Reduction Research and C l i n i c a l Phobias. Psychological  B u l l e t i n , 1978, 85(2), 390-404. Mathews, A., & Shaw, P. Emotional Arousal and Persuasion E f f e c t s i n Flooding. Behaviour Research and Therapy, 1973, 11, 587-598. Mathews, A., & Shaw, P. Cognitions Related to Anxiety: A P i l o t Study of Treatment. Behaviour Research and Therapy, 1977, 15_, 503-505. Mavissakalin, M., & Barlow, D. Phobia: An Overview. In M. Mavissakalin and D. Barlow (Eds.) Phobia: Psychological and Pharmacological Treat-ment. New York: G u i l f o r d Press, 1981 Meichenbaum, D. Cognitive Behaviour Mo d i f i c a t i o n : An Integrative Approach New York: Plexium Press, 1978 Meyer, V., & Reich, B. Anxiety Management - The Marriage of P h y s i o l o g i c a l and Cognitive Variables. Behaviour Research and Therapy, 1978, 16, 177-182. Newman, A., & Brand, E. Coping Response Training versus i n Vivo Desensitizat-ion i n Fear Reduction. Cognitive Therapy and Research, 1980, _4_(4) , 397-407. Nunes, J . , & Marks, I. Feedback of True Heart Rate during Exposure i n Vivo. Archives of General Psychiatry, 1975, 32, 933-936. Ollendick, T., & Murphy, M. D i f f e r e n t i a l Effectiveness of Muscular and Cog-n i t i v e Relaxation as a Function of Locus of Control. Journal of Beh- aviour Therapy and Experimental Psychiatry, 1977, 8, 223-228. Ost, L., Jerremalm, A., & Johansson, J . Individual Response Patterns and the E f f e c t s of D i f f e r e n t Behavioural Methods i n the Treatment of S o c i a l Phobia. Behaviour Research and Therapy, 1981, .19, 1-16. 150 Paul, G. Insight versus Desensitization i n Psychotherapy. C a l i f o r n i a : Stanford U n i v e r s i t y press?, 1966 Popler, K. Agoraphobia: Indications for the Application of the Multimodal Behavioural Conceptualisation. The Journal of Nervous and Mental  Disease, 1977, 64(2), 97-101. Poser, E. The S e l f E f f i c a c y Concept: Some Th e o r e t i c a l , Procedural and C l i n i c a l Implications. Advances in Behaviour Research and Therapy, 1978, 1., 193-202. Rabavalis, A., Boulougouris, J . , & S t e f a n i s , C. Synchrony and Concordance on Subjective and Psychophysiological Measures after Beta Blockade and Flooding i n Obsessive-Compulsive Patients. In J . Boulougouris and A. Rabavalis (Eds.) The Treatment of Phobic and Obsessive Compulsive  Disorders. Oxford: Pergamon Press, 1976 Rachman, S. The Passing of the Two-Stage Theory of Fear and Avoidance: Fresh P o s s i b i l i t i e s . Behaviour Research and Therapy, 1976(a), 14, 125-131. Rachman, S. Courage, Fearlessness and Fear. New S c i e n t i s t , 1976(b), 271-273. Rachman, S. Fear and Courage. San Fransisco: W . Freeman & Co., 1978 Rachman, S. The Concept of Required Helpfulness. Behaviour Research and  Therapy, 1979 (b), 17., 1-6. Rachman, S. The Return of Fear. Behaviour Research and Therapy, 1979(a), .17, 164-166. Rachman, S. Emotional Processing. Behaviour Research and Therapy, 1980, 18, 51-60. Rachman, S., & Hodgson, R. Synchrony and Desynchrony in Fear and Avoidance. Behaviour Research and Therapy, 1974, 12, 311-318. Rachman, S., & Seligman, P. Unprepared Phobias: 'Be Prepared'. Behaviour  Research and Therapy, 1976, .14, 333-338. Reinking, R., & Kohl, M. E f f e c t s of Various Forms of Relaxation Training on P h y s i o l o g i c a l and S e l f Report Measures of Relaxation. Journal of  Consulting and C l i n i c a l Psychology, 1975, 43.(5) , 595-600. Ric k l e s , W . , & Day, J . Electrodermal A c t i v i t y i n Nonpalmar Skin S i t e s . Psychophysiology, 1968, 4_(4) , 321-345. Rosenthal, T. Bandura's S e l f E f f i c a c y Theory: Thought i s Father to the Deed. Advances i n Behaviour Research and Therapy, 1978, 1, 203-209. S a l l i s , J . , L i c h s t e i n , K., & McGlynn, F. Anxiety Response Patterns: A Comparison of C l i n i c a l and Analogue Populations. Journal of Behaviour Therapy and Experimental Psychiatry, 1980, 11 (3), 179-183. Sartory, G., Rachman, S., & Grey, S. An Investigation of the Relation between Reported Fear and Heart Rate. Behaviour Research and Therapy, 1977, 15, 151 435-438. Schwartz, G. Psychphysiological Foundations of Psychotherapy and Behaviour Change. In S. G a r f i e l d and A. Bergin (Eds.) .Handbook of.Psychotherapy  and Behaviour Change. Toronto: John Wiley & Sons, 1978 Schwartz, G., Davidson, R., & Goleman, D. Patterning of Cognitive and Somatic Processes i n the S e l f Regulation of Anxiety: E f f e c t s of Meditation versus Exercise. Psychosomatic Medicine, 1978, __0 (4) , 321-328. Shahar, A., & Merbaum, M. The Interaction between Subject C h a r a c t e r i s t i c s and S e l f Control Procedures i n the Treatment of Interpersonal Anxiety. Cognitive Therapy and Research, 1981,_5_(2), 221-224. Sherry, G., & Levine, B. An Examination of Procedural Variables i n Flooding Therapy. Behavior Therapy, 1980, 11, 148-155. Siddle, D., & Turpin, G. Measurement, Qu a n t i f i c a t i o n and Analysis of Cardiac A c t i v i t y . In I. Martin and P. Venables (Eds.) Techniques i n Psycho- physiology. Toronto: John Wiley & Sons, 1980 Smith, R., Diener, E., & Beaman, A. Demand C h a r a c t e r i s t i c s and the Behaviour-a l Measure of Fear i n Behaviour Therapy Analogue Research. Behavior  Therapy, 1974, 5_, 172-182. Spielberger, C , Gorsuch, R., & Lushene, R. S t a t e - T r a i t Anxiety Inventory  Manual. C a l i f o r n i a : Consulting Psychologists Press Inc., 1970 Stern, R., & Marks, I. B r i e f and Prolonged Flooding: A Comparison in Agora-phobic Patients. Archives of General Psychiatry, 1973, 28_, 270-276. Sutton-Simon, K., & G o l d f r i e d , M. Faulty Thinking Patterns i n Two Types of Anxiety. Cognitive Therapy and Research, 1979, 3.(2), 193-203. Thomas, M., & Rapp, N. P h y s i o l o g i c a l , Behavioural and Cognitive Changes Resulting from Flooding in a Monosymptomatic Phobia. Behaviour Research  and Therapy, 1977, 15., 304-306. Trudel, G. The E f f e c t s of Instructions, Level of Fear, Duration of Exposure and Repeated Measures on the Behavioural Avoidance Test. Behaviour  Research and Therapy, 1979, 17, 113-118. Venables, P., & C h r i s t i e , M. Electrodermal A c t i v i t y . In I. Martin and P. Venables (Eds.) Techniques in Psychophysiology. Toronto: John Wiley & Sons, 1980 Watson, J . , Gaind, R., & Marks, I. P h y s i o l o g i c a l Habituation to Continuous Phobic Stimulation. Behaviour Research and Therapy, 1972, 10, 269-278. Watson, J . , & Marks, I. Relevant and Irrelevant Fear in Flooding - A Cross-over Study of Phobic Patients. Behavior Therapy, 1971, 2_, 275-293. Wiggins, J . Personality and P r e d i c t i o n : P r i n c i p l e s of Personality Assess-ment. C a l i f o r n i a : Addison-Wesley Pub. Co., 1973 Wilson, G. The Importance of Being T h e o r e t i c a l : A Commentary of Bandura's 152 'Self E f f i c a c y : Towards a Unifying Theory of Behavioural Change'. Advances i n Behaviour Research and Therapy, 1978, 1, 217-230. Wilson, G., & O'Leary, K. P r i n c i p l e s of Behaviour Therapy. New Jersey: Prentice H a l l Inc., 1980 Wolpe, J . Self E f f i c a c y Theory and Psychotherapeutic Change: A Square Peg for a Round Hole. Advances i n Behaviour Research and Therapy, 1978, 1, 231-236. Woodward, R., & Jones, R. Cognitive Restructuring Treatment: A Controlled T r i a l with Anxious Patients. Behaviour Research and Therapy, 1980, 18, 401-407. Woolfolk, R., & Lazarus, A. Between Laboratory and C l i n i c : Paving the Two Way Street. Cognitive Therapy and Research, 1979, 3.(3) , 239-244. Appendix 1 I n i t i a l Contact My name i s Michelle Craske and I am a Masters student i n C l i n i c a l Psychology. For my Masters t h e s i s , I am researching musical performance anxiety. I am interested in the various thoughts and f e e l i n g s people experience during solo piano performances. I t i s sometimes found that the peculiar thoughts and f e e l i n g s experienced do not correspond. I want to examine that r e l a t i o n s h i p more p r e c i s e l y . The questionnaire that I ' l l ask you to f i l l in w i l l provide me with some information regarding these issues. Also, t h i s information may lead on to an empirical study, and in the event that I need to contact volunteers for p a r t i c i p a t i o n in that study, i t would be extremely useful to me i f , in addition to your name, you would also write your phone number in the top r i g h t hand corner. 154 Appendix 2 Report of Confidence As a Performer Name: Phone: This questionnaire i s composed of 30 items regarding your f e e l i n g s of cofidence as a solo performer on the piano. Try to decide whether "True" or "False" most represents your f e e l i n g s as associated with your most recent solo performance, and c i r c l e the appropriate l e t t e r . This i n f o r -mation i s completely c o n f i d e n t i a l . Work quickly - f i r s t impressions are most appropriate. Please respond to every statement. 1. I look forward to an opportunity to perform i n p u b l i c . T F 2. My legs tremble when I reach for the pedal. T F 3. I am in constant fear of forg e t t i n g the music. T F 4. Audiences seem f r i e n d l y when I come on stage. T F 5. When preparing a solo performance I am in a constant state of anxiety. T F 6. At the conclusion of a performance I f e e l that I have had a pleasant experience. T F 7. I d i s l i k e to use my body expressively. T F 8. My thoughts become jumbled and confused when I perform before an audience. T F 9. I have no fear of facing an audience. T F 10. Although I am nervous j u s t before getting up I soon forget my fears and enjoy the experience. T F 11. I face the prospect of performing with complete confidence T F 12. I f e e l that I am in complete possession of myself when performing. T F 13. I prefer to have the notes on the piano i n case I forget the music. T F 14. I l i k e to observe the reactions of the audience to my performance. T F 15. Although I perform well before my friends I freeze on stage. T F 16. I f e e l relaxed and comfortable while performing. T F 17. Although I do not enjoy performing i n public I do not p a r t i c u l a r l y dread i t . T F 18. I always avoid playing solos i n public i f possi b l e . T F 19. The faces of the audience are blurred when I look at them. T F 155 20. I f e e l disgusted with myself after performing before an audience. T F 21. I enjoy preparing for a solo performance. T F 22. My mind i s cl e a r when I face an audience. T F 23. My hands f e e l cold and weak before performing. T F 24. I perspire and tremble j u s t before performing. T F 25. My posture f e e l s strained and unnatural. T F 26. I am f e a r f u l and tense a l l the while I am performing before an audience. T F 27. I f i n d the prospect of performing mildly pleasant. T F 28. At the conclusion of my performance I f e e l that I would l i k e to continue performing. T F 29. I am t e r r i f i e d at the thought of performing before a group of people. T F 30. I have a f e e l i n g of alertness i n facing an audience. T F In addition please answer the following question: During any s i t u a t i o n that requires you to perform solo before an audience, do you experience anxiety which i s of s u f f i c i e n t i n t e n s i t y to be a source of concern for you? That i s , do you worry about f e e l i n g s of anxiety before or during performance? Yes No 156 Appendix 3 Subject Data R e l a t i v e l y Anxious RCP Age Sex Grade Level 1 11 20 M 8-9 2 11 18 F ARCT 3 12 18 F ARCT 4 12 21 F ARCT 5 12 17 F ARCT 6 12 19 F ARCT 7 13 23 F ARCT 8 14 19 M ARCT 9 14 18 M ARCT 10 14 21 F ARCT 11 14 30 F 10-11 12 14 21 M ARCT 13 14 20 F 10-11 14 16 16 F 8-9 15 17 22 F ARCT 16 19 21 F ARCT 17 19 22 F ARCT 18 22 24 F 10-11 19 23 33 F ARCT 20 27 18 F ARCT x = 15.5 x = 21.05 M = 4, F = 16 8-9 = 2 10-11 = 3 ARCT =15 R e l a t i v e l y Nonanxious RCP Age Sex Grade Level 1 1 22 M ARCT 2 2 18 M ARCT 3 3 19 F ARCT 4 4 23 F ARCT 5 4 18 M 10-11 6 5 19 F ARCT 7 6 20 F ARCT 8 6 20 M ARCT 9 6 25 M 10-11 10 6 20 F 10-11 11 7 18 M ARCT 12 7 18 F ARCT 13 8 18 F ARCT 14 8 21 M ARCT 15 9 19 M 8-9 16 9 23 F ARCT 17 9 18 M ARCT 18 10 18 M ARCT 19 10 26 M ARCT 20 10 20 M ARCT x = 6.5 x = 20.6 M = 12, F = 8 8-9 = 2 10-11 = 3 ARCT =15 157 Appendix 4 "You Are In Situations Involving Solo Piano Performances" (We are p r i m a r i l y interested i n your reactions i n general to those s i t u a t -ions that involve solo piano performances before an audience) Mark one of the f i v e a l t e r n a t i v e degrees of reaction or at t i t u d e by c i r c l i n g the appropriate number for each of the following nine items. 1. Seek experiences l i k e t h i s 1 2 3 4 5 Very much Not at a l l 2. Perspire 1 2 3 4 5 Not at a l l Perspire much 3. Having an "uneasy f e e l i n g " 1 2 3 4 5 Not at a l l Very much 4. Feel exhilerated and t h r i l l e d 1 2 3 4 5 Very much Not at a l l 5. Get f l u t t e r i n g f e e l i n g i n stomach 1 2 3 4 5 Not at a l l Very much 6. Feel tense 1 2 3 4 5 Not at a l l Very much 7. Enjoy these conditions 1 2 3 4 5 Very much Not at a l l 8. Heart beats faster 1 2 3 4 5 Not at a l l Much faster 9. Feel anxious 1 2 3 4 5 Not at a l l Very anxious 158 Appendix 5A SELF-EVALUATION QUESTIONNAIRE Developed by C. D. Spielberger, R. L. Gorsuch and R. Lushene STAI FORM X-1 N A M E : D A T E D I R E C T I O N S : A number of statements which people have used to describe themselves are given below. Read each state-ment and then blacken in the appropriate circle to the right of the statement to indicate how you feel right now, that is, at this moment. There are no right or wrong answers. Do not spend too much time on any one statement but give the answer which seems to describe your present feelings best. 1. I feel calm ® ® ® © 2. I feel secure © ® ® © 3. I am tense © ® ® © 4. I am regretful © ® ® © 5. I feel at ease © ® ® © 6. I feel upset © ® ® © 7. I am presently worrying over possible misfortunes © ® ® ® 8. I feel rested © ® ® © 9. I feel anxious © ® ® © 10. I feel comfortable © ® ® © 11. I feel self-confident © ® ® © 12. I feel nervous © ® © © 13. I am jittery © ® ® © 14. I feel "high strung" ® © ® © 15. I am relaxed © ® ® © 16. I feel content © ® ® © 17. I am worried © ® ® © 18. I feel over-excited and "rattled" © ® ® © 19. I feel joyful © ® ® © 20. I feel pleasant © ® ® © z o H > > r c o 3 m X > o o s > H B r s c o 159 Appendix 5B SUDS Scale Name: Please indicate the degree of anxiety that you are curr e n t l y experiencing on a xcale of 0 to 100 where 0 represents NO ANXIETY and 100 represents MAXIMUM ANXIETY. C i r c l e the appropriate number: 0 10 20 30 40 50 60 70 80 90 100 Appendix 5C Retrospective SUDS Scale Name: Please indicate the degree of anxiety you were/are experiencing at the following periods on a scale of 0 to 100 where 0 represents NO ANXIETY and 100 represents MAXIMUM ANXIETY: (1) Immediately af t e r you played the f i r s t note (2) Midway through your performance (3) Immediately p r i o r to completing your performance (4) Now 161 Appendix 5D Expectations of Personal E f f i c a c y Scale Name: Date: Part A How c e r t a i n are you that you could now perform the following tasks with your anxiety under control? For each task, please c i r c l e the appropriate number: (1) Playing your favourite piece for family members very 0 1 2 3 4 5 6 completely uncertain c e r t a i n (2) Playing a piece of music for a small group of friends very 0 1 2 3 4 5 6 completely uncertain c e r t a i n (3) Playing three pieces i n an exam with one examiner present very 0 1 2 3 4 5 6 completely uncertain c e r t a i n (4) Playing a single piece of music i n competition for a scholarship in a music f e s t i v a l very 0 1 2 3 4 5 6 completely uncertain c e r t a i n (5) Performing f i v e pieces of music from memory i n a one-person r e c i t a l to a f u l l house. This r e c i t a l i s to be reviewed i n the l o c a l newspapers. very 0 1 2 3 4 5 6 completely uncertain c e r t a i n Part B How c e r t a i n are you that you could not perform the following tasks before an audience with your anxiety under control? On the f i r s t four tasks, assume you have learned the music. For each task, please c i r c l e the appropriate number: (1) Playing i n a piano duet very 0 1 2 3 4 5 6 completely uncertain c e r t a i n (2) Accompanying a small choir of twenty members very 0 1 2 3 4 5 6 completely uncertain c e r t a i n 162 (3) Playing i n a chamber group of three: v i o l i n , c e l l o , and yourself at the piano very 0 1 2 3 4 5 6 completely uncertain c e r t a i n (4) Playing a piece of music from a d i f f e r e n t period (for example, Baroque, Romantic, Modern) from that which you have chosen to play today very 0 1 2 3 4 5 6 completely uncertain c e r t a i n (5) Sightreading a piece of music very 0 1 2 3 4 5 6 completely uncertain c e r t a i n 163 Appendix 5E S e l f - E f f i c a c y Scale Name: How c e r t a i n are you that you could now perform the following tasks i n a s i t u a t i o n i n which you are playing a solo musical piece with no one else present with your anxiety under control? For each task, please c i r c l e the appropriate number: (1) Playing a piece that you have practiced and have mastered well with the written music before you very 0 1 2 3 4 5 6 completely uncertain c e r t a i n (2) Playing a piece that you have practiced and have mastered well without the written music very 0 1 2 3 4 5 6 completely uncertain c e r t a i n (3) Playing a piece of music that you have practiced very l i t t l e and have mastered to some extent with the music before you very 0 1 2 3 4 5 6 completely uncertain c e r t a i n (4) Playing a piece of music that you have practiced very l i t t l e and have only mastered to some extent without the written music very 0 1 2 3 4 5 6 completely uncertain c e r t a i n (5) Playing a piece of music that you have not practiced and have not mastered with the written music before you very 0 1 2 3 4 5 6 completely uncertain c e r t a i n 164 Appendix 5F S e l f - E f f i c a c y Scale Name: How c e r t a i n are you that you could perform the following tasks i n a s i t u a t i o n i n which you are playing a solo musical piece before an audience who are evaluating your performance with your anxiety under control? For each task, please c i r c l e the appropriate number: (1) Playing a piece that you have practiced and have mastered well with the written music before you very 0 1 2 3 4 5 6 completely uncertain c e r t a i n (2) Playing a piece that you have practiced and have mastered well without the written music very 0 1 2 3 4 5 6 completely uncertain c e r t a i n (3) Playing a piece of music that you have practiced very l i t t l e and have mastered to some extent with the music before you very 0 1 2 3 4 5 6 completely uncertain c e r t a i n (4) Playing a piece of music that you have practiced very l i t t l e and have only mastered to some extent without the written music very 0 1 2 3 4 5 6 completely uncertain c e r t a i n (5) Playing a piece of music that you have not practiced and have not mastered with the written music before you very uncertain 0 1 2 3 4 5 6 completely c e r t a i n 165 Appendix 5G Performance Anxiety Self-Statement Scale Name: On the following pages are a number of thoughts which you may have had before, during or a f t e r your performance you j u s t gave. Read each item c a r e f u l l y and decide how frequently you were thinking that p a r t i c u l a r thought. For example: I was a f r a i d I was going to make a l o t of before during mistakes. ( ) ( ) Please follow these i n s t r u c t i o n s very c a r e f u l l y : 1. F i r s t read the statement. 2. Next decide on how frequently the thought occurred during each of the periods of time, and provide that information by entering one of the numbers from the following scale: 0 1 2 3 4 5 6 never almost continuously If the thought did not occur to you at any time, place zero in a l l the parentheses as follows: I was a f r a i d I was going to make a l o t of before during mistakes. ( 0 ) ( 0 ) The following example provides an i l l u s t r a t i o n of a thought that occurred quite frequently both before and during a performance but more frequently during. I t did not occur a f t e r . I noticed that my knees were shaking. before during after ( 4 ) ( 5 ) ( 0 ) Please t r y to answer as accurately as possi b l e . 166 1 2 3 4 never 1. I was reasoning that I was well prepared, 1 my fingers were in good shape, and I was going to do the best I could. 2. I was confident I would play w e l l . 3. I l o s t i n t e r e s t in what I was doing. 4. I had a f e e l i n g that I wasn't going to play w e l l . 5. I was thinking that I could hardly wait t i l l i t was over. 6. I was thinking about the music I was going to play - i t s mood, tempo and f e e l i n g . 7. I was thinking that I was crazy to be there. I was wishing I hadn't come. 8. I was panicking and f e e l i n g completely out of c o n t r o l . 9. I wished t h i s performance were taking place at another time. 10. I was thinking that I was i n c o n t r o l and could meet t h i s challenge. 11. I was not thinking of the past or too far ahead, but was focusing on what I was playing at the moment. 12. I f e l t as i f everything were unreal -as i f I were i n a dream. 13. I was f e e l i n g as i f I'd l i k e to get up and leave. 14. I f e l t I was playing w e l l . 15. I was r e a l l y enjoying myself - thinking I was glad to have the chance to play t h i s music. almost continuously before ( ) before ( ) before ( ) before ( ) before ( ) before ( ) before ( ) before ( ) before ( ) before ( ) before ( ) during ( ) during ( ) during ( ) during ( ) during ( ) during ( ) during ( ) during ( ) during ( ) during ( ) during ( ) during ( ) af t e r ( ) 167 0 1 2 3 4 5 6 never almost continuously 16. I was aware that I f e l t relaxed and comfortable. before during ( ) ( ) afte r ( ) 17. I was worried about disappointing my teacher or my r e l a t i v e s . before during ( ) ( ) afte r ( ) 18. I was f e e l i n g happy. before during ( ) ( ) after ( ) 19. I was thinking how much I l i k e d the piece of music I had chosen to play. before during ( ) ( ) after ( ) 20. I was concerned that the other performers would play much better than I. before during ( ) ( ) afte r ( ) 21. I f e l t that I had been i n c o n t r o l of things throughout the performance. during ( ) afte r ( ) 22. I was thinking that the whole things was a complete d i s a s t e r . during ( ) after ( ) 23. I f e l t that I never again wanted to play i n front of other people. during ( ) afte r ( ) 24. I knew that my mistakes were only a small part of the whole performance. during ( ) afte r ( ) 25. I f e l t l i k e a complete f a i l u r e . during ( ) after ( ) 26. I couldn't believe that I was making mistakes. I was so embarrassed. during ( ) after ( ) 27. I f e l t that I'd t r i e d to communicate something and wasn't j u s t concerned with playing c o r r e c t l y . during ( ) after ( ) 28. I f e l t that an audience would have enjoyed my playing. during ( ) after ( ) 29. I f e l t I had l e t myself down. afte r ( ) 30. I was f e e l i n g pleased with the way I'd played. after ( ) 31. I was ashamed of myself and didn't want to face anyone. aft e r ( ) 32. I r e a l l y enjoyed performing and was kind after of sorry i t was over. ( ) 168 Appendix 5H Performance Anxiety Self-Statement Scale Name: On the following pages are a number of thoughts which you may have had before, during or aft e r your performance you ju s t gave. Read each item c a r e f u l l y and decide how frequently you were thinking that p a r t i c u l a r thought. For example; I was a f r a i d I was going to make a l o t of before during mistakes. ( ) ( ) Please follow these i n s t r u c t i o n s very c a r e f u l l y : 1. F i r s t read the statement. 2. Next decide on how frequently the thought occurred during each of the periods of time, and provide that information by entering one of the numbers from the following scale: never almost continuously If the thought did not occur to you at any time, place a zero in a l l the parantheses as follows: I was a f r a i d I was going to make a l o t of before during mistakes. *( 0 ) ( 0 ) The following example provides an i l l u s t r a t i o n of a thought that occurred quite frequently both before and during a performance but more frequently during. I t did not occur a f t e r . I noticed that my knees were shaking. before during a f t e r ( 4 ) ( 5 ) ( 0 ) Please t r y to answer as accurately as possi b l e . 169 0 1 2 3 4 never I was reasoning that I was well prepared, my fingers were in good shape, and I was going to do the best I could. almost continuously before ( ) I was confident I would play well. 3. I l o s t i n t e r e s t i n what I was doing. 4. I had a f e e l i n g that I wasn't going to play w e l l . 5. I was thinking that I could hardly wait t i l l i t was over. 6. I was thinking about the music I was going to play - i t s mood, tempo and f e e l i n g . before ( ) before ( ) before ( ) before ( ) before ( ) during ( ) during ( ) during ( ) I was thinking that I was crazy to be up there i n front of those people. I was wishing I hadn't come. 9. I wished t h i s performance were taking place at another time. 10. I was thinking that I was i n c o n t r o l and could meet t h i s challenge. 11. I was not thinking of the past or too far ahead, but was focusing on what I was playing at the moment. 12. I f e l t as i f everything were unreal -as i f I were in a dream. before ( ) before ( ) before ( ) during ( ) I was panicking and f e e l i n g completely before during out of c o n t r o l . ( ) ( ) during ( ) during ( ) during ( ) during ( ) 13. I was f e e l i n g as i f I'd l i k e to get up and leave. 14. I f e l t I was playing w e l l . 15. I was r e a l l y enjoying myself - thinking I before was glad to have the chance to share t h i s ( ) music with those l i s t e n i n g . during ( ) during ( ) during ( ) after ( ) 170 0 1 2 3 4 5 6 never almost continuously 16. I was aware that I f e l t relaxed and before comfortable. ( ) during ( ) afte r ( ) 17. I was worried about disappointing my before r e l a t i v e s or my teacher. ( ) during ( ) afte r ( ) 18. I was f e e l i n g happy. before ( ) during ( ) afte r ( ) 19. I was thinking how much I l i k e d the before piece of music I had chosen to play. ( ) during ( ) afte r ( ) 20. I was concerned that the other performers before would play much better than I. ( ) during ( ) after ( ) 21. I f e l t that I had been i n c o n t r o l of before things throughout the performance ( ) during ( ) after ( ) 22. I was thinking that the whole thing was a complete d i s a s t e r . during ( ) afte r ( ) 23. I f e l t that I never again wanted to play i n front of other people. during ( ) afte r ( ) 24. I knew that my mistakes were only a small part of the whole performance. during ( ) afte r ( ) 25. I f e l t l i k e a complete f a i l u r e . during ( ) afte r ( ) 26. I couldn't believe that I was making mistakes in front of those people. I was so embarrassed and was wondering what people would think of me. during ( ) afte r ( ) 27. I f e l t that I'd t r i e d to communicate something and wasn't j u s t concerned with playing c o r r e c t l y . during ( ) after ( ) 28. I f e l t that the people i n the audience enjoyed my playing. during ( ) afte r ( ) 29. I f e l t I had l e t myself down. after ( ) 30. I was f e e l i n g pleased with the way I'd played. af t e r ( ) 31. I was ashamed of muself and didn't want to face anyone. after ( ) 32. I r e a l l y enjoyed performing and was kind of sorry i t was over. after ( ) 171 (1) Appendix 6 D e f i n i t i o n s of Performance Quality Dimensions Touch - t e c h n i c a l e f f i c i e n c y of the performer to achieve c l a r i t y and sound q u a l i t y ( s l u r s , staccatos, legato l i n e s and accents, and p e d a l l i n g ) . (2) Phrasing - c l a r i t y , subtlety, c o n t r o l , length, o r i g i n a l i t y and v a r i e t y of phrasing; degree to which the performer i s aware of and shapes phrases - expressiveness. (3) Pitch and Omission - i n c o r r e c t , omitted or blurred notes or chords (taking into account the r e l a t i v e t e c h n i c a l d i f f i c u l t y of the piece and the importance of the i n c o r r e c t note to the presentations). (4) Rythym - time value of note or r e s t ; c l a r i t y with which the indicated metrical rhythm i s rendered (misplaced b a r l i n e s or s t r e s s e s ) ; c l a r i t y of the rhythmic harmonic progression; rubato; o v e r a l l rhythmic timing of sections (placements and a r r i v a l s of phrases); coordination of attack. (5) Tempo - steadiness of tempo; s t y l i s t i c appropriateness of tempos; consistency and appropriateness of speeds of passages. (6) Dynamics - r e l a t i v e effectiveness of dynamics choses (timing, v a r i e t y , s t y l i s t i c appropriateness and dynamic range); balancing of chords and l i n e s ; i n t e r p r e t a t i o n of indicated (in score) dynamics or accents. (7) Memory - a b i l i t y to remember the piece, and to recover from a memory s l i p (taking into account the importance of the s l i p to the passage). (8) O v e r a l l Quality - the extent to which the performance i s convincing i n i t s t e c h n i c a l , s t y l i s t i c and musical q u a l i t i e s ; the emotional conviction of the performer. Appendix 7 Performance Quality Rating Rater: Subject Number: Rate each dimension on the following scale: 1 2 3 4 5 6 7 8 9 10 T o t a l l y Average Exc e l l e n t Unacceptable 1. Pitch and Omission 2. Rhythm 3. Tempo 4. Dynamics 5. Touch 6. Phrasing 7. Memory 8. Ov e r a l l Quality Grand Total Rate the vigour with which the piece i s played on a a 1 to 10 point s c a l e , where 1 represents VERY LITTLE VIGOUR and 10 represents VERY MUCH VIGOUR by c i r c l i n g the appropriate number: 1 2 3 4 5 6 7 8 9 10 Very Average Very L i t t l e Much 173 Appendix 8 Timed Behavioural Ch e c k l i s t D e f i n i t i o n s of Items (1) Face Deadpan - expressionless face; l i p s are not curled up and are i n t h e i r usual p o s i t i o n ; brow i s not furrowed; eyes are not squinted; face i s not d i s t o r t e d ; eyebrows are not r a i s e d . Deadpan face should p e r s i s t through-out the e n t i r e i n t e r v a l . (2) Moistens Lips - l i p s are parted, tongue protrudes and rests on or runs along e i t h e r top or bottom l i p s . This i s rated i f i t occurs at any time during i n t e r v a l . (3) L i f t s Shoulders - shoulders are l i f t e d at l e a s t one inch i n an upward d i r e c t i o n . Shoulders may be l i f t e d from a drooping or s e m i - l i f t e d p o s i t i o n . This i s rated i f i t occurs at any time during i n t e r v a l . (4) S i f f Back and Neck - back and neck are i n f l e x i b l e and r i g i d , though t h i s does not require an upright p o s i t i o n . Back does not sway from side to side, or back and f o r t h . Neck does not jerk back and f o r t h or from side to s i d e . This i s rated i f i t p e r s i s t s throughout the e n t i r e i n t e r v a l . (5) S t i f f Arms - arms are r i g i d and i n f l e x i b l e , though t h i s does not include gross movements to reach the keys. Arms are not f l a c c i d . This i s rated i f i t p e r s i s t s throughout the e n t i r e i n t e r v a l . Rating Sheet Rater: Subject Number: Inter v a l 1 Interval 2 Interval 3 L i f t s Shoulders S t i f f Back and Neck S t i f f Arms Dace Deadpan Moistens Lips Appendix 9 Audience C h e c k l i s t Rater: Subject: Tick those behaviours which were present at any time during the performance Knees Tremble L i f t s Shoulders S t i f f Back and Neck Hands Tremble S t i f f Arms Pace Deadpan Moistens Lips 175 Appendix 10 Phone Contact Instructions H e l l o , my name i s . I am c a l l i n g i n r e l a t i o n to the study on musical performance which i s a followup on the questionnaire that you recently f i l l e d i n . Do you remember? This study might be of some i n t e r e s t to you as a performer. We are p a r t i c u l a r l y interested i n the s p e c i f i c thoughts that people have during t h e i r performances, and how they r e l a t e to the types of f e e l i n g s experienced. So, the study would provide you with some i n t e r e s t i n g personal feedback on you own reactions during performances, and i t might lead to our being able to help those with serious problems. I ' l l j u s t b r i e f l y describe the study; i t would involve coming to the music department at U.B.C. on two occasions, each l a s t i n g approximately 40 minutes (during the weekends of February 27th to 28th and March 6th to 7th). You w i l l be asked to play the same musical piece -of your own choice - on both occasions. The f i r s t occasion w i l l be a p r a c t i c e session for you to become f a m i l i a r with the procedures, and w i l l j u s t involve performing on your own. On the second occasion you w i l l be asked to play before a small group of people. Also, there w i l l be several questionnaires to f i l l i n on both occasions. Once the study i s f i n i s h e d we can t e l l you what we're interested in and how i t r e l a t e s to your performance. Also, we are o f f e r i n g a payment of $5.00 per session. Would you be interested i n p a r t i c i p a t i n g ? (If the answer i s "YES") Good. F i r s t l y , then, I need to obtain some information - mainly for the purposes of experimental design. (ask the subject d e t a i l s for the following. NB: "grade l e v e l " r e f e r s to the grade of music that they are now mastering) Age: Grade Level: Postal Address: Teachers Name: Between now and the next time I contact you to schedule dates and times could you please choose a piano piece that you have mastered and are f a m i l -ia r with to the extent that you f e e l comfortable playing i t from memory. It must only be 3 to 5 minutes i n length. We w i l l be contacting you within the next 2 weeks. Thankyou. (If the answer i s "NO") Well, i f you do change your mind, please f e e l free to contact us by ringing the C l i n i c a l Psychology Department of U.B.C. 228 5581 and asking for Michelle. Thankyou. Appendix 11 Subject Data Sheet Subject Number: __ . _ Name: — — • Sex: . ! Age: Grade Level: — Score on RCP: — Name of Piece: , _ • Teacher's Name: _____ — Postal Address: . — — — Schedule: Session 1 Date __ Session 2 Date T i m e Appendix 12 Performance Schedule 177 Subject 1 2 3 4 5 6 7 8 9 10 RA, 9.40 9.55 10.10 10.25 10.40 10.55 11.10 11.25 11.40 11.55 9.55 10.10 10.25 10.40 10.55 11.10 11.25 11.40 11.55 12.10 Researcher 9.55 - 10.10 10.10 - 10.25 10.25 - 10.40 10.40 - 10.55 10.55 - 11.10 11.10 - 11.25 11.25. - 11.40 11.40 - 11.55 11.55 - 12.10 12.10 - 12.25 RA 2 10.10 - 10.20 10.25 - 10.35 10.40 - 10.50 10.55 - 11.05 11.10 - 11.20 11.25 - 11.35 11.40 - 11.50 11.55 - 12.05 12.10 - 12.20 12.25 - 12.35 11 12 13 14 15 16 17 18 19 20 1.10 1.25 1.40 1.55 2.10 2.25 2.40 2.55 3.10 3.25 1.25 1.40 1.55 2.10 2.25 2.40 2.55 3.10 3.25 3.40 1.25 1.40 1.55 2.10 2.25 2.40 2.55 3.10 3.25 3.40 1.40 1.55 2.10 2.25 2.40 2.55 3.10 3.25 3.40 3.55 1.40 1.55 2.10 2.25 2.40 2.55 3.10 3.25 3.40 3.55 1.50 2.05 2.20 2.35 2.50 3.05 3.20 3.35 3.50 4.05 178 Appendix 13 F i r s t Set of Subject Instructions - Alone Condition The purpose of today's session i s e s s e n t i a l l y to allow you to become used to the s i t u a t i o n and the procedures that w i l l be involved i n your performance next time. So, we w i l l be going through exactly the same procedures, even though we w i l l not be evaluating the data that we obtain. To gain maximum benefit from t h i s p r a c t i c e session you should t r y to perf-orm as i f you were at home alone. F i r s t l y , I w i l l attach some electrodes which provide measures of heart rate and so on. For today, t h i s i s to ensure that the instruments are operating as they should and to prepare you for the next time, rather than for actual measurement. Then, I ' l l give you some questionnaires to f i l l i n - any information that you provide w i l l remain c o n f i d e n t i a l . Then y o u ' l l be taken to the studio, and when you are seated at the piano alone announce your name and the t i t l e of the piece as i f there were an audience present. You w i l l then hear one tone, followed by another tone. Commence playing when you hear the second tone, before which time you should remain as s t i l l as po s s i b l e . Also, during the per-formance there w i l l be another tone which i s a necessary design factor for the measures we are taking, but ju s t continue playing. When you f i n i s h , Michelle w i l l bring you back here to complete some more questionn-a i r e s . Are there any questions? 179 Appendix 14 Consent Form 1 I, , v o l u n t a r i l y give my consent to p a r t i c i p a t e i n the research project examining musical performance to be conducted at the Univ e r s i t y of B r i t i s h Columbia during the period January to May 1982. The procedures to be followed and the i r purposes have been explained to me and I understand them. I have, made a commit-ment to attend two sessions, each approximately 40 minutes i n length. I know that I w i l l be required to play a solo musical piece on both occasions and w i l l have electrodes attached for the purposes of adaptation i n the f i r s t session and for p h y s i o l o g i c a l data c o l l e c t i o n i n the second session and w i l l complete various s e l f - r e p o r t measures on both occasions. I also know that on the second occasion I w i l l be performing before an audience who w i l l be evaluating my playing and that my performance w i l l be audiotaped and videotaped. I know that any information gathered from me during t h i s study w i l l remain c o n f i d e n t i a l . If I wish to terminate p a r t i c i p a t i o n i n the study, I know that I am free to do so. Signature Date Appendix 15 Debriefing Instructions i n Alone Condition In order to make you f e e l as much at ease as possib l e , we had to ensure that t h i s s i t u a t i o n was i n no way s t r e s s f u l for you. So, we previously t o l d you that we would not be evaluating the data obtained. I t i s very necessary that we obtain a baseline of your reactions without any s t r e s s , which i s used i n analysing the data from your next performance. So, we are interested i n your responses during today's session, and for that reason your performance was videotaped. The only people who w i l l be viewing the videotape are the p r i n c i p l e researchers i n order to make some objective r a t i n g s . Do you have any objections to us using that information? I f so, we can erase the tape. (If the answer i s "NO") It i s very important that you keep t h i s c o n f i d e n t i a l , so could you please not mention the videorecording to anyone. 181 Appendix 16 Consent Form 2 I, , v o l u n t a r i l y give my consent to allow the video and audio recording and p h y s i o l o g i c a l data taken from my performance on the to be retained for an a l y s i s . I am aware that, had I wished, the data could have been deleted. I know that I was not previously informed of these measures i n order to minimize the anxiety I might experience. I know that i n my next performance I w i l l play before an audience who w i l l evaluate my playing while p h y s i o l o g i c a l data i s being c o l l e c t e d and that my performance w i l l be audio and video taped. I also know that any i n f o r -mation gathered from me during t h i s study w i l l remain c o n f i d e n t i a l . If I wish to terminate p a r t i c i p a t i o n i n t h i s study, I know that I am free to do so. Signature Date Appendix 17 F i r s t Set of Subject Instructions - Audience Condition Today you w i l l be performing before a small audience of people who are accomplished p i a n i s t s and experts i n behavioural assessment. They have copies of your music and w i l l be evaluating your performance. In addi t i o n , your performance w i l l be videotaped - t h i s time you w i l l be able to see the camera. Otherwise, we w i l l be following exactly the same procedures as l a s t time; f i r s t l y you w i l l f i l l i n some questionnaires and I ' l l attach the electrodes, and then Michelle w i l l take you into the studio. As before, begin when you hear the second tone before which time you should have already announced your name and the t i t l e of your piece. Also, there w i l l be another tone during the performance for reasons to do with p h y s i o l o g i c a l measurement, but j u s t continue playing. The audience have been instructed not to applaud for standardization purposes, so do not take the absence of applause as a r e f l e c t i o n oh your performance i n any way. When you f i n i s h , Michelle w i l l return to unhook you. Are there any questions? Appendix 18 Video-Camera Questionnaire Were you aware that your f i r s t performance was videotaped before being informed by the research a s s i s t a n t upon completion of that performance Aware Suspicious Unaware 184 Appendix 19 Second Set of Subject Instructions - Audience Condition We would l i k e to thank you for spending time and e f f o r t to p a r t i c i p a t e i n t h i s study. We hope that i t has not been too s t r e s s f u l for you, but instead has been an i n t e r e s t i n g experience. Michelle w i l l be contacting you as soon as a l l of the r e s u l t s have been analysed to discuss with you the findings o v e r a l l and to give you some feedback on your own resp-onses. I t may take some time, but she w i l l d e f i n i t e l y contact you as soon as p o s s i b l e . Thankyou again. 185 Appendix 20 Debriefing Letter Dear , I must f i r s t apologise for the long delay i n contacting you, but there was a great deal of data to analyse . I should begin by explaining more f u l l y the nature of my research project. E s s e n t i a l l y , I am interested i n the manner i n which the three d i f f e r e n t ways of manif-esti n g anxiety are related to each other, to the.degree of anxiety experien-ced while performing, and to one's confidence in one's a b i l i t y to perform. The three major ways are: (1) P h y s i o l o g i c a l e.g. sweating, increased heart rate (2) C o g n i t i v e / s e l f reported e.g. s t a t i n g to oneself (by private thoughts) how anxious one f e e l s (3) Behavioural e.g. s t i f f posture, shaking As you are aware, I took measures in each of these areas i n the study: (1) heart rate, skin resistance and r e s p i r a t i o n were measured while you were playing (2) you completed many questionnaires before and a f t e r playing (3) and from the videos, objective ratings were made of your appear-ance and playing. I remind you that a l l of t h i s information has been and w i l l be kept conf-i d e n t i a l . The degree of anxiety i n t h i s study was contrasted in two conditions -as you remember - once alone, and once with an audience. I t was assumed that for some of you the audience would be more s t r e s s f u l than for others - and i t was. The audience, although a l l interested i n piano and a l l of whom enjoyed the performances very much, were not a l l accomplished p i a n i s t s , as that was not possible for me to arrange. However, as one of the measures your performances were rated on very objective ratings by two very accomp-l i s h e d p i a n i s t s who l i s t e n e d to the audiotapes. The confidence aspect was measured by several of the questionnaires that you completed. One hypothesis that I'm t e s t i n g i s that s e l f confidence predicts the l e v e l of anxiety experienced i n each of those three ways. That i s , low confidence would p r e d i c t a l o t of anxiety. The contrasting hypothesis i s that, because those three aspects of anxiety can change quite independently 186 of each other,.confidence can not always p r e d i c t a l l of them. The r e s u l t s suggest that the l a t t e r i s more often the case than the former. That i s , while confidence may p r e d i c t the way one appears ( i . e . relaxed or tense), i t may not p r e d i c t the amount of p h y s i o l o g i c a l arousal experienced or subjective d i s t r e s s reported. This research has implications for treatment for people experiencing extremely d i s a b l i n g anxiety. That i s , to f i n d the method most s u i t a b l e for t r e a t i n g a p a r t i c u l a r person, one must f i r s t f i n d how the person man-i f e s t s their anxiety. If they mostly become p h y s i c a l l y aroused when anxious, then a treatment focusing upon muscle tension may produce the best r e s u l t s . If t h e i r anxiety i s most obvious i n t h e i r behaviour, then a behavioually oriented treatment method may be best, and so on. Though your l e v e l s of anxiety may not be extremely d i s a b l i n g , i t i s of i n t e r e s t to know more about our own reactions. So, here i s some personal feedback about your reactions when you played i n front of the audience: (1) Your hear rate at r e s t i n g l e v e l was beats per minute, averaged at beats per minute during playing and peaked at beats per minute. Most of the 40 people who p a r t i c i p a t e d i n the study experienced heart rate increases while playing, and much of that i s due to the amount of phy s i c a l energy involved. (2) Your r e s p i r a t i o n rate (breaths per minute) changed from (while resting) to (while p l a y i n g ) . Again, nearly everyone's breathing became much faster and more shallow when they began playing. (3) Your skin resistance (a measure of sweating) was % of the average. Your p h y s i o l o g i c a l r e s u l t s suggest that (4) The r a t i n g of your appearance while playing ( i . e . the degree to which you ilooked relaxed) was , and the average was . That i s , . (5) Your s e l f reported l e v e l of anxiety was % of the aver-age. That i s , . (6) Your s e l f reported l e v e l of s e l f confidence was % of the average. That i s , . I t therefore appears that 187 I would l i k e to thank you once again for helping me i n my research. I hope t h i s information has been of some i n t e r e s t to you. If you would l i k e to arrange some time to view your own videotapes, or to talk further about the research, don't hesitate to c a l l me at 733 9071 (evenings). Sincerely, Michelle Craske 188 Appendix 21 Absolute Values of Heart Rate and Skin Conductance Level R e l a t i v e l y Anxious R e l a t i v e l y Nonanxious Variable M SD M SD SCL1 (/.mhos) Alone 6.64 2.78 ' 6.14 3.46 Audience 6.52 3.60 8.03 5.76 SCL3 (/ttmhos) Alone 8.10 3.53 8.62 5.75 Audience 7.91 4.91 12.53 10.64 SCLE (/tmhos) Alone 7.52 3.40 9.34 10.10 Audience 7.66 4.12 12.82 11.92 HR-pre Alone 96.54 17.51 89.90 12.29 Audience 108.81 17.67 101.56 23.37 HR-during Alone 103.57 16.71 103.69 19.56 Audience 125.08 16.85 121.21 26.62 Most Anxious Least Anxious M SD M SD SCL1 C«mhos) Alone 7.39 3.06 4.06 1.34 Audience 6.79 3.31 5.28 2.69 SCL3 yianhos) Alone 7.85 3.26 7.98 6.17 Audience 7.41 2.72 8.73 7.90 SCLE (/mhos) Alone 7.21 3.12 9.45 13.98 Audience 7.05 2.74 9.53 11.97 HR-pre Alone 92.33 14.84 87.64 10.31 Audience 106.38 19.84 98.90 26.30 HR-during Alone 98.14 14.41 103.21 19.81 Audience 120.98 14.02 120.13 26.69 189 Appendix 22 Summary of Repeated Measures Univariate Analyses of Variance Variable SS MS Error SS MS F jp (df=l,38) PQ Group 809.63 809.63 13889.54 365.51 Condition 0.08 0.08 Group x Condition 102.38 102.38 838.67 22.07 2.22 ns 0.00 ns 4.64 <.04 TC Group 0.1125 0.1125 98.38 2.59 0.04 ns Condition 1.51 1.51 - - 0.37 ns Group x Condition 3.61 3.61 157.38 4.14 0.87 ns State Group 1665.31 1665.31 5971.88 Condition 324.01 324.01 Group x Condition 148.51 148.51 1792.98 157.16 10.06 <.01 6.87 <.02 47.18 3.15 <.084 Level Mean Group Condition Group x Condition Strength Mean Group Condition Group x Condition 8.02 0.006 0.05 1102.61 5.71 32.51 8.02 0.006 0.05 1102.61 5.71 32.51 15.07 3.72 1154.83 131.26 0.40 20.23 <.001 0.06 ns 0.10 0.51 ns 30.39 36.28 <.001 1.50 ns 3.45 9.41 <.01 SUDS Group 1361.25 1361.25 37287.50 981.25 Condition 2311.25 2311.25 Group x Condition 361.25 361.25 8777.50 230.99 1.39 ns 10.01 <.01 1.56 ns RSUDS Group Condition Group x Condition SUDS Tot a l Group Condition Group x Condition 9439.51 9439.51 496366.98 13062.29 14661.10 14661.10 8466.61 8466.61 90694.78 2386.71 0.72 ns 6.14 <.02 3.55 ns 17970.00 17970.00 689707.48 18150.197 0.99 28614.60 28614.60 12325.60 12325.60 111162.27 2925.32 ns 9.78 <.04 4.21 <.05 SS Group Condition Group x Condition 58644.40 58644.40 196840.35 832.05 832.05 320.00 320.00 55026.95 5180.01 11.32 <.01 0.57 ns 1448.08 0.22 ns RSCL1 Group 0.62 0.62 Condition 0.07 0.07 Group x Condition 0.01 0.01 3.06 0.08 0.77 ns 0.73 ns 3.63 0.10 0.11 ns 190 Error Variable SS MS SS MS RSCL3 Group 0.06 0.06 2.17 0.06 1.13 ns Condition 0.01 0.01 - - 0.03 ns Group x Condition 0.01 0.01 7.23 0.19 0.05 ns RSCLE Group 0.11 . 0.11 2.92 0.08 1.42 ns Condition 0.13 0.13 - - 1.41 ns Group x Condition 0.14 0.14 3.54 0.09 1.53 ns RSCLE Mean Group 0.08 0.08 0.93 0.02 3.16 ns Condition 0.06 0.06 - - 0.57 ns Group x Condition 0.04 0.04 3.66 0.10 0.38 ns Amplitudel Group 0.37 0.37 115.91 3.05 0.12 ns Condition 5.48 5.48 - - 2.44 ns Group x Condition 0.41 0.41 85.36 2.25 0.18 ns Amplitude3 Group 0.05 0.05 64.23 1.69 0.03 ns Condition 1.80 1.80 - - 2.32 ns Group x Condition 2.94 2.94 29.40 0.77 3.80 ns Latencyl Group 0.21 0.21 102.45 2.70 0.08 ns Condition 5.95 5.95 - - 3.16 ns Group x Condition 1.62 1.62 71.41 1.88 0.86 ns Latency3 Group 1.21 1.21 51.22 1.35 0.89 ns Condition 0.29 0.29 - - 0.29 ns Group x Condition 1.33 1.33 38.54 1.01 1.32 ns Risetimel Group 1.14 1.14 119.01 3.13 0.36 ns Condition 2.49 2.49 - - 1.23 ns Group x Condition 0.14 0.14 77.05 2.03 0.07 ns Risetime3 Group 3.41 3.41 73.03 1.92 1.78 ns Condition 3.60 3.60 - - 1.68 ns Group x Condition • 0.22 0.22 81.48 2.14 0.10 ns Resp-pre Group 0.42 0.42 304.12 8.00 0.05 ns Condition 5.34 5.34 - - 1.02 ns Group x Condition 0.93 0.93 198.78 5.23 0.18 ns Resp-during Group 33.37 33.37 645.19 16.98 2.14 ns Condition 16.45 16.45 - - 1.86 ns Group x Condition 16.25 16.25 335.49 8.83 1.84 ns 191 Error Variable SS MS SS MS F jo Respiration Mean Group 7.25 7.25 257.13 6.77 1.07 ns Condition 10.14 10.14 - - 2.70 ns Group x Condition 2.36 2.36 142.55 3.75 0.63 ns RHR-pre Group 0.03 0.03 1.79 0.05 0.53 ns Condition 0.76 0.76 - - 17.45 <.001 Group x Condition 0.00 0.00 1.65 0.04 0.05 ns RHR-during Group 0.04 0.04 0.97 0.03 1.58 ns Condition 2.60 2.60 - - 63.19 <.001 Group x Condition 0.10 0.10 1.56 0.04 2.35 ns RHR Mean Group 0.01 0.01 0.56 0.02 0.03 ns Condition 1.54 1.54 - - 43.73 <.001 Group x Condition 0.03 0.03 1.34 0.04 2.30 ns 192 Appendix 23 Summary of Group x Condition x Time of Measurement Analyses of Variance Error Variable SS MS SS MS F P RSCL Group 0.23 0.23 2.79 0.07 3.16 ns Condition 0.16 0.16 - - 0.57 ns Group x Condition 0.11 0.11 10.99 0.29 0.38 ns Time 3.09 1.54 - - 21.87 ns Group x Time 0.01 0.01 5.36 0.07 0.03 ns Condition x Time 0.04 0.02 - - 0.48 ns Group x Cond. x Time 0.05 0.03 3.41 0.05 0.58 ns Amplitude Group 0.35 0.35 147.64 3.89 0.09 ns Condition 0.50 0.50 - - 0.24 ns Group x Condition 0.58 0.58 78.68 2.07 0.28 ns Time 19.82 19.82 - - 23.17 ns Group x Time 0.07 0.07 32.50 0.86 0.08 ns Condition x Time 6.78 6.78 - - 7.14 <.02 Group x Cond. x Time 2.76 2.76 36.08 0.95 2.91 ns Latency Group 0.21 0.21 79.14 2.08 0.10 ns Condition 4.43 4.43 - - 2.48 ns Group x Condition 0.01 0.01 67.88 1.79 0.00 ns Time 40.48 40.48 - - 20.64 <.001 Group x Time 1.21 1.21 74.53 1.96 0.62 ns Condition x Time 1.81 1.81 - - 1.63 ns Group x Cond. x Time 2.94 2.94 42.06 1.11 2.66 ns Risetime Group 4.25 4.25 106.47 2.80 1.52 ns Condition 6.03 6.03 - - 2.17 ns Group x Condition 0.01 0.01 .84.56 2.23 0.00 ns Time 27.66 27.66 - - 12.28 <.01 Group x Time 0.30 0.30 85.57 2.25 0.13 ns Condition x Time 0.05 0.05 - - 0.03 ns Group x Cond. x Time 0.36 0.36 73.97 1.95 0.18 ns Respiration Group 14.49 14.49 514.27 13.53 1.07 ns Condition 20.27 20.27 - - 2.70 ns Group x Condition 4.71 4.71 285.10 7.50 0.63 ns Time 1264.11 1264.11 - - 110.42 <.001 Group x Time 22.30 22.30 435.04 11.45 1.95 ns Condition x Time 1.52 1.52 - - 0.23 ns Group x Cond. x Time 12.47 12.47 249.17 6.56 1.90 ns RHR Group 0.00 0.00 1.12 0.03 0.03 ns Condition 3.08 3.08 - - 43.73 <.001 Group x Condition 0.06 0.06 2.68 0.07 0.91 ns 193 Variable Time Group x Time Condition x Time Group x Cond. x Time Error SS MS SS MS J? p 3.04 3.04 - - 70.29 <.001 0.06 0.06 1.65 0.04 1.49 ns 0.28 0.28 - - 19.51 <.001 0.04 0.04 0.54 0.01 2.45 ns 194 Appendix 24 Summary of S i g n i f i c a n t Univariate Analyses of Variance for Most Anxious and Least Anxious P i a n i s t s Error Variable SS MS .error SS MS F P. PQ Group 1282.56 1282.56 4408.01 244.89 5.24 <.04 Condition 0.01 0.01 - - 0.00 ns Group x Condition 182.76 182.76 298.61 16.59 11.02 < .01 State Group 2030.63 2030.63 1823.85 101.23 20.04 <.001 Condition 429.03 429.03 - - 14.32 < .01 Group x Condition 87.03 87.03 539.45 29.97 2.90 ns Level Group 11.03 11.03 21.45 1.19 9.25 < .01 Condition 0.03 0.03 - - 0.03 ns Group x Condition 0.23 0.23 14.25 0.79 0.28 ns Strength Group 600.63 600.63 835.65 46.43 12.94 < .01 Condition 24.03 24.03 - - 1.81 ns Group x Condition 65.03 65.03 239.45 13.31 4.89 < .05 Mean Level Group 11.74 11.74 5.98. 0.33 35.31 <.001 Condition 0.03 0.03 - — 0.18 ns Group x Condition 0.07 0.07 2.52 0.14 0.50 ns Mean Strength Group 1127.14 1127.14 499.67 27.60 40.60 <.001 Condition 12.47 12.47 - - 3.20 ns Group x Condition 20.07 20.07 70.18 3.90 5.15 < .04 RSUDS Group 46991.00 46991.00 279794.45 15544.14 3.02 ns Condition 16362.00 16362.00 - - 7.92 < .02 Group x Condition 3222.03 3222.03 37171.45 2065.08 1.56 ns SUDS Total Group 75777.00 75777.00 387230.45 21512.80 3.52 ns Condition 28037.00 28037.00 - - 12.45 < .01 Group x Condition 6477.03 6477.03 40551.45 2252.86 2.88 ns SS Group 70644.00 70644.00 85252.25 4736.24 14.92 < .01 Condition 2265.02 2265.02 - - 1.46 ns Group x Condition 46.23 46.23 27942.25 1552.35 0.03 ns 195 Error Variable SS MS SS MS F _p SCL1 Group 0.63 0.63 1.53 0.09 7.38 <.02 Condition 0.02 0.02 - - 0.17 ns Group x Condition 0.02 0.02 2.17 0.12 0.15 ns Amplitudel Group 0.10 0.10 17.61 0.98 0.10 ns Condition 2.78 2.78 - - 4.51 <.05 Group x Condition 1.47 1.47 11.12 0.62 2.38 ns RHR-pre Group 0.06 0.06 0.86 0.05 1.26 ns Condition 0.35 0.35 - - 5.86 <.03 Group x Condition 0.01 0.01 1.08 0.06 0.08 ns RHR-during Group 0.03 0.03 0.53 0.03 1.11 ns Condition 1.35 1.35 - - 30.13 <.001 Group x Condition 0.07 0.07 0.81 0.05 1.48 ns 

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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0095688/manifest

Comment

Related Items