.143 v 4.7. tv 1.... v And to v v v V? tr T1“!!! 0 7 r 71.7 2.. w 3 . 2 VNWCn1§h§¢m¢MN .. 34g... 95 . v 9.”nt a. 7n. quay; H”? . 141%!!vaqu - t v f 7 at .1»... nor... I 25.? a 1 ' I flirocv .4 .uhmnfiwm K... .. w.-.ih¥¥4 a. 4.. 3.”: Fa... .3. ‘ .59 O 1‘ INS-6W fl! . 43...: .. .oflyflhfi 1 .3313. 3 . .. ‘1. . 2.0 x . v 49.9%“... 3.2.; 3 (NH... . . ? flawegf t. . 3m _ w cc t .49 3. 4% “W In... _ ... n » 0v . ' v 1 av . .3? . v s . . ...z§hn.mu§.h-.§. , mu. . $.32. h .. _ ..... . i F O I r . r ‘4'»?! $I!wv¢¥uflt 31...“.fifiv-v 3.... Q :4 > ._ .. . x . . x . . . , .. . 35.59.: g 3 «ml. it!» it. . a . s . . .nwi.:u.h.$m§ “mm... a... J. a. «mu . . , .. . ,iammmwbuii. . tin .. 38a»... 31% , , .. 1% . . 3.... 3.3:: a $3131.... £313. 4 . .. . , . v 9 4n _ Q . w”): , \)A4vfi.(iuuax\1u 75bit , . . . $1.3... Sauna“ 35.3% . 7n . a - . . . A ”((551 . Illéwk 5.. u guihwvfls 1 y .0... . . S S}...,‘ E! ixififlqus 3823.3 . . . . . . .vzfix is has? LRFKZ... . (inn ‘ ; 0‘5 1; 3.6.5514} 2.)) zgkfiécnlvui‘ \ 5:35:54“? huflxi§24~551>i ’bnt Okfi , \. V A .1. .1...- “142(Ru‘gafihflsang V . s ., 3......$:.fl....mfivsdtz1. .éf.i : , g . . _ . i. . . , giiflniiftléuh. fiétztIEiitx..:e§. . .33 . 1.3.2.1.... .is. .2125... is 1 E4 a t . ; . 14:. 3.1.1.0 “uni av... tying; 1&1 - 93.4.14; ... 0.1g? 5 .. 3).... . .. 25.... ~05 1%.... >21...» ,. 4 w... t; ~ V s 740.99: vrv .1 “14.5% goofififlgrcvr $99.50. 7953591410: [It . gégtflgsggg‘xihnxftiga 2 l... fih?) . «L634 a , L, . .. «mm 9% rv 12"...” 0“?! 3.4849,. , t . r... Is“. .. . . ”(Six-u. 2143;.“v4hfgév3ftiabctifia .. 3.2.9.). 1.2) i 52(5) . 1.04”. 441.1547. 1 1a 4149' 1%. cryiiltlrfifitr . . .u \ . . E41435: g5; §§15§gsig 11.5.1... £1.94“. .Eé: 0.0!; g7? 21! 3:4... tilts: 0." a. . L E IBM-23L 4411243.. 33.. \(‘zfifiw’fi d! V. 13.10:- . .- V V . » 1%....u «#835. F.-. .. .34.. .3 t ritzbu! .u. L31... . . . . .zzflvqénfi 323;. 2.: i .2 . . Wm. VB. . . 5.44.51)... V , , . , y “P .g t u 3 '44.. a; II. .4917: v0 .7 31.114124. %\ )1}. l. 1 .§Z¢i ) 5.. 1):}. gram“ 4!. .. it»! vi. .Nrrchvthtr1vbielvl..1zfrz . .0: .412 \t )2. zs\%3~4§~§lg .2. 13245. In“ L H , . . fightehvivgfiéx , lithirixt .. hu.‘ 5.; 424%..iinwiga I... 7) unfit. .. .5872. At, , :li‘fiirrt‘." )fls‘i gi’ii . i in. zhi‘zl , 1h314 ). ,i‘q. a!!! Diwali). . i . . l ‘3‘!“ h- 15‘131 1.0.1. (In!!!) - 90.1.9044}? 0‘ .4324. OVA ' a $441,424“: .4444! I. it.) I. a 44...!) 173" . a)”; o ti ‘ . . 5" A n y . . fry ta... 4%. 3 . >12 > iiik‘iagl. 1§2$s§3 .§23.9+Z~3¢3:A?.|I| .L llL/ll/lllllzllfllflllllljjlilllll“IMMMJJIBHI V y Michigan Sm University? ' This is to certify that the thesis entitled THE EFFECTS OF UNCONTROLLABLE TASK OUTCOMES ON ELECTROMYOGRAPHICALLY MEASURED TRANSIENT FACIAL EXPRESSION presented by Roger Landvoy has been accepted towards fulfillment of the requirements for Ph.D. degree in Department of Coun— seling and Personnel Services Major professor Date February 19, 1979 0-7 639 OVERDUE FINES ARE 25¢ PER DAY PER ITEM Return to book drop to remove this checkout from your record. THE EFFECTS OF UNCONTROLLABLE TASK OUTCOMES ON ELECTROMYOGRAPHICALLY MEASURED TRANSIENT FACIAL EXPRESSION BY Roger Landvoy AN ABSTRACT OF A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Counseling and Personnel Services 1979 ABSTRACT THE EFFECTS OF UNCONTROLLABLE TASK OUTCOMES ON ELECTROMYOGRAPHICALLY MEASURED TRANSIENT FACIAL EXPRESSION BY Roger Landvoy The purpose of this study was to explore for the occurrence of transient changes of facial expression during learned helplessness conditions. To test for the effects of learned helplessness on facial expression, two groups of subjects were treated with a series of un- controllable or controllable stimuli. One group of sub- jects could not escape a repeated presentation of a loud tone (no-escape/learned helpless group); the other (escape group) could. After treatment with the tone, both groups were given the task of trying to solve 20 single-solution anagrams. During the anagram-solving task, subjects were monitored for facial behavior. Hypotheses were tested regarding the effect of treatment with escapable and inescapable tones upon facial expression emitted during the subsequent anagrams task. The dependent variables were based on electro- myogram recordings from four surface facial sites: the Roger Landvoy frontalis, mentalis, masseter, and depressor facial muscles. The measures derived from the electromyograms were: (a) number of changes of facial expression (facial lability); (b) number of repeated facial changes; (c) num- ber of categories of expressions exhibited, and (d) average duration of an expression. The data were analyzed in two trial blocks of 10 anagram trials each. The null hypothe- ses were tested using univariate analysis of variance with a 5% rejection region chosen for all univariate tests. Analyses revealed that the inescapable group exhibited a significantly greater number of changes of facial expres- sion and repeated facial changes during the first 10 ana— gram trials. No significant differences were found during the first trial block for number of categories of patterns displayed nor for average duration of expression. During the second trial block, no significant differences were found between the treatment groups on any of the facial measures. There were no significant main effects for trial block and only one significant trial block—treatment interaction effect; that was for repeated facial changes. Additional univariate analyses of variance were performed to determine if treatment had an effect on sub- sequent anagram task-solving behaviors. No significant differences were found between the treatment groups on the four anagram—solving variables. They were: mean Roger Landvoy response time for correct solution, number of correct solutions, number of incorrect solutions, and number of [Egg gg trials. Based on the results of the study, three general conclusions were made: (1) EMG measures of facial changes were found capable of discriminating between the facial activity of the treatment groups, suggesting that facial EMG may be a useful tool in the quantifying of compacted affect behavior; (2) the facial change findings tend to support the predictions based on compacted affect theory, suggesting that levels of compacting are related to pre- vious painful experiences; and (3) the lack of behavioral differences found on the anagram-solving measures calls into question the helplessness induction procedures used in this study and more importantly the assumptions under- lying the helplessness-inducing theory itself. ACKNOWLEDGMENTS To all those who have helped me in this project, thank you. A special thanks goes to my committee members, Bill Hinds, Larry Van Egeran, Sam Plyler, and Doug Miller. The chance they gave me to share in their ideas and to refine my own was the most valuable experience this, or any study, affords. To Lawrence Sniderman and Jim Abelson for their valued assistance during the running of this project. To Robin Hughes whose editorial assistance made an important contribution to the final product. To the following people who helped me in many different ways along the way: Jeanna Dressel, John Jerome, Pam Bakeman, Cecil Williams, Marty North, Al Hammer, Dave Inman, Pam Ninni, and to those volunteers who participated in the study. And finally to Karen, I owe the biggest thanks. ii TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . LIST OF FIGURES . . . . . . . . . . . CHAPTER I. THE PROBLEM . . . . . . . . . . Introduction . . . . . . . . . Need . . . . . . . . . . . Purpose. . . . . . . . . . . Depression, Compacted Affect, and Helplessness. . . . . . . . Hypotheses. . . . . . . . . . Definition of Terms. . . . . . . Summary. . . . . . . . . . . Overview . . . . . . . . . . II. REVIEW OF THE LITERATURE. . . . . . Introduction . . . . . . . . . Theory . . . . . . . . . . Compacted Affect Theory: An Historical Overview . . . . . . . . . Compacted Affect: Current Theory . Compacted Affect: Research Support. Learned Helplessness——Depression and Emotion . . . . . . . . . III. DESIGN OF THE STUDY . . . . . . . Overview . . . . . . . . . . Sample . . . . . . . . . . . Apparatus . . . . . . . . . . Procedure . . . . . . . . . Orientation. . . . . . . . . The Tone Task . . . . . . . . iii vi 10 ll 12 13 l4 l4 l4 14 28 37 43 51 51 52 53 56 56 57 CHAPTER The Anagrams Task . . . . . . . . . 61 Debriefing . . . . . . . . . . . 62 Stages of Electromyogram Data Gathering . . 64 Design . . . . . . . . . . . . . 66 Hypotheses . . . . . . . . 70 Test of Learned Helplessness: Treatment Effects . . . . . . . . . . . . 72 Summary . . . . . . . . . . . . . 74 IV. ANALYSIS OF THE DATA . . . . . . . . . 75 Hypothesis Testing: Effects of Treatment on Facial Expression . . . . . . . . 75 Hypotheses . . . . . . . . 76 Test of Learned Helplessness . . . . . . 85 Summary . . . . . . . . . . . . . 89 V. SUMMARY, CONCLUSIONS, AND DISCUSSION . . . . 92 Summary . . . . . . . . . . . . . 92 Effects of Treatment on Facial Expression . 96 Effects of Treatment on Anagram—solving Behaviors . . . . . . . . . . . 98 Conclusions and Discussion. . . . . . . 98 The Facial Behaviors . . . . . . . . 99 Anagram-solving Behavior. . . . . . . 103 Limitations . . . . . . . . . . . 110 Implications. . . . . . . . . . . 112 APPENDICES APPENDIX A. SUBJECT SIGN-UP SHEET . . . . . . . . . 116 B. DIAGRAM OF BUTTON BOX . . . . . . . . . 117 C. CORRECT SEQUENCE OF BUTTON PRESSES FOR TONE TASK . . . . . . . . . . . . 118 D. ANAGRAMS . . . . . . . . . . . . . 119 E. DEPARTMENTAL RESEARCH CONSENT FORM. . . . . 120 REFERENCES . . . . . . . . . . . . . . . 121 iv TABLE LIST OF TABLES Experimental Group Means of the Four Facial Measures During Both Anagram Trial Blocks . . . . . . . . . . Summary of Facial Lability Analysis of Variance . . . . . . . . . Summary of Repeated Facial Change Analysis of Variance . . . . . . . . . . Summary of Patterns Displayed Analysis of Variance . . . . . . . . . . . Summary of Mean Duration of Pattern Analysis of Variance . . . . . Experimental Group Means: Anagrams-Solving Behaviors . . . . . . . . . . . 76 78 81 83 86 87 LIST OF FIGURES FIGURE 2.1 Graphical representation of a theory of innate activators of affect (Tomkins, 1962, p. 251) . . . . . . . . 2.2 Summary of features common to learned help— lessness and depression (adapted from Seligman, 1975, p. 106). . . . . 3.1 Placement of surface electrodes. Schematic drawing of the face showing the locations of the frontalis, mentalis, masseter, and depressor muscles, and the locations of the surface recording electrodes. . . . 3.2 Time chart for one trial of the auditory task . . . . . . . . . . . . . 3.3 Anagrams task time chart (one trial) . . . . 3.4 Stages of electromyogram data gathering (Flow chart depicts stages of data gathering for one anagram task trial) . . 3.5 Muscle pattern numbers (Muscle sites ranked in order of decreasing activity). . . . 3.6 Design: Effects of treatment on facial behavior. . . . . . . . . . . . 3.7 Learned helplessness test. . . . . 4.1 Facial lability group means . . . . . . . 4.2 Repeated facial changes: Group means. . . . 4.3 Number of facial patterns displayed: Group means. . . . . . . . . . . 4.4 Facial pattern duration: Group means. . . vi 3O 44 55 60 63 65 67 69 73 78 81 83 86 CHAPTER I THE PROBLEM Introduction Since S. Freud's (1894) early definition of psy- chological "defense" as the struggle against painful or unendurable ideas and affects, psychotherapeutic concern for client affects and their related defenses has been constant. The repeated use of distorted emotional behavior is variously seen as a symptom, cause, or attri- bute of psychological maladjustment. Anna Freud (1937) noted that for defenses to occur, the affects "must sub- mit to all the various measures to which the ego resorts in its efforts to master them, i.e., they must undergo metamorphosis" (p. 34). Tomkins (1962) uses the term "affect transfor- mation" to describe this metamorphosis. Tomkins (1962, 1963) and others (Kell & Mueller, 1966; Jacobson, 1967; Haggard & Isaacs, 1966; Hinds, 1976) have noted transfor- mations on the faces of clients which miniaturize and combine affect to form a part of the clients' defenses. The nature of the transformation is the compressing and combining of single affects over very short (two seconds or less) time spans. Kell and Mueller (1966) termed this compressing and combining of affect "compacted affect." The dynamics of compacted affect relate to the temporal reduction of the experience of one affect by the expres- sion of another. The purpose of this transformation is the termination or avoidance of unwanted affect. Kell and Mueller (1966) believe that compacted affect helps to create and maintain psychOpathology. They go on to sug- gest that diagnosis, elaboration, and understanding of these compacted affect patterns by clients is beneficial and perhaps crucial to a client's behavior change process. They state, "The client has compressed his affect for good reason and one of those is that in its present form, the affect although anxiety provoking, is not as painful as it would be if it were expanded" (Kell & Mueller, 1966, p. 62). They concentrate on the compacted "theme" (the recurring tone, attitude, or View one has of himself that has come to symbolize the affect content) in their efforts to understand client behavior. In addition they believe exploration into the theme allows reliving and expansion of the initial compacted affect, resulting in a decom- pression of affect behavior. The importance of this exploration is clearly stated by Kell and Mueller: Unless the encysting threads of the compacted feel- ings are unwoven, thus facilitating the expansion of experiences, the counseling relationship will terminate with both client and counselor feeling dissatisfied as a result of having undergone a sig— nificant interpersonal failure. (Kell & Mueller, 1966, p. 62) . . .. ._..-. _._...——-_. ..-_ . . A view similar to Kell and Mueller's is expressed by Haggard and Isaacs (1966). After reviewing slow motion films taken during psychotherapy interviews, they conclude that clients seem to emit more micromomentary facial expressions when they are dealing with active conflict. They found that micromomentary expressions tend to be associated with denial statements and verbal blocking, and incongruent with concurrent verbal content and adja- cent facial expressions. They also found that micromomen- tary expressions usually go unobserved when they occur and when film is played at normal speed. However, they find that these expressions can be observed when film is played at about one-sixth normal speed. Haggard and Isaacs believe that these compacted expressions may be a form of "temporal censorship" that acts as a safety valve to permit the very brief expression of unacceptable affects. They state: From this point of View the micromomentary facial expressions have a double benefit to the indi- vidual: he may indulge in some impulse expression with minimal risk of retaliation or rejection by the observer, and without arousing the anxiety that would ensue if he himself were aware of, or of having expressed the unacceptable impulse or affect. (Haggard & Isaacs, 1966, p. 165) Haggard and Isaacs (1966) conclude that relatively little clinical or research attention has been given to micromomentary facial expressions. They speculate that the reason for the inattention may be due to a "per- ceptual defense" on the part of psychologists. They ask: "Have we as adult observers learned BEE to see some aspects of what occurs before our eyes . . . [for] if one were to observe them they would activate impulses, conflicts, or anxieties which one had succeeded in for- getting" (pp. 164—165). A similar View is expressed by Loeb (1966). He believes that therapists cannot avoid the impact of compacted affects by ignoring them. He states: "The psychiatrist might behaviorally respond to the patient's compacted anger without the idea that the patient is angry ever coming into his consciousness" (p. 617). Loeb states that these rapid changes of affect are one means by which, as Freud stated, "the Unconscious of one human being can react upon that of another without passing through consciousness" (see Loeb, 1966, p. 614). In summary, these clinicians have noted (a) that compacted facial affect may play a major role in the maintenance of psychopathology, (b) that psychotherapy clients may need to expand their compacted affects before other behavior changes can take place, and (c) that therapists may be unconsciously reacting to their clients' expressions of compacted affect and thus unknowingly impacting the therapeutic process (Kell & Mueller, 1966; Haggard & Isaacs, 1966; Loeb, 1966; Hinds, 1970). Need Although psychological clinicians have suggested that compacted affect serves a defensive function that helps maintain maladaptive behavior, little research has been conducted to determine under what conditions and to what degree this affect transformation takes place. As previously mentioned, in their study of facial expres— sion, Haggard and Isaacs (1966) found that rapid changes of expression occur when psychotherapy clients are in "active conflict." They recommend that persons in situ— ations other than psychotherapy should be studied to determine the extent to which rapid changes of facial expression occur. To date no one has undertaken a study to analyze compacted facial affect in persons in other active conflict situations. The present study seeks to determine the extent to which compacted facial expressions occur in a nontherapy conflict situation. Hinds (1976) and Inman (1976) have identified another need relevant to the study of compacted facial affect. They note a lack of instruments capable of measuring patterns of compacted expressions of affect. This is not a new problem. Investigators for over 100 years have had to contend with ways to measure rapid changes of facial expression. Darwin, in his book The Expression of the Emotions in Man and Animals (1872), states: The study of expression is difficult, owing to the movements being often extremely slight, and of a fleeting nature. A difference may be clearly per- ceived and yet it may be impossible, at least I have found it so, to state in what the difference consists. (p. 12) Since Darwin's time, few studies have been conducted to determine the psychological meaningfulness of transient expressive behavior, and virtually no instruments have been developed to help in these efforts. One measurement approach that shows promise is facial electromyography (EMG). Electromyograms have been found capable of providing subtle, quick, and con- tinuous observations of facial activity. In his book, The Biology of Emotion, Jacobson (1967) makes a strong statement supporting the use of electromyographical recordings for the measurement of subtle affect. He writes: According to the ultrasensitive measurements during mental activities [including emotional ones] which have continued almost daily in my laboratory these past thirty-five years, a specific neuromuscular (EMG) pattern marks the character of each and every moment of their occurrence. . . . I here go further, suggesting that the nature of mental activities can be accounted for and depends upon specific recordable and repeatable neuromuscular patterns. (Jacobson, 1967, pp. 118-119) Schwartz et a1. (1973), in their studies of facial expres— sion of affect, found that facial electromyography could accurately discriminate between depressed and nondepressed patients. In another study (Schwartz et al., 1974), they found discriminations could be made between happy, sad, and angry thoughts generated by subjects. They concluded that facial electromyography can provide a useful objective index of subtle emotional states. But, they note, their "preliminary but strong support" for the hypothesis that electromyography can act as a qualitative and quantitative index of facial expression needs further study and clarifi- cation. Thus, electromyograms show promise in the study of subtle facial behavior. What is needed now, according to Hinds (1976), is the further development of procedures that may lead to a systematic research approach to the analysis of compacted facial affect. PUI‘EOSE Compacted and sequenced facial affect has been noticed and often described but rarely studied in the last 100 years. Two important areas have been identified as needing further attention. They are the need to study the expression of compacted affects in persons other than psychotherapy clients and the need to develop instruments capable of accurately measuring compacted affect behavior. The purpose of this study is to test for the occurrence of compacted facial affect after treatment with uncontrollable outcomes. The main thrust in designing the study was aimed at refining strategies, measures, and procedures in the study of compacted facial affect. In particular, these efforts were directed towards adapt- ing the use of electromyograms to the study of transient facial expression. The main question this study will address is whether expressions of compacted affect are associated with maladaptive behavior patterns. Depression has been identified as a major maladaptive behavior pattern maintained by compacted affect (Tomkins, 1962, 1963). However, this View is not universally held. Seligman (1975) has suggested depression is maintained by the reduced expression of all affect. These contrast- ing views will be briefly reviewed in the following section. Depression, Compacted Affect, and Helplessness Tomkins (1962) believes depression to be a sequenc- ing of positive and negative affect, and that this type of sequencing creates and maintains this and other forms of psychOpathology. He claims that depression is a con- tinual sequencing of affect consisting of "an oscillation between increase and decrease of positive affect which alternately activates distress or anger and shame" (Tomkins, 1962, p. 290). In his View, depression can become chronic and, eventually, a stable character trait. This occurs when the oscillating sequences become com- pacted, unconscious, unassociated with the activating event, and used to defend against new experiences. (An extensive review of Tomkins' theory will be contained in Chapter II of this report.) Another View of depression is set forth by Selig- man (1975) as the learned helplessness theory of depres- sion. He calls depression the common cold of psychopathology: it is something that touches all of us, yet is the least understood and most inadequately investi- gated of all the psychopathologies. He believes that the cause of depression is the belief that action is futile. Seligman (1975) states, "The depressed patient believes or has learned that he cannot control those elements of his life that relieve suffering, bring gratification, or provide nuture--in short he believes he is helpless" (p. 93). The primary characteristic of learned helpless- ness-depression is believed to be the generalization of a behavior pattern of responselessness that was previously associated with an uncontrollable situation to a current, controllable situation. Laboratory and clinical evidence finds the following traits associated with learned help— lessness: the belief in an external locus of control, a lack of motivation to respond, and undermined ability to perceive success, and an initial period of heightened emotionality followed by responselessness (Hiroto, 1974; Seligman, 1975). In this View, "depression" (the human correlate of Seligman's 1975 animals' responses to con— ditions of learned helplessness) is equated with an initial increase in emotion, followed by a general lack of responding, or, an emotional deficit. In Seligman's (1975) words, it is "not doing anything at all." 10 When a traumatic event occurs, it causes a heightened state of emotionality that can loosely be called fear. This state continues until one of two things happens: if the subject learns that he can control the trauma, fear is reduced and may disappear altogether; or if the subject learns he cannot control the trauma, fear will decrease and be replaced with depression. (Seligman, 1975, pp. 53-54) Seligman's View of the etiology of depression is consistent with Tomkins' View up to the point where Selig- man claims that "heightened emotionality" gives way to "not doing anything at all." Tomkins' theory would account for the reduction of overt affect activity during depression that Seligman describes by hypothesizing that the final stage of a transformation process compacts and combines facial affect to form an emotional "heiroglyphic." This "heiroglyphic" is generally unobservable to both respondent and observer and thus capable of being mistaken for "not doing anything at all." The purpose of this study is to put to an experi— mental test these opposing views of emotional activity during depression. This test will be guided by the following hypotheses. Hypotheses Four hypotheses will be used to guide this explor— ation. The hypotheses reflect the three major character- istics of compacted affect. They are (a) that people shorten the duration of their affect and (b) repeat sequences of affect to (c) avoid the experiencing of new and painful, but potentially growth-producing, emotion. 11 The hypotheses are: :1} Subjects treated with uncontrollable outcomes will exhibit significantly more changes of facial expres- sion than will subjects treated with controllable outcomes. {12} Subjects treated with uncontrollable outcomes will exhibit significantly more repeated changes of facial expression than will subjects treated with control- lable outcomes. H_3_‘ Subjects treated with uncontrollable outcomes will exhibit significantly fewer types of facial expres- sion than will subjects treated with controllable outcomes. 52‘ Subjects treated with uncontrollable outcomes will exhibit facial expressions of significantly shorter duration than will subjects treated with controllable outcomes. Definition of Terms In this study a few special terms are used. These terms and their definitions are: Affect.—-A complex concept with physiological, neuromuscular, and phenomenological aspects. Affect forms a subsystem of personality having motivational and homeostatic qualities which help give direction and 12 meaning to life. In this study, the terms "affect" and "emotion" will be considered synonomous. Facial affect.-—A major component of all affect experience. It includes innate and learned patterns of electrochemical, muscular, and felt facial activity. Affect transformations.—-Changes in the innate pattern of activation or expression of affect, usually achieved through learning. Compacted affect.-—Denotes a transformation con— sisting of at least two affects shortened in duration, combined one after another in a stereotyped sequence, and used repeatedly in response to an activating event. Summary The study was designed to explore the occurrence of compacted facial affect during learned helplessness conditions. Tomkins (1962) as well as Kell and Mueller (1966), Haggard and Isaacs (1966), and Hinds (1976) sug- gest that people use compacted sequences of affect to maintain stability when threatened with the experience of painful affect. To ward off the pain they call upon sequences of affect that, through repetition, have become so miniaturized and unconscious that they generally go undetected by both observer and exhibitor. The important question this research study will address is whether the "emotional responselessness" (that 13 Seligman hypothesized accompanies learned helplessness) may consist of the repeated use of miniaturized sequences of affect that Tomkins believes helps to create and main- tain depressive psychopathology. Overview Chapter II contains a review of related literature supporting the View that compacted affect plays a role in the maintenance of maladaptive behavior. Also to be found in Chapter II will be a review of the literature pertinent to the expression of emotion during learned helplessness. In Chapter III, the design, procedures, measures, and methods of analysis that are employed in the study are described. Chapter IV contains reports of the results of data analysis, and Chapter V will contain a discussion of the results and a summary. CHAPTER II REVIEW OF THE LITERATURE Introduction The temporal dimension of emotional expression received theoretical attention during the nineteenth century. In the first section of this chapter, the works of Charles Darwin (1872), William James (1884, 1890), Sigmund Freud (1893, 1894, 1895a, 1895b, 1896a, 1896b, & 1899), and G. B. Duchenne (1876) will be reviewed. They provided the ground work for the current theoretical formulations reviewed in the following section. The final section Of this chapter contains a review of learned helplessness theory and research linking learned helpless- ness to emotion and depression. Theory Compacted Affect Theory: An Historical Overview The basis for recent facial affect theory was formed in the nineteenth century. Theorists such as Darwin (1872), Duchenne (1876), Bell (1806), Spencer (1855), Pediret (1867), and James (1890) gave attention to the fact as a source, expresser, and transformer of 14 15 affect. Darwin (1872) described affects ranging from joy to despair. He and his associates, through collective observation, catalogued 30 primarily facial descriptions of affect. Many of the observational methods Darwin developed for the study of emotion are still in use today. He believed that facial displays of affect had survival value in that they communicate the internal state of the organism to other members of the species. Also, he was one of the first to emphasize the importance of the tem- poral association of the emotions. He likened emotional expression to musical expression in that each individual emotion, like tones in a melody, is dependent upon the surrounding emotion for its expressive effect. But Darwin chose not to study the temporal dimension of emotion, noting that the task would be extremely diffi— cult "owing to the movements being often extremely slight and of a fleeting nature" (Darwin, 1872, p. 12). William James (1890) was another investigator who believed facial and other patterns of muscular movements played an important role in emotional experience. James paid particular attention to the subtle feelings of change which accompany emotion. James states, The changes are so indefinitely numerous and subtle that the entire organism may be called a sounding— bgard, which every change of consciousness, however slight, may make reverberate. The various permu- tations and combinations of which these organic activities are susceptible make it abstractly pos- sible that no shade of emotion, however slight, should be without a bodily reverberation. . . . 16 The next thing to be noticed is this, that every one of the bodily changes, whatsoever it be, is FELT, acutely or obscurely, the moment it occurs. (James, 1890, pp. 150-151) While Darwin chose to study the character of the indi- vidual expressions of emotion, James believed that the study of transient patterns of expression is essential to understanding emotion. He criticized psychologists for not paying enough attention to the transient affec- tive behaviors. James claimed the affect literature describing individual emotions (he termed them the "golden eggs" of emotion) generated by Darwin was "tedious," and "its pretenses to accuracy a sham," and he believed Darwinian methods of observation could not be used to study the more important transitive affect behaviors: the movement, order, succession, pacing, intensity, and duration of emotion. Another observation made by James, which is important to the study of com— pacted affect, is his notation that emotions were subject to "blunting." By blunting, he referred to the loss or reduction, achieved Via learning, of physiological activity associated with emotion. He saw in blunting a means by which emotional expression could easily be reduced and transformed. Finally, James noted a close link between imagery and the muscular movements of expression. He believed that an image of the original "exciting fact" could cause an "actual revival" of the emotions associated with the imaged fact. James states, 17 "The cause is now only an idea, but this idea produces the same organic irradiations, or almost the same, which were produced by its original, so that the emotion is again a reality" (James, 1890, p. 474). He believed that an emotional temperament and lively imagination were the "necessary and sufficient conditions for an abundant emotional life." His descriptions of types of images (tactile, motile, visual, and auditory), and the effect of imagery upon affect experiences, gives a sense of the diversified imagery process that Tomkins (1962, 1963) and Jacobson (1967) describe as capable of transforming and organizing affects into stable behavior patterns. Three years after the publication of James' Th3 Principles of Psychology (1890), Freud published the first of his many articles on psychopathology (1893). In this work, and in a number of succeeding publications (on hysteria, 1896; the defence neuropsychosis, 1894, 1896; anxiety-neurosis, 1894, 1895; and obsessions and phobias, 1895), Freud presented his early View of the relationship of affect and affect transformation to psychopathology. Freud (1893), writing about hysteria, noted his and Breuer's original contention that recollection, in detail, of the original exciting trauma and its accompanying affect was essential to alleviating and eliminating hysterical symptomatology: "Recollection without affect is nearly always quite ineffective" (p. 28). Freud l8 believed that "hysterical patients suffer principally from reminiscences," and that these reminiscences were maintained by associated affects. He characterized normal and hysterical reminiscences in the following manner. After an accident, for example, the remembrance of the danger and the subsequent (weakened) repro- duction of the terror is accompanied by the memory of the sequel, of the rescue and the consciousness of present security. The memory of an injury to the feelings is corrected by an objective evaluation of the facts, consideration of one's actual worth and the like, and thus the normal man succeeds by means of associations in dissipating the accompanying affect. . . . Our observations have shown that those memories which give rise to hysterical phe- nomena are retained with wonderful freshness and with full affective tone for a long period. (p. 31) In "The Defence Neuro-psychosis" (1894), Freud extends his affect theory along new lines. First, he hypothesizes that in hysteria, persons learn to convert their painful affect-ideational excitations into some other bodily expression, and that when these same affect— ideational excitations are not "converted," there is a predisposition to other pathologies. If the capacity for conversion does not exist in a person predisposed to hysteria and yet the separ- ation of its affect from an unbearable idea is nevertheless undertaken as a defence against the latter, then this affect must persist in the psy- chical sphere. Thus weakened, the idea remains present in consciousness, detached from all associ- ations; but its affect, now freed from it, attaches itself to other ideas which are not in themselves unbearable, but which through this "false connection" grow to be obsessions. This is shortly the psycho- logical theory of obsessions and phobias. (pp. 65-66) Freud notes that the primary characteristic of obsessional affect is that it is "dislocated or 19 transposed." In "The Anxiety Neurosis" (1894), Freud extends further the concept of transposed affect to consider it a major mechanism in the maintenance of phobias. He states that the difference between phobias and obsessions is found in the fact that the transposed emotionality accompanying phobias is always morbid anxiety, while obsessions are maintained by persistent emotional states which can be made up of any of the innate emotions, such as fear, anger, and remorse. Another difference is that phobias do not rely upon a substitution or conversion of the affect-ideational excitation to create the pathology, but rely directly on an associational process that links the transposed anxiety to its original source. In a further important discussion of the role of transposed affect, which is of particular interest to the study of compacted affect, Freud (1896) maintains that pathological affect can be transformed from any one affect to any other discrete affect for the purpose of defensive relief. He gives the following example: The affect of reproach can with the aid of some psychical reinforcement transmute itself into any other unpleasant affect; when this has happened there is nothing to hinder the substituted affect any longer from becoming conscious. Thus self— reproach (for having performed a sexual deed in childhood) can easily transform itself into shame (lest another person should come to hear about it), into hypochondriacal anxiety (lest some bodily injury should result from the action which evoked the self- reproach), into dread of the community (fear of punishment by the world at large for the lapse), 20 into religious anxiety, into delusions of reference (dread of betraying the deed), into dread of temp- tation (justified mistrust in the personal strength of moral resistance), and so on. (p. 165) In The Interpretation of Dreams (1899), Freud explains the role of unconscious processes in the formation of transformed affect and resulting psychopathology. Once ideas and associated affects are transformed, via repres- sion, and experienced only unconsciously, they can be further transformed by conflict generated in the uncon- scious. This occurs when the original pleasurable impulses that were thwarted, and which caused the now—avoided pain, seek new outlets. The result can be a multi-level series of trans- formed ideas and associated affects. This complex system of associations, originally intended to avoid pain, can result in relatively stable pathological states. The theory of the psychoneuroses asserts as an indisputable fact that only sexual wishful impulses from infancy, which have undergone repression (i.e., a transformation of their affect) during the develop- mental period of childhood, are capable of being revived during later developmental periods . . . and are thus able to furnish the motive force for the formation of psychoneurotic symptoms of every kind. (Freud, 1899, p. 645) Freud (1899) claimed that the role of transformed affect in psychoneuroses could best be understood by studying the technique of dream work. He claimed that in the dream work, the "quality" of the affect could not be changed, and that the transposed associational ideas (in latent form) were present together with the original 21 affect. The fact that affects could be repressed, trans- posed, and transformed, yet maintain their original "quality" in the dream work, was not a contradiction to Freud. It was just this seeming contradiction that made the interpretation of the pathological ideational material possible. A brief summary of Freud's theory of affect trans- formation follows. First, a painful impulse is subject to-control by efforts to terminate its expression. The associated ideas and the affect itself are then stored in memory where the ideational material can readily be worked upon by any defense mechanism. Then, once the idea has been repressed or otherwise defended against, the affect itself is subject to substitution or reduction of quantity by the expression of other affects. These "transformations of affect" are the second step in a con- tinuing and complex process aimed at the successful con- trol of the original painful impulse and its associated affect. The transformed associated affects are therapeu- tically important because they must be reunited with the dissociated memories, and clinically they are important because, as Freud states, when working with the psycho- neuroses, "the affects are the constituent which is least influenced and which alone can give us a pointer as to how we should fill in the missing thoughts." Thus Freud, in his early theoretical works, laid down a framework for understanding psychopathology, based (-v~.~ _ 22 upon a foundation of transposed and transformed painful affects. Freud expanded on these ideas throughout the rest of his life, and the psychoanalytic movement in general has been the most prominent theoretical source of ideas about the role of affect in psychopathology. From the analytic point of view the avoidance of painful affect has come to be known as the primary moti- vator for the psychological defense mechanisms employed by those suffering from all forms of psychopathology. Fenichel (1945) writes, "Thus, in the last analysis any defense is a defense against affects. 'I do not want to feel any painful sensation' is the first and final motive of defense" (p. 161). It is surprising that the psychoanalytic movement, which has based much of its theory of psychopathology on affect defenses, and the resulting transformation of affect, and which has been a major source of ongoing speculative theory about affect behavior, has given little attention to the particulars of affect expression. An important nineteenth century contribution to under- standing the particulars of facial affect expression came from G. B. Duchenne. Duchenne (1862, 1876) took meticu- lous care in describing the effect of over 100 pairs of facial muscles on facial affect expression. Duchenne suggested that most pairs of facial muscles were expres— sers of discrete affects. His precise use of electrical ~- ._ .. “u_‘. .1-- 23 stimulation and photographs to produce and record affect expression remains one of the most comprehensive efforts to catalogue discrete affect expression. Duchenne was a French neurologist who said his work was generated by one ambition: "To demonstrate as never before a living anatomy in man." His major tool in this endeavor was the stimu— lation of facial nerves with electrical current and the recording of the resulting muscle and expressive activity. Duchenne's studies are important to the study of compacted facial affect in three respects: (a) his method of study--electrical stimulation and photographic recording of muscle activity--pioneered the use of electro- physiological recording techniques; (b) he charted the nerve and muscle structure of the face for later researchers; and (c) his work demonstrated the subtle and important effect of individual muscle action upon facial expression. Duchenne believed the expressions of the face are produced by contractions of four classes of muscles: completely expressive, incompletely expressive, comple— mentary expressive, and inexpressive muscles. By com- pletely expressive, he meant those muscles which have the exclusive role of representing completely a specific expression by their isolated action. The other three categories are made up of muscles that act either in succession or in combination to form a complete expression. , _., 7,. 24 Duchenne refers to the expressions produced by these four categories of muscles as "primary" expressions. He states: The essential expressions of the face (produced either by the partial contractions of the com- pletely expressive, or by the combination of the incompletely expressive muscles with the comple- mentary expressive muscles) are primary, because in association they can produce a harmonious ensemble and create other expressions, the sig- nificance of which is more extended and which are complex expressions. (Duchenne, 1862, p. 584) What Duchenne meant by a complex expression—-a singular expression created by two or more muscles capable of individual expressiveness acting in associ- ation—-no longer is the common meaning of the term. Complex expression has come to mean the facial activity consisting of two or more expressions expressed simul- taneously on the face (Ekman & Friesen, 1967, 1969; Ekman, Friesen, & Ellsworth, 1972; Nunnenmaa, 1958; Plutchik, 1962; Tomkins & McCarter, 1964). This "given instant" interpretation of complex expression has become so prevalent that most studies of transient facial behavior are efforts to determine the variety of dis- crete affects expressed over time, rather than the nature of the rapid sequencing of expressions of discrete affects (Izard, 1971; Schwartz, 1973, 1974). Thus, researchers have tended to neglect one of Duchenne's main assertions: that emotional expression is primarily a process of associated singular expressions. 25 In summary, Darwin, James, and Freud believed that changes in innate affect expression played a role in the maintenance of painful psychological states. In addition, Darwin, James, Duchenne, Spencer, Bell, and Pediret speculated that the expressions of the face had a direct relationship with affect and offered a theoreti- cal framework from which to explore the hypothesized relationship. While these nineteenth century works offered a rich field for further investigation into transient affect expression, changes in the way investi- gators conceptualized human behavior drew attention away from the face and emotion. The discrediting of the then- popular "science" of facial physiognomy and the advent of "behaviorism" were two of many developments that led to the diminution of research in facial affect between the late 18003 and early 19505. One result of this change is that research psy- chologists, for most of this century, have discounted the affects, seeing them only as internal, subjective, and unmeasurable phenomena. Izard (1971) attributes this lack of scientific attention to three things: (a) the tendency of behavioral scientists to View emotion as a global, unitary concept which makes operational defi- nition so difficult as to preclude it as a researchable problem, (b) the dominance of logical positivism and s-r drive reduction principles in psychological theory, and 26 (c) the lack of an adequate theory dealing with separate and distinct affects that can be studied by repeatable and specific operations. But perhaps the major problem has been that when scientists have chosen to study facial affect, they have concentrated their attention on verifying the existence of categories or dimensions of facial expression. The problematic history of this Darwinian type "golden egg" approach has had a depressive effect on affect research. Initially, such research concentrated on labeling cate- gories of naturally occurring or posed expressions. Later other approaches included the study of the expressive value of different portions of the face; how to determine deceptive from natural categories of expression; the phylogenetic and ontogenetic development of categories of facial expression; the best means of recording facial expression; how to train judges to accurately differen- tiate expressions; poser and/or judger variables that effect ratings; the differentiation of learned and innate expressions; and cross-cultural studies to determine the universality of different expressions. The resulting collection of inconclusive and oftentimes contradictory data, based on widely divergent methodologies, left reviewers of the research wondering why so much attention had been given to an area of study that had consistently proven unrewarding. 27 Duffy (1941) in his review of the research sug- gested that because of the problem in measuring emotion, the concept should be abandoned altogether and other con- cepts developed where the meanings would not be so idio- syncratic, and research would prove more feasible. Facial studies conducted between 1900 and 1950 were reviewed by Bruner and Tagiuri (1954), and they concluded that there are no invariable facial patterns of emotional expression. Brown and Farber (1951) note that all previous efforts to classify discrete emotions based upon facial expression had ended in failure. And Klopfer (1975), in assessing the research conducted in the area of facial affect expression in the last 100 years, believes we are still largely where Darwin (1872) left us in his study of affect expression. Klopfer states, "In sum, even the post Dar- winian studies show little improvement in conception or design from their predecessors, nor have they advanced us significantly from the time of Darwin" (p. 406). A number of recent develOpments have caused a change in that trend. Theoretical and empirical support for transient facial affect and the role played by affect transformations in human behavior have provided a broader base for investigative efforts into the behavioral ex- pressive realm of the affects (Tomkins, 1962, 1963; Izard, 1971, 1972, 1975; Ekman, 1972; Ekman & Friesen, 1975, 1976; Ekman, Friesen, & Ellsworth, 1972; Gellhorn, 1964; Jacobson, 1967; Schwartz, 1973, 1974; Plutchik, 1962). 28 Recently, Sylvan Tomkins (1962, 1963), in his two- volume work Affect Imagery and Consciousness, has used and refined the theories of nineteenth century investigators to form a comprehensive theory of affect behavior. Like Darwin, he lists and describes a number of primary and secondary affects and goes on to assert that these affects are innate and that the face is the primary site of their expression. He also posits, as did Duchenne, that the face, with its extensive network of nerve and muscle tissue, is a source as well as an expresser of affect. Tomkins, like Freud and James, believes that affect dynamics and affect transformations have a modifying affect on human behavior. His View that awareness and control of the face is important when trying to change problematic behavior is of clinical and theoretical importance (Kell & Mueller, 1966; Haggard & Isaacs, 1966; Jacobson, 1967; Izard, 1971). Tomkins' work details the effect of the transformations that he terms "miniaturized" and "combined" affect. In this area, his theory forms the theoretical basis for this study. His work will be reviewed in depth in the following section. Compacted Affect: Current Theory This study of compacted facial affect rests pri— marily on Tomkins' (1962, 1963) theory of affect behaviors. In particular, it is based upon his statements about min— iaturization of affect, central control of imaged affect, 29 consciousness of transformed affect, and the results of these transformation processes. Two significant aspects of Tomkins' theory are (a) that he considers the affect system to be a major motivating system and (b) that he defines affect behavior as primarily facial behavior. He lists eight primary affects that can be differentiated facially and that are universal across cultures. They are: joy, interest, surprise, fear, anger, shame, contempt, and distress. Tomkins believes that there is a wide variety of innate activators of these affects but that the common- alities of the activators can be summarized with a few simple principles. He believes affects can be activated by stimu- lation (density of neural firing) that is on the increase, on the decrease or that maintains a steady level. These innate activators of affect respond to and amplify the pattern of stimulation created by the person—environment interaction and thus provide motivation for further action. Figure 2.1 graphically represents the relation— ship of density of neural firing over time to affect activation. The innate system of activation and expression of affect can be changed dramatically by a number of learned transformation that can permanently alter the ongoing experiencing of affect. Tomkins gives extended 30 treatment to numerous transformations which can affect the compacting process. The effects of miniaturization, imagery, and consciousness will be reviewed. Startle Fear Interest / Anger i\<:\\\// //r /// Distress Density of Neural Firing Laughter Joy Time Figure 2.1. Graphical representation of a theory of innate activators of affect (Tomkins, 1962, p. 251). Miniaturization. Miniaturization is the charac- teristic and repetitive shortening of duration and expression of an affect. Tomkins (1962) describes it in the following way: A transformation similar to habituation, yet dis- tinct, is what we have called miniaturization, in which the innate complex organization is further and further compressed almost to the point of invisibility to both the respondent and the observer. Thus the smile may be emitted for a half second rather than a few seconds. At the same time, the mouth does not open and the broad- ening characteristic of the smile may be reduced to a slight widening--no more than a quick tug on 31 the lips. Similarly the affect of shame may be miniaturized into a very slight lowering of the eyes, or even a partial lowering of the lids or a slight relaxation of the neck muscles which sup- port the head in an upright position. . . . There are innumerable other ways in which all, or crucial parts, of innately patterned affective responses are compressed. (pp. 307-308) Imaged affect. Innate affect can be transformed and stored centrally as an image which is subject to recall. Tomkins believes that these remembered exper— iences provide another means for reducing and compacting affect. Centrally stored imaged affect can be experienced upon retrieval with much greater speed than can normal affect. Furthermore, internal control of such retrieval allows tremendous freedom to compose combinations of affect. Such conscious control over affect imagery also makes possible another kind of modification of the innate affects-—that of combinations of affects. Although each affect is innately pat- terned, one may learn to emit two affects to the same stimulus or to emit two affects simultaneously as continuing moods somewhat independent of the environment. The combinations of affects leave each expression intact, side by side, but they may also produce resultants which correspond to no simple affect. It is in just such combinations that the common language is most likely to be deficient. To the extent to which affective phe- nomena of this kind escape language they will also remain opaque. Indeed much of the incommunica- bility of experience, mystical and prosaic, is a consequence of somewhat idiosyncratic combinations of affect which the individual finds as difficult to analyze for himself as to communicate verbally to another. (Tomkins, 1962, pp. 317—318) 32 Single affects, once miniaturized, can be stored, selectively combined with others, and then stored as a combined set. The set itself is then subject to mini- aturization, storage, and recall. The interplay of mini- aturization and imagery allows for a continual transfor— mation of innate patterns of affect expression, particu- larly in regard to temporality and potency of expression. Conscious control. The process of compacting affect can be complicated by loss of control over com— pacted affect experience. The loss of consciousness is achieved by the habitual use of the compacting process, which, like any skill that is well practiced and committed to memory, can lead to partial surrender of voluntary control. One result of this loss of consciousness is that learned sequences of miniaturized affect can become stable character traits that are relatively invulnerable to behavior change strategies. Tomkins believes that such transformations, which are the consequence of a defensive strategy to control affect, "may become so skilled a strategy that the individual becomes unconscious of what was once a voluntary act and ultimately unable to decontrol his habitual posture of defense" (Tomkins, 1962, pp. 320-321). Tomkins summarizes by pointing out that these transformations are inter—affect dynamics and that efforts to change and reverse their courses need to start with an 33 understanding of the sequencing, rather than of external, seemingly related events. He states: It should be noted in all of these types of learning that these are inter-affect dynamics unrelated to other aspects of experience. It is of course pos- sible to learn that whenever one feels distress another individual will do something to comfort one and thus reduce the feeling of distress. We are calling attention to a type of learning somewhat different from this, in which the activated affect becomes the stimulus for its own intensification and gradual reduction, or for its own sudden reduc- tion through becoming the stimulus for the activation of a particular competing affect. Whenever such learning becomes stabilized, the individual's account of the rise and fall of his own feelings either mis— takenly relies on plausible events going on at the time or seems to him frankly beyond explanation. Thus if distress is learned to activate more dis- tress, the individual may believe that the problems which initiated his distress are more serious and more insoluble than he at first realized. When one examines such phenomena closely, one will find that an individual whose affects have been so transformed by learning always encounters problems which appear to become more difficult as he struggles with them and that this is an artifact of the learned sequence in which mild distress activates more intense and more enduring distress. In order to diagnose this type of learning, one must establish invariance of the affective sequences which are independent of the course of external events. (Tomkins, 1962, pp. 325- 326) (Emphasis added.) Carroll Izard (1971) takes Tomkins' views further and speculates about the impact on behavior of imaged and micromomentary affect. Izard suggests that micromomentary imaged expres- sions of affect may be maladaptive because of their lack of precision, specificity, and complexity, and, depending upon the completeness of the affect experience they 34 afford, may be a crucial variable in human creativity and talent. He states, I doubt that the mediating memory image could ever be as precise, specific, and complex as the neuro— muscular activity of the face and the face-to-brain feedback. Consequently, the subjective experience would be less precise and less complete. The degree of precision and completeness which would accrue from the addition of the diminished or micromomentary facial activity to the memory image may make the difference between the person who can experience and respond to the subtleties and nuances of emotion and the one who cannot. (Izard, 1971, p. 193) Edmond Jacobson (1967) offers a different approach to the study of compacted affect. Jacobson does not hypothesize a transformation process that shortens emotion. He suggests instead that all emotion is in part miniscule neuromuscular contraction patterns. He believes these contraction patterns are goal-oriented responses that are central to man's being. He believes that the emotion process is often triggered by imagery. Jacobson suggests that imagery may be the key to understanding micromomentary affect expression. He believes that maladaptive emotion is based in part on imagery below threshold that escapes autosensory observation. Jacobson claims that "States of anxiety and depression are triggered by pertinent visual imagery which may be persistent or intermittent, clear or vague, steady or fleeting, observable or unob— servable" (Jacobson, 1967, p. 143). To remedy such mal- adaptive imagery based emotion, Jacobson believes people must be taught first to discriminate imagery which 35 pertains to maladjustments. If such discriminations can be made, then one can learn to selectively relax imagery pertaining to maladjustment. The problem entailed in this behavior change pro- cess is that the more "vague," "intermittent," "fleeting," and "unobservable" the images are the more likely they are to be "subvisualized" (below the lower limits of visualization) and thus the less they are subject to conscious autosensory discrimination and ultimately to any behavior change strategy (i.e., relaxation). Paul Ekman (1971) presents a social—cultural explanatiOn for the occurrence of micromomentary affect. Ekman suggests affect may be controlled and transformed by a learned system of "display rules." He believes display rules govern facial behavior on a habitual basis. Display rules transform affect by intensifying, deintensi- fying, neutralizing, or masking one affect by the expres- sion of another. Ekman and Friesen (1969) and Wilson (1976) suggest that these transformation processes may result in "micro-displays" of affect. They believe micro- displays are defensive attempts to govern one's affect via deception. Wilson suggests that if self-deception is the aim of micro-displays of affect, repression is the psychodynamic process at work. (Freud believed affects themselves could not be repressed but that repression of ideas associated with instincts or affects could suppress 36 the development of affect and transform affect expression; thus suppression and control of unwanted affect is the prime aim of repression (Freud, 1915). Thus, micro- displays would seem to be an indication that either affect suppression or affect control has taken place. In this sense the concept of micro-displays would seem similar to Tomkins and Izard's concept of miniaturized affect. In summary, regardless of the concept employed (miniaturized, miniscule, micromomentary, or micro-display) or the purpose hypothesized (motivation, evaluation, com- munication, or deception) the impact of patterns of com- pacted affect on person-environment interactions is believed to be maladaptive. These theoretical formulations have explanatory power for the process of learned helplessness. For if vague and fleeting emotion-evaluation processes triggered by imagery are transferred from one uncontrollable situ- ation, the ability to adjust to the new situation will be limited by one's ability to control the vague and fleet— ing image-induced emotion; emotion that is generally con- ceded to be outside of conscious control efforts. Research adding support to the above theoretical formulations will be reviewed in the following section. 37 Compacted Affect: Research Support Landis and Hunt (1939) in their studies of the startle response offer insight into the expression of miniaturized and combined affect. Using high-speed film (up to 3,000 frames per second) they studied over 100 subjects' body and facial reactions in response to a revolver shot. They found the following behaviors associ- ated with the startle response: mean reaction time of eye blink from beginning to full closure was 15 thousandths of a second, the mouth was widened for 67 thousandths of a second, and it took 83 thousandths of a second to initiate head movements. The average time for the total response to take place and for facial features to return to pre-startle conditions was 1/4 of a second. Of greater interest to this study is their finding that following the startle response comes a number of secondary responses. These responses are fleeting and oftentimes stylized versions of emotions. Within this secondary period affects such as surprise, fear, disgust, annoyance, and interest can be perceived in fleeting patterns of expres— sion. These secondary behaviors were observed to change with age. Children having clearer and fewer responses, basically consisting of fear, but with advancing age they observed increasing use of stylized and variable emotional responses. 38 They conclude their observations about secondary expressions by stating, Thus there are facial expressions of amusement, surprise, disgust, and others which, while they are not as immediate and uncontaminated by learning as is the startle pattern, come as quick, involun- tary responses to the stimulation and may be regarded as more primary and primitive than the voluntary and communicative usage of facial expressions mentioned above. The study of the minor changes in expression, from an involuntary, though learned, expression of surprise up to the voluntary use of the expression as a means of deliberate social communication, will help to bridge the gap between the innate and the acquired, between the primitive individual response and the social stereotype. It is in this gap that our secondary behaviors lie. (Landis & Hunt, 1939, p. 136) Landis and Hunt also set out to analyze the question of the mixing or blending of discrete secondary expressions. Taking the then popular concept of "inter- woven" secondary reactions (blended or mixed expression consisting of one expression appearing before another abates), they set out to analyze using high-speed film the occurrence of interwoven expressions. Landis and Hunt conclude: With the "interwoven" sort of reaction there is some confusion between the two behaviors if the photography is done at normal camera speeds, but ultra-rapid motion pictures separate the consecu- tive behaviors much more clearly and dispel the confusion that bothered Straus. For this reason we see little value in keeping the temporal cate- gories of "interwoven" and "postponed" secondary behavior. (Landis & Hunt, 1939, p. 144) Schwartz, Fair, Greenberg, Forna, Klerman (1974a), in a series of studies using electromyograms (EMG) to examine facial activity in response to self-generated 39 affect states, conclude: EMG provided a sensitive indi— cator of self-generated mood states and discriminated between subjects' happy, sad, and angry imagery. In one study, using a within subject design, 13 subjects were asked to generate happy and sad imagery and feelings while facial electromyograms were taken from sites above the right corrugator, depressor, zygo- matic, and mentalis muscles. Results indicated that happy and sad imagery resulted in significantly different patterns and that these imaged patterns were miniatures of the overt facial expressions of these emotions. In a variation of the previous study (1974), Schwartz et a1. added the affect of anger and included the measurement of the temporalis and orbicularis muscles. This study supported the original findings. That is, imaged-induced covert facial expression of the affects of anger, sadness, and happiness can be discriminated between, using electromyography. In another study, Schwartz, Fair, Greenberg, Mandel, and Klerman (1974) compared the facial expressions displayed by symptomatically depressed, clinically depressed, and nondepressed subjects. Twelve female volunteer subjects with Zung Self Rating Depression Scale (Zung, 1965) scores below 50, comprised the normal group. A group of six females with scores above 50 comprised the "symptomatic depressed" 40 group, and 12 clinically depressed subjects (with scores above 56) comprised the third group. EMG's were taken from sites above the corrugator, frontalis, depressor, and masseter facial muscles. All subjects were asked to image sad, angry, happy, and "typical day" feelings. The results indicate that normal subjects' happy, sad, and angry imagery can be differentiated electromyo- graphically, even when no overt expressions can be per- ceived. They state: "unbeknowns't to either the casual observer or the subjects themselves, they are actually generating small but distinct patterns of facial muscle activity that can be recorded and quantified by facial EMG." When normals are compared to clinically and symptomatically depressed subjects, the depressed sub- jects show similar sad and angry EMG patterns but dis- similar happy and "typical" facial patterns. Depressed subjects' typical patterns resembled sadness while normals resembled happiness. The authors conclude that EMG can provide a sensitive and objective measure of subtle facial adjust- ments. Additionally they conclude that it is possible to discriminate between normal and depressed subjects in a manner that confirms clinical diagnosis. 41 Haggard and Isaacs (1966) used slow motion film to study micromomentary facial expressions (MME's) of psychotherapy clients. MME's are facial expressions that are so short lived that they seem to be "quicker—than— the—eye." The authors had initially noticed that in analyzing films of psychotherapy hours, clients would occasionally change their facial expression from a smile to a grimace and back to a smile in less than 1/5 of a second. They then set out to analyze this phenomenon more systematically. Rather than speeding up the photography (as Landis and Hunt did), the authors chose to analyze slow motion film. They adopted the following "objective" scoring system: (a) judges were told to assume a state of passive alertness; (b) film was divided into scoring units with 120 judgments required for each 20—second unit; and (c) each expression Change should involve movement from one identifiable affect to another. They then compared scores obtained at slow speed to changes identified at normal speed and found that judges observed 2% times as many changes when the film was run at slow speed. They also found that many of the observed changes frequently occurred in "bursts," compounding the difficulty for those judgments made at normal speed. Of interest to this study is their observation that one depressed client showed great variability from session to session. During 42 one hour the patient emitted 98 MME's but during another hour, "characterized by a stubborn depressive mood," only emitted three MME's. Thus suggesting depression is characterized by reduced expression of micromomentary affect. They offer a number of suggestions for future exploration. They question: (a) whether MME's may inter- fere with other behaviors, such as speech or cognition; (b) whether situational factors affect the expression of MME's; (c) whether MME's may be efforts to express unacceptable affects; or (d) whether MME's may be trans- ference, counter-transference, phenomena that signal therapeutic progress or impasse. They conclude by asking, might micromomentary expressions indicate the existence of an ego mechanism which we may call temporal censorship. Thus, in addition to the ego controls which regulate or even inhibit the expression of particular impulses or affects or the awareness of particular fantasies or memories we assume that these controls also have a temporal dimension. (Haggard & Isaacs, 1966, p. 165) Inman (1976) using a similar format asked trained judges to identify on the faces of taped subjects, the number and type of changes in affect expression displayed (in slow motion and normal speed tape playback). He found that judges observed a significantly greater number of changes at slow speed than at normal speed but that their accuracy of identification of affect 43 scores were no greater at slow speed. These findings tend to support Haggard and Isaacs' contention that there may be a separate temporal dimension to the ego control of affect expression. The research exploring the expression of compacted affect is limited and inconclusive. All that can be said at this time is that the findings tend to support the contention that the display of compacted affect coincides with psychological stress and maladaptive behavior. Learned Helplessness--Depression and Emotion Seligman (1973, 1975) presents a theory of depres- sion that suggests that depression results from learning that response and outcome are independent. Thus, the individual believes events are uncontrollable and he feels helpless. The features that Seligman believes are common to both learned helplessness and depression are summarized in Figure 2.2. Seligman and associates have conducted a number of studies designed to test the learned helplessness- depression assumptions. Miller (1974; Miller & Seligman, 1975) treated depressed and nondepressed subjects with either escapable noise, inescapable noise, or no noise. During subsequent anagrams testing the depressed no noise subjects and the nondepressed inescapable subjects per- formed similarly on the anagrams test. Both groups' 44 'h—ad— .7; 7,-1 , Learned Helplessness Depression SYMPTOMS Passivity Passivity Difficulty learning that Negative cognitive set responses produce relief Dissipates in time Time course Lack of aggression Introjected hostility Weight loss, appetite Weight loss, appetite loss, social and sexual loss, social and sexual deficits deficits Norepinephrine depletion Norepinephrine depletion and cholinergic activity and cholinergic activity Ulcers and stress Ulcers (?) and stress Feelings of helplessness CAUSE Learning that responding Belief that responding and reinforcement are is useless independent CURE Directive therapy: forced Recovery of belief that exposure to responses that responding produces produce reinforcement reinforcement Electroconvulsive shock Electroconvulsive shock Time Time Anticholinergics; Anticholinergics (?); norepinerphrine stimu- norepinephrine lants (?) stimulants Figure 2.2. Summary of features common to learned helplessness and depression (adapted from Seligman, 1975, p. 106). 45 performances were significantly different from the other groups studied, thus indicating that depressed and ines- capable conditions may result in similar symptomatic behavior, at least in regard to subsequent task perfor- mances. In another learned helplessness study comparing subjects pretreated with escapable or inescapable noises, Gatchel, Paulus, and Maples (1975) found that subjects treated with an escapable noise and who demonstrated behavioral deficiencies on an anagrams task initially showed significantly heightened depression anxiety and hostility scores as measured by the Multiple Affect Adjestive Check List but that these differences dissi- pated after completion of the anagrams task. They con- cluded that the results support Seligman's (1975) con- tention that learned helplessness is associated with a heightened state of emotionality. They also noted that the heightened scores dissipated after successful solving of anagrams and that this suggests that regaining control over outcomes may be critical in efforts to successfully treat affect disorders induced through uncontrollability. Klein, Pencil-Morse, and Seligman (1975) treated depressed and nondepressed subjects with an unsolvable, solvable, or no—problem task. They found that the depressed no-problem subjects performed similarly to nondepressed unsolvable subjects on a subsequent anagrams 46 task. They concluded that learned helplessness and depression cause similar behavioral deficits. Gatchel, McKinney, and Koebernick (1977) treated depressed and nondepressed subjects (as measured by the Beck Depression Inventory) with escapable, unescapable, and passively listened to aversive tones. Then all groups were asked to solve a series of anagrams. This study supports the findings of Miller and Seligman (1975), Gatchel, Paulus, and Maples (1975), and Hiroto and Selig— man (1975). They found that depressed subjects and non- depressed inescapable subjects demonstrated parallel deficits on the anagram tasks, thus in their words "demonstrating a similarity of impairment in naturally occurring depression and laboratory induced learned helplessness" (Gatchel, McKinney, & Koebernick, 1975, p. 25). The authors administered the Multiple Affect Adjective Check List before and after the experimental treatments. They found similar elevated scores for inescapable subjects after treatment and depressed sub- jects prior to treatment. They concluded that these results also indicate a similarity between learned help- lessness and naturally occurring depression. Skin resistance was continuously monitored throughout the experiment and revealed that learned helplessness was associated with less phasic skin 47 conductance responding, while depression was associated with greater responding to uncontrollable adversive tones. The authors conclude that the data indicate that there is not a direct parallel between learned helplessness and naturally occurring depression with respect to elec- trodermal responding. Miller, Seligman, and Kurlander (1975) tested a depressed anxious, a nondepressed anxious, and a non- depressed nonanxious group of subjects with a set of discrimination problems. They found that depressed sub- jects evidenced a lack of expected control during the discrimination task as well as performance deficits on discrimination problems, as compared to nondepressed anxious and nondepressed nonanxious subjects. On another measure, time taken to shut off an alarm, the groups did not differ. The authors hypothesized, based on the learned helplessness model of depression, that depressed subjects would show behavioral and expectancy deficits of a greater magnitude than nondepressed subjects. They concluded that the results tended to support the learned helplessness model of depression. In an effort to determine if helplessness (like depression) generalizes across tasks and motivational systems, Hiroto and Seligman (1975) conducted a series of experiments where they pre-treated subjects with (a) escapable, inescapable, or no tone, followed by 48 anagram testing; (b) solvable, unsolvable, or no problem, followed by anagrams testing; (c) escapable, unescapable, or no tone, followed by shuttle-box tone avoidance test— ing; and (d) solvable, unsolvable, or no discrimination problems, followed by shuttle box tone testing. The results of the experiments indicate that there was sig— nificant interference on all the test tasks (both ana- grams and tone) for those pre-treated with unescapable tone when compared to those pre-treated with escapable or no tone. They performed more poorly on all the ana— gram and tone solving measures: trials needed to find anagram (or shuttle box) pattern, number of failures to escape, and mean response latency. They also found significant differences between those pre-treated with unsolvable problems when compared to those treated with solvable and no problems for both anagrams and shuttle box testing. Thus, the four experi- ments resulted in test task interference on all the measures. The authors conclude, . . . since it is general across motivations and tasks we suggest the process induced by uncontrol- lability may be the rudiment of a [depression like] "trait" . . . this speculation entails the conse- quence that our procedure for producing debilitation should also produce those symptoms associated with mild depression. (Hiroto & Seligman, 1975, p. 327) The results of the helplessness studies, while not always conclusive, taken together, tend to offer strong support for the learned helplessness model of 49 depression. In addition, when the Multiple Affect Adjec- tive Check List was given (Seligman, 1975; Miller & Selig- man, 1975; Gatchel, Paulus, & Maples, 1975; Gatchel, McKinney, & Koebernick, 1977) both depressed and learned helplessness subjects (depressed subjects before treatment and inescapable subjects after treatment) tended to evi- dence higher depression, anxiety, and hostility scores than control subjects. The heightened anxiety and hos-l tility scores support Seligman's contention that learned helplessness results in an initial state of heightened emotionality but tend to contradict his belief that depression is a state in which affectivity is character— ized by responselessness. In summary, compacted facial affect has received theoretical attention for over 100 years, but the diffi— culty scientists have had in quantifying the temporal dimension of affect expression has hindered scientific research attempts to study the compacting process. Recent theories of psychopathology and psycho- therapy have brought back to attention the need to study compacted facial affect. Recent theorists suggest that compacted affect plays a major role in the creation and maintenance of maladaptive behavior and hypothesize that those suffering from such maladaptive behavior need first to de-compress their affect before other behavior changes can take place. 50 Tomkins (1963) hypothesizes that depression is a maladaptive behavior pattern that is maintained by heightened levels of compacted facial affect, while Seligman (1975) suggests that depression is a state typified by reduced levels of affect expression. To date, no studies have focused directly upon the question of whether depression consists of increased or decreased levels of transient facial affect. CHAPTER III DESIGN OF THE STUDY Overview The purpose of this study is to test for the occurrence of transient changes of facial expression during learned helplessness conditions. To test for the effects of learned helplessness on facial expression, two groups of subjects were treated with a series of un- controllable or controllable stimuli. One group of sub- jects could not escape a repeated presentation of a loud tone (no-escape/learned helpless group); the other (escape group) could. After treatment with the tone, both groups were given the task of trying to solve 20 single-solution anagrams. During the anagram-solving task, subjects were monitored for facial behavior. Data derived from facial electromyograms taken from sites above the frontalis, masseter, depressor, and mentalis facial muscles were used to test for the effects of treatment on transient changes of facial expression. Data collected were analyzed using analysis of variance procedures. The remainder of this chapter describes the design of the study. 51 52 Sample Thirty—four students volunteered to participate in the study. But, due to a noisy electrical ground which caused the computer to fail erratically, data on 16 sub- jects (the lost data subjects and their yoked counterparts) were lost. The following description of the sample relates to the remaining 18 subjects. Eighteen volunteer subjects, who were students attending classes at Michigan State University or Lansing Community College during the summer of 1977, signed up to participate in the study (see Appendix A). The sub- jects were told that the experimenter was interested in studying physiological and psychological responses to problem—solving tasks; that the study consisted of an auditory and a cognitive task; that the auditory task included listening to a loud tone that was mildly unpleasant; that they would be given a sample of the tone prior to beginning their participation in the study; and that they could discontinue participating at any time during the course of the experiment. In addition, subjects were told that since the experiment involved listening to and responding to a tone, students with hearing impairments or disabilities could not be included as subjects in the study. Finally, the subjects were also informed that physiological recordings would be 53 taken during the tasks and that after completion of the experiment the experimenter would go over their record- ings with them. Apparatus The "helplessness" factor was manipulated by the presentation of a series of 20 trials with an escapable or inescapable 3000 Hertz 95 decibel tone (i.5 db). The tone was produced by an audio generator and was delivered to subjects through earphones. On each trial, the tone could be terminated by subjects in the escapable group by pressing two buttons on a button box (see Appendix B) in a predetermined "correct" order (see Appendix C). The inescapable group subjects were given the tone independent of button-pressing. Twenty single—solution five-letter anagrams (Appendix D) were used to test for the effects of the experimental treatment on subsequent task—solving ability. The anagrams were presented on a closed-circuit television monitor, with the numbers one through five printed below the letters. For example, the word "apron" was presented R O P N A 1 2 3 4 5' by pressing the five numbered buttons on the response as the anagram: The anagram could be solved panel in the correct sequence to spell the word "apron." Thus, the correct pressing sequence of "apron" is 5(A)- 3(P)-1(R)-2(O)-4(N). The number sequence 5-3-1-2—4 solved all 20 anagrams. ‘ —.s.. ._._ I 54 Electromyograms (EMG) were taken from the skin surface above the frontalis, mentalis, depressor, and masseter facial muscles (see Figure 3.1). The EMG read- ings were used to identify facial activity. Four pairs of miniature Beckman Ag/AgCl surface electrodes were used to record the EMGs for six consecutive seconds at the beginning of each anagram trial. EMG activity was channelled through a Grass polygraph and into a computer (PDP 11/40 DECLAB System) for storage. The EMG record- ing was sampled at 250 samples per second (Hz) by the computer. The EMG voltage was integrated over each 200 msec of each six—second sample, which yielded 30 integrated EMG scores. The data gathered from the four facial sites were used to test for the expressive effects of the experimental treatment. EMG recordings from the four facial sites included surface activity in the general area above the designated muscle; however, for reasons of convenience, activity will be reported as muscle- specific activity. The laboratory where the experiment took place consisted of two rooms: the Subject Room and the Record- ing Room. The Subject Room contained a chair, a closed- circuit television (CCTV) monitor, earphones, a subject response panel, and behavior-monitoring devices for recording facial activity, heart rate, respiration, and digital blood volume activity. The latter three recording 55 Frontalis (Ml) Masseter (M3) Depressor (M4) Mentalis (M2) Figure 3.1. Placement of surface electrodes. Schematic drawing of the face showing the locations of the frontalis, mentalis, masseter, and depressor muscles, and the locations of the surface recording electrodes. 56 devices were used to help mask the facial focus of the study. The Recording Room contained the Grass polygraph and the Digital Equipment Corporation PDP 11/40 Computer System. The computer was connected to the equipment in the Subject Room. Thus, once the subject was "wired" to the monitoring devices and received initial instructions for each task, all following instructions, feedback, and recording of physiological and task-solving behaviors proceeded automatically. The subject was visually moni- tored through a glass partition between the Subject and Recording Rooms. Procedure The 18* subjects were randomly assigned to the two treatment groups. Subjects were run individually, and "paired," such that the duration of tone received by a subject in the inescapable group was identical to the duration of tone received by the preceding subject in the escapable group. Orientation When the subject arrived for participation in the study, he was led into the Subject Room. There he was asked to read the following "orientation" sheet outlining the study: * Additional subjects were run, but due to a noisy electrical ground which caused the computer to fail, data on these subjects (and their yoked counterparts) were lost. 57 We are interested in psychological and physio- logical processes involved in thinking and problem- solving. The present experiment consists of three tasks: an auditory task (listening and responding to tones), a cognitive task (rearranging mixed letters, called ana rams, to Spell out a word), and an imagery task * (imagining some scenes). The instrumental and cognitive tasks will last about 20 minutes each, the imagery task about 10 minutes. You will be given detailed instructions for the tasks later. The entire experiment will last 90- 120 minutes. Some physiological responses will be recorded during the experiment. We use standard recording methods and the recordings will not be painful or harmful. It is important, however, that you remain still during the experiment, and avoid abrupt changes in breathing (holding your breath, sighing, or taking deep breaths), so that we can record your physiological activity accurately. Why are we doing this experiment? We are inter- ested in understanding psychological and physio- logical changes which may accompany thinking and problem-solving. When the experiment is over we will show you your physiological recording and dis- cuss it with you, if you are interested. If you call the lab at the end of the experiment (end of August) after we complete analysis of the data, we will be glad to tell you what we learned in this experiment and explain more details of the research. We ask that you do not share information about the experiment with other participants in your class until after they have participated. The Tone Task Once the subject had read the orientation sheet, he was asked to listen to a sample of the tone that was used in the experiment. If he desired to continue par- ticipation, he was asked to sign a consent form (Appen- dix E) and was then fitted with the physiological monitor- ing devices by a lab assistant. Subjects were seated in ** The imagery task was a separate experiment and will be reported elsewhere. 58 the chair with the CCTV situated in front of them. The subject's response panel was attached to the right arm of the chair. Subjects in both groups were given the following printed instructions for the auditory task: From time to time you will hear a loud tone. When the tone comes on there is something you can do to stop it. If you press the red and green buttons in front of you in the correct sequence and the tone stops, the words YOU STOPPED TONE will be shown on the T.V. screen in front of you. If you do not press the correct sequence, after a period of time, which will vary from trial to trial, the tone will stop automatically and the words TONE STOPPED AUTOMATICALLY will be shown on the T.V. screen. From time to time you will be given a new problem (new sequence of button presses) to solve. The number of button presses in the new sequence will be presented on the T.V. screen before each new problem begins, e.g., the message "4 BUTTON PRESSES" will indicate that the correct sequence on the following trials requires a total of four presses of the green and red buttons. Your job is to discover the sequence of button presses that will terminate the tone. During the course of the experiment, please move as little as possible since movement may interfere with our physiological recording. TRY TO AVOID UNNECESSARY MOVEMENTS . After the subject had read the instructions, he was informed that the experiment would start in about five minutes and that if he needed to communicate with the experimenter, talking out loud would be heard in the next room (via intercom) and the experimenter would respond. During this five-minute period, four six—second pretreatment baseline electromyogram recordings were taken. These were manually Schmitt-triggered and edited for interfering movements (artifacts) such as yawns, 59 eyeblinks, and swallows. Once the baseline measures were taken, the treatment task began. The treatment was computer-programmed to proceed automatically with presentation of tone, monitoring of button-pressing, tone termination, and presentation of feedback. Each subject was instructed that the tone could be terminated if he pressed the correct button sequence (see Appendix C). However, only the subjects in the escapable group could actually terminate the tone by pressing the correct button sequence; subjects in the inescapable group could not. The time of tone duration of subjects in the escapable group was computer-recorded, stored, and presented to the immediately following inescapable group subjects. Subjects in the escapable group who pressed the correct sequence of buttons termi- nated the tone and received the message YOU STOPPED TONE on the CCTV monitor. Otherwise, they received five to seven seconds of tone and the message TONE STOPPED AUTO- MATICALLY on the CCTV screen. The paired subjects in the inescapable treatment group always received the same duration of tone on each trial as the preceding escapable subject received on that trial, but they were always given the message TONE STOPPED AUTOMATICALLY on the CCTV screen. The process is illustrated in Figure 3.2. 60 Tone Button Pressing ---—-‘dp—— Feedback Screen #1 or #2 Message r————-—-—] } To Tx Tx+3 ‘q’_-- H-—-——~—— --- ------------- Elapsed Time-----------—-- v—3 II tone begins 0 TX = five to seven seconds, or completion of correct sequence of presses ITI = Intertrial Interval = 20 to 30 seconds * Screen Messages: #1: YOU STOPPED TONE #2: TONE STOPPED AUTOMATICALLY Figure 3.2. Time chart for one trial of the auditory task. 61 The Anagrams Task After each subject completed the auditory task, the lab assistant returned to the room and handed the following written instructions for the anagrams task to the subject: Your task is to figure out a series of anagrams. An anagram is a word which is presented with its letters in a scrambled order. Your job is to figure out the word and then spell it out cor- rectly using the keyboard in front of you. Notice that the buttons on your keyboard are labelled with numbers. The anagrams will appear on the TV screen before you and beneath each letter there will be a number. Once you have figured out the word you will have to determine the proper sequence of numbers to correctly spell the word on your keyboard. The idea is best illustrated by an esample. A typical anagram as it appears on the T.V. screen will look like this: U G A R S l 2 3 4 5 First, you should ignore the numbers and try to figure out the word. In this case, the word is "sugar." Once you know the word, look at the numbers to determine the proper sequence of numbers to correctly spell the word. In this case, the first letter is "s" so the first number is 5. The second letter is "u" so the second number is l. The third letter is "9" so the third number is 2. The fourth letter is "a" so the fourth number is 3. The fifth letter is "r" so the fifth number is 4. The sequence of numbers to spell "sugar," then, is 5, 1, 2, 3, 4 and you must press the buttons on your keyboard in this order——first button 5 (for "s"), then button 1 (for "u"), then button 2 (for "g"), then button 3 (for "a"), and then button 4 (for "r"). When an anagram first appears on the screen you will see the word THINK printed below it. A few seconds later the word THINK will be replaced by RESPOND. DO NOT begin to press buttons until the word RESPOND appears, even if you figure out the word immediately. We are interested in how quickly you can solve each anagram, so press your buttons as soon as you know the word and the mes- sage RESPOND comes on. 62 If you hit a wrong button while spelling a word you can press button number 8 to erase what you have pressed so far. Then you can start spelling that word over from the beginning. How- ever, once you have pressed the fifth and last button for a word the erase button will no longer work, so you must catch your mistakes before you finish spelling the word. There may be a pattern or principle by which to solve the anagrams, but that is up to you to figure out. After you have pressed the buttons, the word you typed in will be shown on the T.V. screen. There is a time limit of 45 seconds for solving each anagram. If you fail to figure out a word in 45 seconds, the words TIME UP will appear on the screen. There will be a brief wait between anagrams. TRY TO AVOID UNNECESSARY MOVE- MENTS. After the subject read the instructions, the ana- grams task began. Again, as in the tone task, the ana- grams task was computer-controlled to present the anagrams on the video screen, present feedback, and store problem— solving and physiological behavior. Figure 3.3 is a time chart illustrating the sequencing of the anagrams task. Debriefing After the subject completed the experiment, the experimenter entered the room, removed the monitoring devices, and answered any initial questions. Then, the general purpose of and rationale for the procedures were explained. Special attention was paid to explaining and answering questions about the treatments. This included exploring any reactions the subjects may have 63 Anagram Display 41 , i I o I I I ‘ i I ' I I I ' I Think I + I I I I Sign ' I I L I I 1 I I ' I ' I I I I I I I I . EMG Sample : I I : I I t———.' . F IL I I I I I I I Respond ‘ I : I ' Sign I L I I r J ;—F 77 I I I I 1 ' I I I I I I L I A # I I ' ‘ I screen I #1 1 #2 f #3 #4(.1,.2,.3) Messages . J I T0 T1 T7 T8 TX Tx + 3 Tx + 13 ITI TO = Program Display Begins TX = 45 seconds if subject has not pushed five buttons. = less than 45 seconds if subject has completed five button presses. ITI = 22 to 37 seconds. Screen Messages: . : ————— . ————— R 4.3 TIME UP #4 l CORRECT OR #4 2 INCORRECT O # Figure 3.3. Anagrams task time chart (one trial) 64 had to the procedures. The debriefing concluded with reviewing the polygraph printout of the subject's physio- logical behavior. Stages of Electromyogram Data Gathering Electromyograms (EMG) were continuously recorded at each of the four muscle sites during the six seconds (T to T7) of each of the 20 anagram test trials. 1 The first stage was the conversion of continuous EMG voltage to EMG scores (numerical representations of integrated EMG voltages). The computer sampled the con- tinuous EMG voltage entering the computer 250 times per second and converted each voltage sample to a number. The yield from each six-second EMG sample was 1500 (250 per second X 6 seconds) numerical representations of raw data. The computer then applied an integration formula to each set of 50 numbers for each ZOO-msec time period of each six—second sample. This yielded 30 inte— grated EMG scores (5 per second x 6 seconds) (Figure 3.4A). The next stage was the transformation of inte- grated EMG scores into Z-scores. Each EMG data point was compared to the mean at-rest baseline EMG for each site to determine a change score (C/EMG) (Figure 3.4B). A distribution was formed for each muscle site, using all of the C/EMG scores from that site. Z-scores (based on C/EMG scores) were then determined for each C/EMG data point (Figure 3.4C). H.HmHHu xmmu Ecnmmcw mco HOM mcHHmzumo mump mo mommum mHOHmmc pumgo Bonv .vcHHmnumm mumc EcuqoxfiouuomHo mo mommum .v.m wHDmHm lllll mccoomeHHHs oooonuuuu uuuuunnmecooomHHHHs oooouuuuuuu InunmccoommHHHHs ooooununn cm...... m N H cm....... m N H om..... m N H 0.. v O 00. O 0 0 v .0. v N N H H 2 mm H mm H oN N HN N z N N N N z ... m . ... . . . m ... m H H v m z 00 N 00 N om H Nm H z N N N N z ... N . ... . . . N ... N v a N a 2 mm H mm H No N Ho H z N N N N z ... H . ... . . . H ... H m m m N 2 mm H mm H 00 N «H N z N N N N z H>HH>Huom R. w>HuHmom ammummHmHHv mmHoomIN 02m HMOHHwauom>m no moHoomIN 02m "O r0 mmuoomIN 02m Scum nonfixcmm "m uuuuuuuuuu mccoommHHHHs oooouuunuuuuiu qunumecoommHHHHs oooouunun cm . . . . . m N H om...... m N H ozm\o . . . ozm\o ozm\o ozm\o ozm ... ozm ozm 02m Hmmmoudmc a: ozm\o . . . ozm\o ozm\o ozm\o ozm ... 02m ozm ozm Hmummmms m2 ozm\o . . . ozm\o ozm\o ozm\o 02m ... ozm 02m ozm mHHchms N: ozm\o . . . ozm\o ozm\o ozm\o ozm ... ozm 02m ozm mHHmucoum Hz mmuoom mmcmno ozm “ mmuoom ozm 36m .< 66 EMG Z-scores were used to rank each of the muscle sites, ranging from greatest change activity to least change activity for each ZOO-msec time period. Using hypothetical numbers, Figure 3.4D and E illustrates this process for three time periods. This was carried out for all 600 time periods (30 time periods per trial X 20 trials). The rank orders were used to determine muscle order and muscle perseveration (runs) for EMG Z—score activity. Muscle pattern was determined by ranking (from l to 4) the facial recording sites for each period in descending order of EMG Z-score activity. Each possible order (pattern of muscle activity) was assigned a number (Figure 3.5). Muscle pattern numbers were assigned to all 600 time periods for each subject. Using the rank orders from Figure 3.4E, the first three muscle patterns would be 20, 21, 17. Musclegpattern perseveration was determined by the number of time periods that a muscle pattern was maintained once it was expressed. The dependent measures were derived from EMG Z-score muscle pattern and muscle pattern perseveration data. Design The design tested for the effects of the helpless- ness induction procedure on changes of facial expression. Pattern Number (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (ll) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) Figure 3.5. 67 Possible EMG Z Muscle Orders 1 2 3 = M1 M2 M3 M4 = M2 M2 M4 M3 = M1 M4 M2 M3 = M1 M4 M3 M2 = M1 M3 M2 M4 = M1 M3 M4 M2 = M2 M1 M3 M4 = M2 M1 M4 M3 = M2 M3 M1 M4 = M2 M3 M4 M1 = M2 M4 M1 M3 = M2 M4 M3 M1 = M3 M1 M2 M4 = M3 M1 M4 M2 = M3 M2 M1 M4 = M3 M2 M4 M1 = M3 M4 M1 M2 = M3 M4 M2 M1 = M4 M1 M2 M3 = M4 M1 M3 M2 = M4 M2 M1 M3 = M4 M2 M3 M1 = M4 M3 M1 M2 = M4 M3 M2 M1 M1 2 M2 3 m M Q —---Order-- Muscle pattern numbers. sites ranked in order of decreasing activity.) 4 = Rank Order Frontalis Mentalis Masseter Depressor (Muscle 68 That is, it tested whether treatment with escapable XE- inescapable stimulus (the tone) affected facial expression during later task-solving activities. The data were analyzed in two trial blocks of 10 anagrams trials each. The decision to analyze the EMG data in two sets of 10 trials each was based on Seligman's (1975) observation that there are two stages of affective reaction to learned helplessness induction. The first stage is a state of heightened emotionality, followed by a state of reduced emotionality. Even though Seligman (1975) believes that the stages of affectivity following uncontrollable events are solely an induction phenomenon, it was decided that trial block analysis could give additional information about the affectivity, and the possibility of affect stages, accompanying the process of helplessness general— ization. This 2 X 2 design is illustrated graphically in Figure 3.6. To test for the effects upon the facial changes of expression, the following dependent variable measures were used: (a) facial lability (FL), defined as the number of times each facial muscle pattern changed from one ZOO-msec time period to the next, divided by the total number of possible changes; (b) repeated facial pattern changes (RFC), defined as the number of times that a facial muscle pattern change was the repeat of a previous change, divided by the total number of possible pattern changes; 69 com :prpwm wmmso>< mo oEHB HoH>mnmn HMHOMM co pcmfiummuu mo muommmm "cmflmmo .c.m stmHm coHumHDQ mqumpumm 02m HMHomm mo HoQESZ H mcnmuumm mmmcwnu Hafiomm cmumwmmm NHHHHan HmHomm 0mm Ah NH m mmmommloz .I m H mdmomm H COHHMHDQ mcumpumm 0mm Hm GOHHmHSQ mcumuwmm 0mm Am mm Bzm281 W m "a, 5. .3 rfl’_,,LH—eev””'flflflwmfifld’flflEscapable Q 4. 2 H B 4.) 3. m m ”a 2. H m g 1. 5 z #1 #2 Trial Blocks Figure 4.3. Number of facial patterns displayed: Group means. Table 4.4 Summary of Patterns Displayed Analysis of Variance Factor F = P = Block #1 TRl X TR2 1.09 .15 Block #2 TRl X TR2 .293 .30 Block #1 X Block #2 1.59 .11 Interaction BL X TR .25 .32 TRl = Escapable Treatment Group TR2 = Inescapable Treatment Group (One-tailed tests) 84 Ho4a: There will be no significant difference in the duration of facial patterns displayed by the escapable and in- escapable groups in trial block number one. Ho4b: There will be no significant difference in the duration of facial patterns displayed by the escapable and in- escapable groups in trial block number two. The alternative hypotheses are: Ha4a: The facial patterns displayed by the inescapable group will be of significantly shorter duration than those displayed by the escapable group in trial block number one. Ha4b: The facial patterns displayed by the inescapable group will be of significantly shorter duration than those displayed by the escapable group in trial block number two. No significant differences were found between groups in the average duration of facial patterns dis- played for either trial block (trial block number one: F = 2.78, df 1/16, p .056; trial block number two: F = .027, df 1/16, p .43). Therefore, the null hypotheses Ho4a and Ho were not rejected. Further 4b analysis revealed no main effect for trial block (F = .024, df 1/16, p = .44) nor was there an interaction effect (P = 2.32, df 1/16, p = .07). 85 Figure 4.4 depicts the duration scores of the two treatment groups over trial blocks. Table 4.5 summarizes the mean duration of pattern univariate analyses. Test of Learned Helplessness To test whether treatment with unescapable noise had an effect on later task-solving behaviors, four anagram—solving behaviors were recorded. The behaviors were: (a) response time for correct anagram solutions; (b) number of anagram trials with correct solutions; (c) number of anagram trials with incorrect solutions; and (d) number of trials terminated with the TIME UP message presented on the video screen without the first button being pressed. These measures formed the depen- dent variables for the four hypotheses used to test for learned helplessness. Analyses of variance were calcu- lated with a 5% rejection region selected for all uni- variate F tests. Table 4.6 gives the means for all four anagram-solving measures by treatment group. The first hypothesis tested for the effects of treatment on response time for correct anagram solutions. H05: There will be no significant difference between the escapable and inescapable groups in mean response times for correct anagram solutions. Has: The inescapable group will have significantly longer mean response time for correct solutions than will the escapable group. 1000 900 800 700 600 500 400 300 Duration in msecs 200 100 Escapable Inescapable #1 #2 Trial Block Figure 4.4. Facial pattern duration: Group means. Table 4.5 Summary of Mean Duration of Pattern Analysis of Variance Factor F = P = Block #1 TRl X TR2 2.78 .056 Block #2 TRl X TR2 .027 .43 Block #1 X Block #2 .024 .44 Interaction BL X TR 2.319 .07 TRl = Escapable Treatment Group (one-tailed tests) TR2 = Inescapable Treatment Group 87 Table 4.6 Experimental Group Means: Anagrams-Solving Behaviors RegPonse Correct Incorrect TIME UP ime Solutions Solutions Solutions (Seconds) Escapable Group 8.5 8.8 2.7 6.6 Inescapable Group 10.5 9.4 2.0 6.4 No significant difference between the two treat- ment groups in regard to mean response times for correct anagram solutions was observed (F = .49, df 1/16, p = .25). Thus, the null hypothesis was not rejected at the p < .05 significance level. The second hypothesis tested treatment effects on the ability to correctly solve the anagrams. H06: There will be no significant difference between the escapable and inescapable groups in the number of anagram trials with correct solutions. Ha6: The inescapable group will have significantly fewer trials resulting in correct solutions than will the escapable group. 88 There was no significant difference between groups in number of correct anagram solutions (F = .06, df 1/16, p = .44). Thus, the null hypothesis was not rejected at the p < .05 level of significance. The third hypothesis tested treatment effects on the number of anagram trials with incorrect solutions. H07: There will be no significant difference between the escapable and inescapable groups in the number of anagrams trials with incorrect solutions. Ha7: The inescapable group will have significantly more trials resulting in incorrect anagrams solutions than will the escape group. No significant difference between groups was found in this measure of anagram-solving behaviors (F = .41, df 1/16, p = .27). The null hypothesis was not rejected at the p < .05 level of significance. The last hypothesis tested for differences between groups in the number of anagrams trials terminated with the TIME UP message presented on the video screen. 89 H08: There will be no significant difference between the escapable and inescapable groups in the number of trials terminated with the TIME UP message presented on the video screen. Ha8: The inescapable group will have significantly more trials terminated with the TIME UP message presented on the video screen than will the escape group (response time = 45 seconds without first button being pressed). There was no significant difference between groups for this variable (F = .01, df 1/16, p = .46). The null hypothesis was not rejected at the p < .05 level of significance. Summary Hypotheses were tested regarding the effect of treatment with escapable and inescapable noise upon facial expression emitted during a subsequent anagrams task. The dependent variables were based on electro- myogram recordings from four facial sites: the frontalis, mentalis, masseter, and depressor facial muscles. The measures derived from the electromyograms were: (a) num- ber of changes of facial expression (facial lability); (b) number of repeated facial changes; (c) number of categories of expressions exhibited, and (d) average duration of an expression. The data were analyzed in two trial blocks of 10 anagram trials each. The null 90 hypotheses were tested using univariate analysis of variance with a 5% rejection region chosen for all uni- variate tests. Analyses revealed that the unescapable group exhibited a significantly greater number of changes of facial expression (F = 4.56, df 1/16, p = .023) and repeated facial changes (F = 7.11, df 1/16, p = .008) during the first 10 anagram trials. No significant differences were found during the first trial block for number of categories of patterns displayed (F = 1.09, df 1/16, p = .15) nor for average duration of expression (F = 2.78, df 1/16, p = .056). During the second trial block, no significant differences were found between the treatment groups on any of the facial measures (changes of expression: F = .66, df 1/16, p = .28; repeated changes: F = 2.70, df 1/16, p = .31; patterns displayed: F = .293, df 1/16, p = .30; duration: F = .027, df 1/16, p = .43). There were no main effects for trial block and only one significant trial block—treatment interaction effect; that was for repeated facial changes (F = 4.11, df 1/16, p = .025). Additional univariate analyses of variance were performed to determine if treatment had an effect on sub- sequent anagram task-solving behaviors. No significant differences were found between the treatment groups on the four anagram-solving variables. They were: mean 91 response time for correct solution (F = .49, df 1/16, p .25); number of correct solutions (F = .06, df 1/16, p = .44); number of incorrect solutions (F = .41, df 1/16, p = .27); and number of TIME UP trials (F = .01, df 1/16, p = .46). CHAPTER V SUMMARY, CONCLUSIONS, AND DISCUSSION In this chapter, the study is summarized and con- clusions based upon the data analyses are presented, fol— lowed by a discussion relating the results of the study to facial affect theory and learned helplessness theory. Summary Early in his career Freud (1899) claimed, "as an indisputable fact," that all adult psychoneurotic symptoms were based on the re—experiencing of repressed "childhood sexual wishful impulses" which had been maintained by the transformation of the impulse-related affects (p. 645). Since that time the repeated use of transformed emotional behavior has been variously seen as a symptom, cause, or attribute of psychological maladjustment. Fenichel (1945) summarized the psychoanalytic View of affect transfor- mations by noting that the defenses against affects not only transform affects but invariably lead to rigid pat- terns of affect expression that are symptomatic of a person's psychopathology. 92 93 Recently, Loeb (1966), Haggard and Isaacs (1966), and Hinds (1976) have noted an affect transformation on the faces of psychotherapy clients which miniaturizes and combines affect to form an important part of the clients' symptomatic behavior. The nature of the transformation is the compressing and combining of single affects over very short (two seconds or less) time spans. Kell and Mueller (1966) termed this compressing and combining of affect "compacted affect." The dynamics of compacted affect relate to the temporal reduction of the experience of one affect by the expression of another. The purpose of this transformation is hypothesized to be the termi- nation or avoidance of unwanted affect. In addition to these clinical impressions, Tomkins (1962—1963), Ekman (1971), Jacobson (1967), and Izard (1971) have offered theoretical models of transient affect behavior that coincide closely with the clinical observations. Although psychological clinicians and theorists have suggested that compacted affect serves a defensive function that helps maintain maladaptive behavior, little research has been conducted to determine under what con- ditions and to what degree this affect transformation takes place. The purpose of this study was to test for the effects of uncontrollable task outcomes on transient 94 changes of facial expression. A secondary purpose of the study was to explore the use of electromyograms in the measurement of transient facial expression. To test for the effects of uncontrollable out- comes on facial expression, two groups of subjects were treated with either a series of uncontrollable or con— trollable tones. One group of subjects could not escape the repeated presentation of a loud tone (inescapable/ learned helplessness group); the other (escape group) could. This design format, using the inescapable and escapable treatment groups from Seligman's learned help— lessness "triadic" design, was chosen because learned helplessness following treatment with uncontrollable out— comes had been linked theoretically and empirically to depression (see Seligman, 1975, Chapter II). After treatment with the tone, both groups were given the task of trying to solve 20 single-solution anagrams. During the anagram-solving task, subjects were monitored for facial behavior. Data derived from facial electromyo- grams (EMG) taken from sites above the frontalis, mas- seter, depressor, and mentalis facial muscles were used to test for the effects of treatment on transient changes of facial expression. Eighteen subjects who were students at Michigan State University or Lansing Community College volunteered to participate in the study. Subjects were randomly 95 assigned to groups and run individually. The subjects in the two groups were paired so that each pair of sub- jects received identical instructions and amounts of noise. To measure transient changes of facial expression, pairs of Beckman miniature Ag/AgCl surface electrodes were placed above the four facial sites. Electromyograms were taken six seconds before the beginning of each ana— gram trial. The EMG recordings were computer-sampled at 250 samples per second. The EMG voltages from each site were integrated over each ZOO-msec period then trans- formed to Z-scores and ranked in order of greatest posi- tive change from resting baseline scores. The resulting 24 possible rank orders were termed facial "patterns." The following facial measures were derived from the integrated and ranked EMG data: (1) facial lability (FL); (2) repeated changes of facial expression (RFC); (3) "patterns" of expression displayed; and (4) duration of expression. To determine if there were different stages of emotionality during the anagrams task, the resultant data were divided into blocks of the first 10 and second 10 anagram trials. Based upon facial affect theory, which suggests that people under depressive conditions will compact and combine their expressions of affect, it was hypothesized that subjects in the inescapable group would express a greater number of changes of facial expression, repeated 96 changes of facial expression, fewer facial patterns, and would exhibit expression of shorter duration than escapable subjects. Analysis of variance was used to analyze the data. A 5% rejection region was chosen for all univariate tests. To determine whether treatment with escapable and inescapable noise had a behavioral effect on subsequent task—solving ability (produced a state of learned helpless- ness), four anagram—solving behaviors were recorded. They were: (1) response time for correct solution, (2) correct solutions, (3) incorrect solutions, and (4) time-up trials with no solution attempted. It was hypothesized, based on learned helplessness theory, that inescapable subjects would have longer response times, fewer correct trials, and a greater number of incorrect and time up trials. Analysis of variance was used to o\0 analyze the data. A 5 rejection region was chosen for all univariate tests. Effects of Treatment on Facial Expression Univariate F-tests were calculated to determine if there were differences between the escapable and inescapable treatment groups in regard to facial lability, repeated facial changes, patterns of facial expression, and duration of facial expression while trying to solve a series of anagrams. 97 Lability. It was found that subjects treated with inescapable noise displayed significantly more changes of facial expression (FL) during the first 10 anagram trials (p < .05) than did subjects treated with escapable noise. There was no significant difference found between the treatment groups during the second 10 anagram trials. There was no main effect for trial block nor was a trial block X treatment interaction found. Repetition. During the first trial block, sub— jects treated with inescapable tones repetitively used sequences of expression (RFC) more often than did subjects treated with escapable noise (p < .05). No difference was found during the second trial block. There was no main effect for trial block, but there was a significant (p < .05) trial block X treatment interaction. Patterns. During each of the trial blocks, sub- jects treated with inescapable noise were not signifi- cantly different from subjects treated with escapable noise in regard to the variety of patterns of expression displayed. There was no main effect for trial block or interaction effect found. Duration. The mean duration of a facial expres- sion displayed by the inescapable group was not signifi- cantly different from those displayed by the escapable 98 group during either trial block. There was no main effect for trial block or trial block X treatment inter- action effect found. Effects of Treatment on Anagram- solving Behaviors Univariate F-tests were calculated to determine whether there were differences between the inescapable and escapable groups in regard to the following anagrams task behaviors: response time for correct solutions, correct solutions, incorrect solutions, and "TIME UP" trials. Subjects treated with inescapable noise did not significantly differ (p > .05) from subjects treated with escapable noise in the time taken to solve the anagrams, nor in their number of correct, incorrect, and "TIME UP" trials, indicating that over the 20 anagram trials, the treatment groups did not differ in the degree of learned helplessness behavior displayed. Conclusions and Discussion The results of the analyses indicate that prior lack of control over task outcomes (the variable believed to be responsible for inducing learned helplessness) did significantly effect some facial behaviors but did not significantly effect subsequent task-solving ability. Based on the results of the study, three general conclusions were made: (1) EMG measures of facial changes were found capable of discriminating between the facial 99 activity of the treatment groups, suggesting that facial EMG may be a useful tool in the quantifying of compacted affect behavior; (2) the facial change findings tend to support the predictions based on compacted affect theory, suggesting that levels of compacting are related to pre- vious painful experiences; and (3) the lack of behavioral differences found on the anagram-solving measures calls into question the helplessness induction procedures used in this study and more importantly the assumptions under— lying the helplessness-inducing theory itself. These conclusions will be discussed in the following pages. The Facial Behaviors The most important findings of this study are that during the initial stages of the anagrams task sub- jects treated with inescapable tones exhibited signifi- cantly more changes of facial expression and repeated changes of facial expression than did subjects treated with escapable tones. These two findings have impli— cations for facial affect theory. These implications will be discussed in the following paragraphs. Facial expression of compacted affect is believed to have an important effect on human motivation and behavior (Tomkins, 1962; Kell & Mueller, 1966; Haggard & Isaacs, 1966; Loeb, 1966; Jacobson, 1967; Izard, 1971; Hinds, 1976). These authors hypothesize that adaptive behavior is interferred with by the repetitive expression 100 of sequences of compacted affect that serve as a defense against re-experiencing affect associated with prior pain- ful events. The findings that after uncontrollable tones inescapable subjects expressed a greater number of changes of expression and repeated changes of expression than escapable subjects lend themselves to the support of the part of the hypothesis that suggests that compacted affect is a defense against re-experiencing emotion associated with painful events. (The first part of the hypothesis-- that compacting interferes with adaptive behavior-~will be discussed in the "anagram—solving behavior" section.) To elaborate, in this study inescapable subjects treated with uncontrollable aversive tones showed initially (first trial block) heightened scores for facial lability and repeated facial changes when compared to the escapable subjects; but they lowered their levels of compacting following the experience of being able to control the outcomes of the first anagrams trial block, while the escapable group showed the opposite tendency. One inter- pretation that can be given to these results is that the direction of change in levels of facial compacting fol- lowed the direction predicted by compacted affect theory. Uncontrollable aversive outcomes were followed by compara- tively heightened levels of compacting. Then after exper— iencing controllability during the first trial block of the anagrams task, differences subsided. 101 The finding that inescapable subjects did not sig— nificantly differ from their escapable counterparts in the number of categories of expressive patterns displayed suggests that the uncontrollability of an aversive stimu— lus does not inhibit the variety of expressions displayed in future situations that are similar to the original aversive situation. This finding fails to support the theoretical assumption that compacting is associated with a limited number of categories of expression and with any assumed resultant psychological debilitation. This finding would also fail to support Seligman's assumption that the state of heightened emotionality following uncontrollable events consists primarily of one category of emotion, i.e., fear. In this study, subjects given uncontrollable outcomes generated greater transience and repetitive changes of EMG but did not exhibit a reduced number of patterns of expression when compared with controllable subjects. This suggests that no single type of expression dominates during periods of heightened levels of transience. This interpretation could be misleading. The pos- sibility exists that even if both groups displayed a simi— lar mean number of affect patterns, subjects in either group could have used one pattern 90% of the time and used the other patterns 10% of the time, while the other group used all patterns an equal number of times. The 102 repeated facial change comparisons indicate that the inescapable group did, in fact, use particular pattern sequences significantly more often than did the escapable group. Thus, the theoretical position may need to be adjusted so as not to suggest a limited emotional scope but a limited variability of expression within that sc0pe. Seligman's explanation of the etiology of depres- sion fits the overall results of the facial analyses. Seligman states: When a traumatic event first occurs, it causes a heightened state of emotionality that can loosely be called fear. This state continues until one of two things happens: if the subject learns that he can control the trauma, fear is reduced and may disappear altogether; or if the subject finally learns he cannot control the trauma, fear will decrease and be replaced with depression. (1975, pp. 53-54) The results of this study indicate that a state of heightened emotionality, comprised of repetitive sequences of compacted affect, was greater for those treated with an uncontrollable traumatic stimulus. The heightened facial lability and repeated facial change scores obtained during the first trial block by those treated with inescapable noise (the traumatic event) would be consistent with Seligman's etiological theory. The fact that significant differences were not found during the second trial block also seems to support Seligman's contention that if the subject learns he can control the trauma, the heightened state of 103 emotionality is reduced. During the first trial block, subjects did have the chance to control the trauma and went on to reduce their expression of facial changes and repeated facial changes in the second trial block. Seligman's contention that the "heightened state of emotionality" could "loosely be called fear" needs to be explored further. The muscles measured in this study are not commonly associated with fear but are associated with the expression of grief, sadness, and distress. Still, it may be that fear is an integral part of the compacting-sequencing process. Along a similar vein, Hinds (1976) has suggested that compacted affects are "interfaced with fear"; this, however, needs further study. An implication of the results of this study is that in order to understand the emotionality accompanying uncontrollable traumatic events and subsequent generalized behavior, researchers may need to study the compacting and sequencing of affect, as well as the affect content accompanying such traumatic situations. Anagram-solving Behavior The results of the anagram-solving comparisons indicate that the groups did not differ in their degree of learned helplessness—depression during the anagrams task. One of the original goals of this study was to provide evidence that would help determine what, if any, 104 relationship exists between depression and compacted facial affect. But the lack of significant group dif- ferences on the anagram-solving measures permits only speculative discussion about the relationship between depression and facial expression. Necessarily left un- answered is the original research question that asked whether depression is associated with an increase (as suggested by the compacted affect literature) in the expression of compacted patterns of facial affect. The following discussion will focus on possible explanations for the learned helplessness findings, including a critique of the standard learned helplessness paradigm. Eckelman (1976) states that for helplessness- inducing procedures to be effective, they need to meet the following conditions: (1) subjects expect that they will be able to control the outcomes, (2) subjects are exposed to a high amount of helplessness training (uncon- trollable outcomes), and (3) subjects believe that the outcome is important. These conditions may not have been met in this study. This study's procedures, while essentially similar, differed significantly from what could be described as a standard helplessness-inducing paradigm and could have resulted in a softening of treatment effects. A common procedure in learned helplessness studies (Seligman, 1975) is for the treatment groups to initially be given the task 105 of trying to solve a series of task trials with one solution-behavior, capable of being learned, that will solve all trials. For instance, during treatment with tone, it is usually necessary to discover only one solving-behavior, e.g., one series of button presses. For this study, it was necessary to solve seven different patterns of button presses in order to terminate all 20 tones. This was planned to give subjects a "real life" and meaningful series of controllable or uncontrollable events in the belief that this would result in stronger treatment effects. The reverse may have been true. Inescapable sub- jects may 393 have expected that they could control the outcomes of the latter tone trials. Inescapable subjects may have initially believed they could control the tone but because the task (the number of button presses neces- sary to solve the task) changed often (depending upon the success of the paired escapable subject), they may have come to believe that they simply did not have enough chances to learn the controlling behavior. In addition, the inescapable subjects were given the exact stimulus duration of tone as the preceding escapable subject had received for each trial. This resulted in tone of very short duration on many trials (where escapable subjects had found the sequence of presses) and could have resulted in a belief of no control over outcomes because of a lack of reasonable time to solve the task. 106 The combination of multiple solving-behaviors and short time periods in which to solve the task could have left inescapable subjects with the feeling that it was unreasonable to expect that anyone could have gained control over all outcomes. Thus, rather than concluding that they were unable to gain control, they could have concluded that their initial evaluation of expected con- trol was erroneous. In other words, they may have felt "set up" rather than "helpless." This would have vio- lated condition number one (expected control) of an effective helplessness-inducing procedure. Additionally under these conditions, 20 tone trials may not have been enough uncontrollable outcomes (Condition #2) to create generalizable feelings of "helplessness." Another and perhaps more telling interaction of this "lessening" of treatment effects calls into question the underlying assumptions of the learned helplessness paradigm. That is that helplessness induction procedures may not be producing behavioral deficits as a result of feelings of helplessness, but perhaps what is witnessed is a motivational deficit leading to withdrawal from transgressor (in this case the experimentor) associated requests for behavior that could be related to feelings of power defiance or other reactive feeling states. In this study pilot inescapable subjects were quizzed regarding their feelings following treatment with a 107 standard (one relatively simple series of button presses given repetitively) helplessness-inducing task. This procedure was changed to the one used in this study when pilot subjects reported that they with- drew from the (anagrams) task because of feelings of "being tricked," or "mistrust" and not because of help- less feelings. The procedures eventually used in this experiment were intended to give the inescapable sub- jects repeated controllable tasks that would be exper- ienced as potentially real and honest as well as con- trollable. Thus, the helplessness—inducing procedure util- ized in this study was just as uncontrollable as the standard procedures but had a lessened impact perhaps because it reduced mistrust of the experimentor and not because it produced reduced levels of helplessness feel- ings (in fact, it would seem plausible to assume it caused more feelings of helplessness). The point being made is that there is reason to be concerned with the underlying assumptions that suggest that behavioral deficits result from feelings of help- lessness following uncontrollable events. The tone task-solving procedures may also have had an effect on the escapable group subjects that could have led to a lessening of the differential effects of treatment. Most of the escapable group subjects 108 experienced initial success during the tone task, but this always led to an increasingly difficult task (a lengthier series of presses), which led to additional failures. This may have discouraged the transferring of task-solving behaviors to the anagrams task. In other words, these subjects may have been subjected to helplessness-inducing conditions. They believed they had control over outcomes (Condition #1), and they often experienced repeated and increasing amounts of negative outcomes which may have been perceived as uncontrollable as the task became more difficult (Condition #2). Another possible explanation of the lack of effects found would be that the anagram-solving task was different in important ways from previous studies using anagrams as a test task and that this affected the results of the anagrams task. For example, in most helplessness studies using anagrams as the test task, the subject has the full prescribed time period to try to solve the ana- grams, regardless of how many trials and errors were made in an effort to solve the anagram. In this study, the subject had but one chance: once the subject pushed the fifth button spelling out a five-letter answer, the trial was over. This may have had a differential effect upon the treatment groups' subsequent solving-behaviors; i.e., it may take a larger number of trial and error/success 109 responses with feedback than the test-task afforded before the effects of a helplessness-inducing treatment can take hold. This analysis "fits" the behavior of what Selig- man termed "the paradigmatic learned helplessness finding." Seligman states that when a dog pre-treated with inescap- able shock was placed in a shuttle box from which it could escape, . it ran around frantically for about thirty seconds. But then it stopped moving; to our sur— prise, it lay down and quietly whined. After one minute of this we turned the shock off; the dog had failed to cross the barrier and had not escaped from shock. On the next trial, the dog did it again; at first it struggled a bit, and then, after a few seconds, it seemed to give up and to accept the shock passively. On all suc- ceeding trials, the dog failed to escape. (1975, p. 22) Thus, the dog had (and perhaps needed) the chance to make "frantic" adjustments to the environment before the effects of the previous helplessness-inducing treatment could be transferred to the new situation. It is interesting to speculate as to whether free— dom of frantic evaluation and reaction may be a necessary condition for the induction and transferring of learned helplessness behavior. If this were so, it would have important clinical implications. That is, controlling the conditions that allow for quickened evaluation and reaction may be, at times, a more efficient approach to countering the effects of prior uncontrollability than re- training the organism to believe that it has control. 110 Limitations This study is subject to a number of limitations which affect its generalizability. One limitation to the generalizability of the study is based on the fact that no other means for measuring transient facial expression, other than elec- tromyograms, were used in this study. Therefore, gen- eralizations to other methods of determining facial expressive behavior (e.g., self-reports, introspection, visual observations, and judgments based on visual recordings) are unwarranted. (Jacobson, 1967, and Schwartz et al., 1973, 1974, found that subtle facial EMG activity was generally not observable Visually nor recountable introspectively without extensive training.) Additionally, since only four facial muscle sites (of more than 100 muscle sites) were monitored with elec- tromyograms, generalizations to other surface muscle sites, or the "face" per se, is unwarranted at this time. Subjects in this study were college students who volunteered to participate after they had been informed of the study's general purpose. They were told that it was a study of psychological and physiological activity during problem-solving situations, that physiological measures would be taken, what types of tasks would be expected of them; and they were given a chance to hear the "mildly unpleasant tone" before they were asked to 111 make a final decision about participation. Thus, this was a highly restricted subject sample. Generalizations to other populations would be questionable judgment. Also, it must be noted that laboratory "staged" uncontrollable events often bear little resemblance to their "real-life" counterparts, so generalizing from laboratory conditions to in yiyg conditions is question- able. A design oversight and the omission of an addi- tional analysis limited information that could have been gathered in this study. The differential trial block analyses performed on the facial behaviors (two trial blocks) and the anagram-solving behaviors (one block) limited interpretative direct comparisons between the behaviors. And the omission of a test for the simple effects of trial block on the facial behaviors of both groups precluded discussion about the significance of the changes observed. Finally, a threat to internal validity could have been introduced when all EMG facial recordings were screened for major facial reactions such as yawns, eye blinks, and head movements. When such movements were observed, the six-second data sample was thrown out. This could have resulted in a differential loss of transient facial behavior from the groups. For example, it was assumed the yawns were random displays of nonexpressive 112 behavior, but it could be that one or the other of the groups was yawning to cover up displays of emotion stimu- lated by the pretreatment. Implications Psychological clinicians have been vocal about the importance of compacted affect behavior to mental health. They believe that the recalcitrance of patterns of compacted affect makes treatment of maladaptive behavior extremely difficult. The inability of clini- cians to observe transient behavior and identify exactly what affects are being displayed and in what order inter- feres with their diagnosis and behavior change efforts. Continued exploratory research in the following three areas could give clinicians needed information about transient affect behavior. 1. Electromyography has been claimed by Schwartz (1974) to be a sensitive and objective means for quantita- tive and qualitative measurement of subtle facial adjust- ments. This research supports that claim and goes further: not only is electromyography a means of measuring subtle- covert facial adjustments, it also provides a means for measuring transient adjustments of facial expression. Further exploratory research into the use of electromyograms seems warranted. Such study might be aided by a cross-validation of electromyographically measured expression to affect with other instruments, 113 such as high-speed film and videotape. It would seem that high-speed film could be used to identify the subtle and covert facial configurations displayed at transient levels, while electromyograms could measure specific patterns of muscle activity associated with transient expressions. The following comments relate to the facial EMG procedures used in this study. The procedures were exploratory in the sense that they broke new ground in the study of facial EMG activity. The discriminations found between the treatment groups suggest that the pro— cedures used in this study deserve more attention and support Schwartz's work and conclusions about the use- fulness of facial EMG in the study of subtle facial affect behavior. One suggestion for refining the measure is that great attention needs to be given to observing and elimi- nating interfering EMG artifacts such as eyeblinks and yawns. For example, in efforts to measure the surface EMG activity above the corrugator muscle it was impossible to eliminate such artifacts from the recordings, calling into question the usefulness of using the corrugator muscle as a surface recording site (as Schwartz, 1973, 1974, has done in his studies of facial EMG). 2. The study of the emotionality accompanying the induction, generalization, and reversal of maladaptive 114 behavior is needed. The learned helplessness paradigm (whether it is helplessness or some other maladaption that is manifested) seems well suited for the exploratory study of emotionality during traumatic conditions. Not only does it allow for the study of emotion present under such conditions, but it offers the Opportunity to study the emotional process accompanying "helplessness" induc- tion, generalization, and reversal. This type of "time series" analyses would seem ideally suited to garner information useful to clinicians who must deal with the effects of such debilitating learning processes daily. The results of this study indicate that "help— lessness-emotion" studies would be aided by the following considerations: (a) that the more "real-life" the help— lessness-inducing treatment is, the more attention needs to be given to insuring the following conditions are met: initial belief in the controllability of the outcomes, large amounts of uncontrollable outcomes, and strong meaning attached to the ability to control outcomes; (b) since emotional reactivity to uncontrollable events theoretically occurs in stages, the research design needs to incorporate means to measure the stages of emotionality following helplessness induction; (c) that pre and post- testing (or the addition of an untested control group; this option would amount to Seligman's "triadic" design) are needed to determine the direction of change in 115 transient emotional behavior following uncontrollable events. Without this change in design, it cannot be determined if inescapable subjects are demonstrating mitigated reductions or accented additions of transience levels following uncontrollable events; (d) that an intensive design may be a more economical approach to exploring the parameters of emotionality following uncon— trollable events than are group designs. An intensive (N=l) study could provide detailed descriptive material about the emotional process accompanying traumatic learn- ing situations that could guide future research efforts. 3. Finally, clinicians need to know more about the meaning given to transient affect by those who exper- ience it. Emotion has been hypothesized to be a way in which humans evaluate their environment and make judgments about what action to take. The possibility that compacted affect is accompanied by compacted evaluation, judgment, and action needs to be explored. The study of the pheno- menology of compacted affect could supply information about the meaning of this segment of affective behavior and supplement the findings gained through observational methods. APPENDICES APPENDIX A SUBJECT SIGN-UP SHEET APPENDIX A SUBJECT SIGN-UP SHEET Dear student: We are looking for students to participate in a research study into problem-solving behavior. We are interested in understanding psychological and physiologi- cal changes that may accompany problem solving. The experiment consists of two tasks: an instrumental task (listening and responding to noise) and a cognitive task (observing and rearranging mixed printed symbols). The instrumental task will last about 15 minutes and the cog- nitive task about 30 minutes. The entire study will take about 90 minutes. Some physiological responses will be recorded during the problem solving (heart rate, pulse, breathing along with forehead, chin, and cheek muscles). When you finish, we will show you your physiological recording and discuss it with you. If you can participate you will be helpful to us in our efforts to better understand problem-solving behavior. For those of you who will be conducting your own research in the future you may find the procedures and the physiological recordings of added interest. If you would like to participate please leave your name, open time periods and the dates you could partici- pate, and I will contact you. Sincerely, Name Phone no. Roger Landvoy (Counseling Psychology Doctoral Candidate, Times and dates Tel. 351—1747) available William C. Hinds (Professor, Counseling and Personnel Services) 116 APPENDIX B DIAGRAM OF BUTTON BOX APPENDIX B DIAGRAM OF BUTTON BOX Buttons numbered one and two were operable during the tone task and buttons numbered one through five and number eight were operable during the anagrams task. 117 APPENDIX C CORRECT SEQUENCE OF BUTTON PRESSES FOR TONE TASK APPENDIX C CORRECT SEQUENCE OF BUTTON PRESSES FOR TONE TASK (a) 1, 2, 2, l (b) 2, l, l, 2 (C) 1, 1, 1, 2, 1 (d) 2, 2, 2, 1, 2 (e) 1, l, 2, 2, 1, 1 (f) 2, 2, 1, 1, 2, 2 (9)1, 1! 2! 2! 1! ll 2 118 APPENDIX D ANAGRAMS APPENDIX D ANAGRAMS Word (a) (b) (C) (d) (e) (f) (g) (h) (i) (j) (k) (l) (m) (n) (o) (p) (q) (r) (S) (t) APRON OPIUM WAGON MUSIC NOBLE GIANT BATON ROACH BATCH PEONY HABIT PATIO FLOUR COUGH FAULT POUND UNCLE PANIC HUMAN CLERK (Adapted from Tressalt and M. Press.) S. 119 Anagram '—l 2 NZZOCCGOHWOHPHVWmOI—Iw WWHUZHOGHHZOOOZBHOCO HCWZO>OWIUPM>O>HOC>Ww 3 4 NZOMUHEWOHKEEZHMOZZ’Z OZEFUCSIU'TJOWIUE'UUJWUJOZZSOS’ 5 Mayzner, 1966, Psychonomic APPENDIX E DEPARTMENTAL RESEARCH CONSENT FORM 1. APPENDIX E DEPARTMENTAL RESEARCH CONSENT FORM Michigan State University Department of Psychiatry I have freely consented to take part in a scientific study entitled: Psychophysiology of Problem Solving being conducted under the supervision of: Dr. Lawrence Van Egeren Academic Title: Associate Professor The study has been described to me and I understand the description that has been given and what my participation will involve. I understand that I am free to discontinue my partici- pation in the study at any time without penalty. I understand that the results of the study will be treated in strict confidence and that I will remain anonymous. Within these restrictions, results of the study will be made available to me at my request. I understand that my participation in the study does not guarantee any beneficial results to me. I understand that, at my request, I can receive addi- tional explanation of the study after my participation is completed. I have listened to a sample of the tone that will be used in the study and after hearing the tone agree to continue participation. Signed: Date: 120 REFERENCES REFERENCES Bell, C. Essgys on the anatomy of expression in painting. London: Longmans, Green, 1806. Brown, J. S., & Farber, I. E. Emotions conceptualized as intervening variables with suggestions toward a theory of frustration. Psychological Bulletin, 1951, 48, 465-495. Bruner, J. S., & Tagiuri, R. The perception of people. In G. Lindzey (Ed.), Handbook of social psy- chology. Cambridge: Addison-Wesley, 1954. Darwin, C. The expression of the emotions in man and animals. Chicago: University of Chicago Press, 1965. (Originally published, 1872.) Duchenne, G. B. Mecanisme de la physionomie humaine. Paris: Bailliere, 1876. Duchenne, G. B. [Physiology of motion demonstrated by means of electrical stimulation and clinical observation and applied to the study of paralysis and deformities] (E. Kaplan, M.D., J.B., Ed. and trans.). New York: Lippincott and Co., 1949. Duchenne, Mechanisme de la physionomie humaine ou Anaylse electro-physiologique de la expressiones des passiones, 1862. Ekman, P. Universals and cultural differences in facial expressions of emotion. Nebraska Symposium on Motivation, 1971, 19, 207—283. Ekman, P. Universals and cultural differences in facial expression. In J. K. Cole (Ed.), Nebraska Sym- posium on Motivation, 1971. Lincoln: University of Nebraska Press, 1972. Ekman, P., & Friesen, W. V. Head and body cues in the judgment of emotion: A reformulation. Percept. and Mot. Skills, 1967, 24, 711-724. 121 122 Ekman, P., & Friesen, W. V. Nonverbal leakage and clues to deception. Psychiatry, 1969, 32(1), 88-106. Ekman, P., & Friesen, W. V. Nonverbal behavior in psy- chotherapy research. In J. Shlien (Ed.), Research in Psychotherapy (Vol. III). American Psycho- logical Association, 1968. Ekman, P., & Friesen, W. V. Unmasking the face. Engle- wood Cliffs, N.J.: Prentice-Hall, Inc., 1975. Ekman, P., Friesen, W. V., & Ellsworth, P. Emotion in the human face. New York: Pergamon Press Inc., 1972. Ekman, P., Friesen, W. V., & Tomkins, S. S. Facial affect scoring technique: A first validity study. Semiotica, 1971, 3, 37-58. Fenichel, O. The psychoanalytic theory of neurosis. New York: W. W. Norton & Company, Inc., 1945. Freud, A. The ego and the mechanisms of defence. London: The Hogarth Press, 1966. Freud, S. On the psychical mechanism of hysterical phe— nomena (1893). Collected papers (Vol. I). New York: Basic Books, 1959. Freud, S. The defence neuro-psychoses (1894). Collected papers (Vol. I). New York: Basic Books, 1959. Freud, S. A reply to criticisms on the anxiety-neurosis (1895). Collected papers (Vol. I). New York: Basic Books, 1959. Freud, S. Further remarks on the defence neuro-psychoses (1896). Collected papers (Vol. I). New York: Basic Books, 1959. Freud, S. The aetiology of hysteria (1896). Collected papers (Vol. I). New York: Basic Books, 1959. Freud, S. The predisposition to obsessional neurosis (1913). Collected papers (Vol. II). New York: Basic Books, 1959. Freud, S. Repression (1915). General p§ychological theory. New York: Collier Books, 1963. Freud, S. The unconscious (1915). General psychological theory. New York: Collier Books, 1963. 123 Freud, S. The interpretation of dreams. New York: Avon Books, 1971. (Originally published 1899.) Gatchel, R. J., McKinney, M. E., & Koebernick, L. F. Learned helplessness, depression and physiologi- cal responding. Psychophysiolqu, 1977, 44(1), 25-31. Gatchel, R. J., Paulus, P. B., & Maples, C. W. Learned helplessness and self-reported affect. Journal of Abnormal Psychology, 1975, 84(6), 732-734. Gatchel, R. J., & Proctor, J. D. Physiological correlates of learned helplessness in man. Journal of Abnormal Ppychology, 1976, 88, 27-34. Gellhorn, E., & Loofbourrow, G. N. Emotion and emotional disorders: A neurpphysiological study. New York: Harper & Row, 1963. Haggard, E. A., & Isaacs, F. S. Micromomentary facial expressions as indicators of ego mechanisms in psychotherapy. In L. A. Gottschalk & A. H. Averback (Eds.), Methods of research in psycho- therapy. New York: Appleton-Century-Crofts, 1966. Hinds, W. C. Compacted affect: A theraputic concept. Paper presented at the 84th Annual Conference of the American Psychological Association, Washing— ton, D.C., September 7, 1976. Hiroto, D. S., & Seligman, M. E. P. Generality of learned helplessness on man. Journal of Personality and Social Psychology, 1975, 84, 311-327. Hiroto, D. S. Locus of control and learned helplessness. Journal of Experimental Psychology, 1974, 102, 187-193. Hunt, W. A. The relation of bright and dull pressure to affectivity. American Journal of Psychology, 1931, 48, 87-92. Inman, D. J. Investigation of affective facial expres- sions through slow motion and normal speed video- tape techniques. Ph.D. dissertation, Michigan State University, 1976. Izard, C. E. Face of emotion. New York: Appleton— Century-Crofts, Merideth Corporation, 1971. 124 Izard, C. E. Patterns of emotions. New York: Academic Press, 1972. Izard, C. E., & Tomkins, S. S. Affect and behavior: Anxiety as a negative affect. In C. D. Spiel- berger (Ed.), Anxiety and behavior. New York: Academic Press, 1966. Jacobson, E. Biology of emotions. Springfield, 111.: Charles C. Thomas, 1967. Jacobson, E. Progressive relaxation. Chicago: Uni- versity of Chicago, 1929. James, W. The principles of psychology. New York: Holt, 1890. (Originally published 1844.) James, W. What is emotion? Mind, 1884, 8, 188—204. Klein, D. C., Pencil-Morse, E., & Seligman, M. E. P. Learned helplessness, depression, and the attribution of failure. Unpublished manuscript. Landis, C., & Hunt, W. A. The startle pattern. New York: Farrar & Rienhart, 1939. Loeb, F. F., Jr. The microscopic film analysis of the function of a recurrent behavioral pattern in a psychotherapeutic session. The Journal of Neurosis and Mental Disease. Williams and Wilkins Co., 1966. Mayzner, M. S., & Tresselt, M. E. Anagram solution times: A function of word transition probabilities. Journal of Experimental Psycholpgy, 1962, 88(5), 510—513. Mayzner, M. S., & Tresselt, M. E. Anagram solution times: A function of multiple-solution anagrams. Journal of Experimental PsychologY, 1966, 14(1), 66-73. Mayzner, M. S., & Tresselt, M. E. Normative solution times for a sample of 134 solution words and 378 associated anagrams. Psychonomic Monograph Sup- plements, 1966, 4(15), 293-298. Miller, W. R. Psychological deficit in depression. Psychological Bulletin, 1975, 88(2), 238-260. Miller, W. R., & Seligman, M. E. P. Depression and learned helplessness in man. Journal of Abnormal Psychology, 1975, 84, 228-238. 125 Miller, W. R., Seligman, M. E. P., & Kurlander, H. M. Learned helplessness, depression, and anxiety. Unpublished manuscript. Nummenmaa, T. The language of the face. Jyvaskyla studies in education, psychology, and social research. Jyvaskyla, Finland: Jyvaskylan Yllopistoyhdistys, 1964. Nummenmaa, T., & Kauranne, U. Dimensions of facial expression. Academiae Paedagogicae Jyvaskyliensis Research Report, No. 20, 1958. Pederit, T. Mimik und physiognomik. Detmold, 1925. (Originally published 1867.) Peiper, A. Cerebral function in infancy and childhood. New York: Consultants Bureau, 1963. Plutchik, R. The emotions: Facts, theories, and a new model. New York: Random House, 1962. Plutchik, R. The role of muscular tension in maladjust- ment. Journal of General Psychology, 1954, 88, 45-62. Plutchik, R. What is an emotion? The Journal of Psy- chology, 1965, 81, 295-303. Schwartz, G. E., Fair, P. C., Greenberg, P. S., Friedman, M. J., & Klerman, G. C. Facial electromyography in the assessment of emotion. Unpublished paper read to the Society for Psychophysiological Research, October 1973. Schwartz, G. E., Fair, L., Greenberg, P. S., Mandel, M. R., & Klerman, G. L. Facial expression and emotion: An electromyographic study. Unpublished paper read at meeting of American Somatic Society, Philadelphia, Pa., 1974. Schwartz, G., Fair, P., Salt, P., Mandel, M., & Klerman, G. Facial muscular patterning to affective imagery in depressed and non-depressed subjects. Science, April 30, 1971, 192. Seligman, M. E. P. Depression: Fall into helplessness. Psychology Today, June 1973, 1(1), 43—48. Seligman, M. E. P., & Miller, W. R. Depression and learned helplessness in man. Journal of Abnormal Psychology, 1975, 84(3), 228-238. 126 Seligman, M. E. P. Learned helplessness. Annual Review of Medicine, 1972, 2;, 407—412. Spencer, H. The principles of psychology. Vol. I. New York: Appleton, 1890. (Originally published 1855.) Tomkins, S. S. Affect, imagery, consciousness. Vol. I. The positive affects. New York: Springer, 1962. Tomkins, S. S. Affect, imagery, consciousness. Vol. II. The negative affects. New York: Springer, 1963. Tomkins, S. S., & McCarter, R. What and where are the primary affects? Some evidence for a theory. Perceptual and Motor Skills, 1964, 18, 119-158. Zuckerman & Lubin, B. Manual for the multiple affect adjective checklist. San Diego, Calif.: Edu— cation and Industrial Testing Service, 1965. V. HICHIGRN STQTE UNI LIB 11ll1111111HIHWIHHIIWI 1111111 312931031609 RQ IES 1| 78 :11: fights; ~—