The Open Collections website will be unavailable July 27 from 2100-2200 PST ahead of planned usability and performance enhancements on July 28. More information here.

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Projection as a mechanism of defense Dyck, Murray James 1980

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


831-UBC_1981_A8 D93.pdf [ 2.85MB ]
JSON: 831-1.0095154.json
JSON-LD: 831-1.0095154-ld.json
RDF/XML (Pretty): 831-1.0095154-rdf.xml
RDF/JSON: 831-1.0095154-rdf.json
Turtle: 831-1.0095154-turtle.txt
N-Triples: 831-1.0095154-rdf-ntriples.txt
Original Record: 831-1.0095154-source.json
Full Text

Full Text

Projection as a Mechanism of Defense by Murray James Dyck B. A. (Honours) University of Winnipeg, 1978 A Thesis Submitted in Partial Fulfillment of l the Requirements for the Degree Master of Arts in the Department of Psychology We accept this thesis as conforming to the required standard The University of British Columbia 0 December, 1 9 8 0 In present ing th is thes is in p a r t i a l fu l f i lment of the requirements for an advanced degree at the Un ivers i ty of B r i t i s h Columbia, I agree that the L ibrary sha l l make it f ree ly ava i lab le for reference and study. I fur ther agree that permission for extensive copying of th is thes is for scho la r ly purposes may be granted by the Head of my Department or by his representat ives . It is understood that copying or pub l i ca t ion o f th is thes is fo r f inanc ia l gain s h a l l not be allowed without my wri t ten permission. Department of The Univers i ty of B r i t i s h Columbia 2075 W e s b r o o k P l a c e V a n c o u v e r , C a n a d a V6T 1W5 i Abstract Eighty-four male undergraduate students were divided into groups of higher and lower defensive subjects. Subjects were then assigned to one of three experimental conditions: experimental projection, no-projection control (NPC)., and no-threat control (NTC). After viewing male homosexually explicit slides, experimental subjects made attributions to similar and dissimilar others, and then completed anxiety and defensive compensation measures. Control groups f o l -lowed the same procedure except that NPC subjects were not allowed to make attributions, and NTC subjects were not ex-posed to the slides. Results indicated that higher defen-sive subjects attributed higher levels of negative charac-t e r i s t i c s to the similar target person than did lower defen-sive subjects, and obtained lower anxiety scores following this projection. The hypothesis i s made that defensive projection i s a functional method of coping with high levels of anxiety. i i Table of Contents page Abstract i Table of Contents i i L i s t of Tables i i i Introduction 1 Methods Subjects 18 Materials 18 Procedure . 2 1 Results 2Zf Discussion 37 Reference Notes 46 References k7 Appendix I 5 n Appendix II . 53 Appendix III 55 Appendix IV 57 i i i L i s t of Tables Table 1: Summary of Order Effect Analyses for Attribution Checklist Data p. 26 Table 2: Summary of ANOVA's for Attribution Checklist Data in which Experimental Condition was Factor One and Target Favourability was Factor Two p. 27 Table 3: Summary of ANOVA's for Attribution Checklist Data in which Level of Defensiveness i s Factor One and Target Favourability i s Factor Two p. 29 Table k: Summary of ANOVA's for Attribution Checklist Data in which Experimental Condition i s Factor One and Level of Defensiveness i s Factor Two P. 30 Table 5: Summary of ANOVA1s for Attribution Checklist Data in which Level of Defensiveness i s Factor One and Trait' Ratings are Factor Two p. 31 Table 6: Summary of Order Effect ANOVA's for ADS Data p. 35 Table 7: Summary of Analyses of Defensive Compensation Data P. 36 1 Introduction Despite the recent burgeoning research effort in the whole area of attribution and person perception, there re-mains one sub-area within the f i e l d which has to date re-ceived scant attention. The neglected topic to which I re-fer i s "defensive projection," that i s , the systematically-biased attribution to others of negatively valued t r a i t s or characteristics by an individual in order to defend his characterization of himself. Thus, in contradistinction to other forms of attribution, defensive projection has been assumed to be both a systematic and a self-serving Since i t s popularization as a distinctly psychological concept by Freud, projection has, as an explanatory tool, been used in widely varied contexts and has been defined in startlingly numerous ways (cf. Murstein and Pryer, 1959; Mintz, 1956; Murray, 1 9 5 1 ) . So abused and overused was the term that Henry Murray was f i n a l l y forced to exclaim, " i f projection i s everything, then i t i s nothing." It has probably been for this reason that research into the valid-i t y of the concept has been so restricted and the paucity of relevant ,data so pronounced. However, the efforts of a handful of investigators have, over the last two decades, so c l a r i f i e d and defined the f i e l d that continuing constructive research i s now. pos-sible. One of the chief among these has been David Holmes who, in his f i r s t review of the projection literature (Holmes, 1 9 6 8 ) , identified four distinct forms of projection on the basis of manifest characteristics relative to two . descriptive dimensions. The dimensions used by Holmes were as follows: whether the subject i s aware or unaware of the t r a i t which results in projection, and whether the individual projects his own t r a i t or a different (complementary) t r a i t . The four resultant forms of projection, then, are these: 1) similarity (classical) projection, the attribution of a tr a i t which the individual i s unaware of or denies in him-self; 2) attributive projection, the attribution of a t r a i t which the individual i s aware of or self-ascribes; 3) com-plementary projection, the attribution of a t r a i t different from (complementary to) one which the individual self-ascribes; and k) "Panglossian-Cassandran" projection, the attribution of a t r a i t different from one which the individ-ual i s unaware of or denies in himself. On the basis of his f i r s t review of the available evidence on projection, Holmes ( 1 9 6 8 ) reached the following conclusions. "There i s strong evidence for the projection of subject's own t r a i t or the complement of this t r a i t i f subject i s aware that he possesses the t r a i t . There i s no evidence for any type of pro-jection resulting from a t r a i t which subject i s not aware that he possesses (from the Abstract, Holmes, 1 9 6 8 ) . " Ten years after the appearance of these f i r s t conclu-sions, Holmes ( 1 9 7 8 ) published another review of the l i t e r -ature, this time focussing his evaluation on projection as a defense mechanism. This reevaluation led to a:set of more negative judgments consisting of the following: "(a) there i s no evidence that projection protects persons from the awareness that they possess undesir-able t r a i t s , (b) the use of projection follows only some of the predicted patterns, (c) there i s no evidence that projection results in undesirable t r a i t s being reevaluated as more positive, and (d) there i s no reliable or strong evidence that projection results in stress reduction (from the Abstract, Holmes, 1 9 7 8 ) . " The appropriateness of these conclusions has, however, been called into question (Sherwood, Note 1 ) . Holmes based his view that projection does not keep undesirable thoughts out of awareness largely on the outcome of his earlier review. In that paper he argued that there was no empirical support for classical projection -the at-tribution of a t r a i t which i s denied in oneself——so that i t i s now pointless to suggest that this denied t r a i t i s being defended against. In other words, i t i s pointless to speak of the defensive function of a mechanism which has i t s e l f not been demonstrated. Holmes then states that "no new evidence has appeared that would necessitate a change in the conclusion that was drawn earlier (p. 3 7 8 ) . " Holmes' assertion i s challenged by at least two stud-ies (Epstein and Baron, 1969; Halpern, 1977) extant at the time his review was published, and by one more recently published study (Sherwood, 1979). Halpern (1977) divided subjects into higher and lower defensive groups based on their responses to his Sexual Defensiveness Scale. After rank ordering a series of target photographs from most to least favourable, the subjects spent time viewing porno-graphic pictures. Following this, subjects rated themselves ' and the previously identified least favourable target person on a set of t r a i t rating scales; only the t r a i t l u s t f u l was used in the analysis. The results showed that higher defen-sive subjects attributed less lustfulness to themselves and more lustfulness to unfavourable target persons than did lower defensive subjects. These findings clearly support the classical projective mechanism. In the Epstein and Baron (1969) study, subjects were given experimentally manipulated GSR feedback to sentences they were required to construct around twenty stimulus words (ten of which were blatantly hostile). The GSR feedback, which was supposed to reflect unconscious h o s t i l i t y , was varied according to whether the subject was part of a "strong confrontation" condition in which the galvanometer indicated strong h o s t i l i t y , or whether the subject was part of a "mild confrontation" condition i n which case the galvanometer i n d i -cated mixed feelings. At this point, subjects estimated the galvanometer readings for hostile and non-hostile words ap-pearing in several taped TAT stories read in either a Negro or a Caucasian voice, rated the reader oh a number of seman-ti c differential scales loaded on evaluative dimensions, and rated their own h o s t i l i t y . Two significant results emerged from the measures of projection in this study. F i r s t , "strongly confronted subjects attributed more 'un-conscious h o s t i l i t y 1 to the stimulus groups than weakly confronted subjects. ... (Second), a s i g n i f i -cant f i r s t order interaction between confrontation and race was obtained. This interaction indicated that strongly confronted subjects rated the Negro stimulus (a) more negatively than the white, and (b) more negatively than weakly confronted subjects (Epstein and Baron, 1 9 6 9 , PP. 1 7 7 - 1 7 8 ) . " Even though there were no differences between GSR scores attributed to black and white stimulus persons, the tendency to deny h o s t i l i t y i n themselves by subjects when evaluating an outgroup member as opposed to when evaluating an ingroup member, i s consistent with what would be expected during classical (similarity) and attributive projection. Sherwood ( 1 9 7 9 ) induced dissonance in his subjects by giving them bogus personality feedback indicating that they were highly neurotic. Subjects were then informed that i t i s possible to make accurate judgments of other persons based on 6 the sound of their voice alone. At this point they listened to tape recordings of two voices, ostensibly those of a nine-teen year old female student (favourable ingroup target) andc. a thirty-one year old female philosopher charged with child abuse (unfavourable outgroup target). Following the audition subjects rated both the two target persons and themselves on neuroticism. Subjects who increased their self-ratings r e l -ative to previous self-ratings were labelled self-ascribers, while a l l other subjects were labelled deniers. The results were as follows: "in accordance with classical projection, those who denied a higher level of neuroticism in themselves attributed more of i t to the unfavourable target person than did subjects who ascribed neuroticsm to themselves. In accordance with attributive pro-jection, those who self-ascribed a higher level of neuroticism attributed more of i t to a favourable target person than did those who denied neuroticism in themselves (from the Abstract, Sherwood, 1979)." Given that each of the studies just described yielded evidence clealy consistent with the hypothesis of a "mechan-ism" of classical projection, the basis of Holmes* f i r s t conclusion i s c r i t i c a l l y undermined. Holmes' (1978) second conclusion deals with the "pat-terning of projection." This refers to the hypothesis which arose from dissonance theory which predicted that among 7 persons who self-ascribe a negative t r a i t , projection w i l l occur onto persons whom they regard as favourable. In this way, rationalization i s believed to be fa c i l i t a t e d by a t t r i -butive projection. For example, an individual might say, "certainly I possess this negative characteristic, but so does X whom I admire." Following his reevaluation of four studies (Bramel, 1962, 1963; Edlow and Kiesler, 1966; Secord et a l . , 1964) which he had previously cited (Holmes, 1968) in support of attributive projection and one additional study (Holmes, Note 2), Holmes (1978) suggests that, "in four (Bramel, 1962, 1963; Edlow and Kiesler, 1966; Holmes, Note 2) of the five relevant exper-iments, i t was found that when subjects were con-vinced that they possessed an undesirable t r a i t , they projected the t r a i t onto undesirable as well as desirable persons. This set of findings i s i n -consistent with the defensive theory of projection because seeing one's own undesirable t r a i t in un-desirable persons would appear to reaffirm the tr a i t ' s undesirability and thus maintain or enhance the threat to self-esteem imposed by the possession of the t r a i t (Holmes, 1978, p. 681)." However, Holmes has been c r i t i c i z e d for his indiscrim-inate use of the concept of projection (Sherwood, Note 1). In his review of the evidence, Holmes has not taken care 8 to distinguish projection from halo-based attributions. The halo effect refers to the tendency of subjects to make judgments which accord with their f i r s t impressions, "attributing high levels of the undesirable charac-t e r i s t i c s to target persons described in unfavourable terms (negative halo bias) and low levels of that characteristic to target persons decribed in fav-ourable terms (positive halo bias)(Sherwood, Note 1, P. 5)." Given that this tendency i s in operation during projection studies, the attribution of an undesirable characteristic to an undesirable target person should not be an unexpected outcome, indeed, i t was anticipated by Bramel (1962, 1963). Bramel (1962), arguing that the extent to which one projects i s , at least in part, a positive function of the amount of dissonance created by self-attribution of a nega-tive t r a i t (e.g. unconscious homosexual arousal), manipulated subjects* self-esteem so as to either enhance or attenuate the degree to which dissonance would be induced. Subjects were confronted with evidence (rigged GSR feedback) which indicated homosexual arousal, and then asked to rate the arousal of their experimental partner, for whom they had previously provided a favourability rating. In accordance with halo effect expectations, high levels of the negative characteristic were attributed to unfavourable partners by both the high and low dissonance groups, and moderate levels 9 of the negative characteristic were ascribed by both groups to moderately favourable partners. However, when i t came to the most highly favourable partners, only the low dissonance group, that i s the group from which the least projection had been anticipated, continued to make halo based attributions (low levels of arousal). The high dissonance subjects a t t r i -buted significantly more homosexual arousal to these most favourable partners. According to Sherwood (Note 1 ) , both the Bramel ( 1 9 6 3 ) and the Holmes (Note 2 ) studies exemplify the same issue. If one f a i l s to take into consideration the operation of the halo effect, the results appear to belie the hypothesized projective patterns. However, when the halo principle i s taken into account when interpreting the results of these studies, these patterns clearly emerge. The validity of the concept of patterning i s further underscored by the results of a study by Secord et a l . ( 1 9 6 4 ) where i t was found that "subjects projected their undesirable t r a i t s onto desirable persons but not onto undesirable persons (Holmes, 1 9 7 8 , p. 6 8 1 ) . " Holmes' third conclusion was that undesirable tra i t s are not reevaluated as being more positive following a t t r i -butive projection, that i s , attributive projection does not serve a dissonance reducing function. Although Holmes' specific conclusion appears to be ju s t i f i e d , we must take care not to generalize this finding to other possible 10 mechanisms for restoring cognitive balance. For example, the results of the Secord et a l . (1964) study led i t s authors to suggest that, "by attributing a need to a friend, a person i n effect reduces the prominence of the need in himself. Through this action he may assume that he 'is like others' in possessing the t r a i t . Although the ranked value of the t r a i t remained the same, i t i s also pos-sible that attribution of the t r a i t to a friend changes the referents for the value scale as well. Associating a negatively valued object with a posi-tively valued object may through induction reduce the magnitude of the negative valence. ... The shift in referents for need attribution and need value, however, would support each other in the direction of a reduc-tion in imbalance (Secord et a l . , 1964> p. 4 4 6 ) . " Additional means whereby attributive projection may serve a defensive function w i l l be discussed i n the following section. Holmes' (1978) f i n a l conclusion i s that the use of projection does not result in stress reduction, the defen-sive function traditionally associated with the mechanism. In the case of classical projection, since no empirical tests of the stress, reduction hypothesis have yet appeared in the literature, no conclusions are warranted. However, with regard to attributive projection, Holmes has been dir-ectly challenged by Sherwood (Note 1) who believes that 11 "there i s reliable evidence indicating that attributive projection has a stress reducing effect (Sherwood, Note 1 , P. 1 ) . " Sherwood (Note 1) made his affirmation of the defen-sive function of projection following his reevaluation of the six studies (Bennett and Holmes, 1 9 7 5 ; Zemore and Greenough, 1 9 7 3 ; Burish et a l . , 1 9 7 8 ; Holmes and Houston, 1 9 7 1 ; Stevens and Reitz, 1 9 7 0 ; Hellman and Houston, Note 3 ) cited by Holmes ( 1 9 7 8 ) and an additional study by Heilbrun. ( 1 9 7 8 ) . Both reviews agree that the studies by Bennett and Holmes ( 1 9 7 5 ) and Burish et a l . ( 1 9 7 8 ) 1 suggest a reduction in emotional distress following attributive projection. Of two studies (Holmes and Houston, 1 9 7 1 ; Hellman and Houston, Note 3 ) cited by Holmes ( 1 9 7 8 ) as be-ing inconsistent with the stress reduction hypothesis, Sherwood (Note 1) c r i t i c i z e d the Holmes and Houston ( 1 9 ? 1 ) study as being an invalid test of the hypothesis on method-ological grounds (a comparison group which failed to ex-clude the possibility of projection) and discounted the Hellman and Houston (Note 3 ) study since neither the data nor the manuscript are available for public examination. Although both the Stevens and Reitz ( 1 9 7 0 ) and the Zemore and Greenough ( 1 9 7 3 ) studies failed to obtain evi-dence of anxiety reduction on their direct measures, their 1 According to Sherwood (Note 1 ) , the study which Holmes refers to as the Burish et a l , ( 1 9 7 8 ) paper actually formed the basis of the Burish and Houston ( 1 9 7 9 ) report. 12 results are nonetheless suggestive of defensive functioning. Stevens and Reitz ( 1970) experimentally manipulated failure on an anagram solving task and then assigned subjects to conditions which either encouraged projection (e.g. rate other students and the general public on how well they would perform on the task) or discouraged i t (perform a distrac-tion task). Although subjects in the two conditions did not differ on anxiety reduction, the results indicate a signif-icant correlation between the use of projection and anxiety reduction. In the Zemore and Greenough ( 1973) study, male subjects' were told that they had scored low on masculinity and high on femininity on a previously administered test. Subjects i n a projection and a no-threat control group rated a target person on a number of t r a i t s including masculinity/ femininity while the no-projection subjects performed a dis-tractor task. Results showed that the use of projection did not result in anxiety reduction. A second measure of stress reduction (defensive compensation) required subjects to sel -ect an exercise level from a series of graduated d i f f i c u l t y . Subjects in the no-projection group chose exercise levels significantly more d i f f i c u l t than did subjects in either of the other two conditions, presumably attempting to compensate for the low masculinity which had been ascribed to them. Finally, a recent study by Heilbrun ( 1978) has ad-duced additional evidence for a defensive function for pro-jection. Heilbrun was concerned with the relationship 13 between projective and repressive styles of dealing with aversive information about the self. In an i n i t i a l session, subjects rated the favourability of a l i s t of 120 adjectives. Later, subjects listened to 22 triads of favourable, neutral and unfavourable words purportedly used by mothers asked to describe their college age children. Subjects then attempted to recognize these 66 descriptors on the original adjective checklist and indicated whether i t was more descriptive of themselves or similar others. After performing a distractor task, subjects once more were asked to identify the descrip-tors on the checklist. From this procedure two scores were derived: a projection score based on the extent to which subjects attributed recognized unfavourable words to others rather than to themselves, and a repression score based on the number of unfavourable descriptors not recognized ,on the second t r i a l which had been recognized on the f i r s t t r i a l . The results indicated an inverse relationship between the use of projection and repressive forgetting of negative t r a i t s . Heilbrun concluded that these results support the assumption of individual differences in the u t i l i z a t i o n of the major defense mechanisms. As Sherwood (Note 1) has pointed out, however, the "fact that projection was the f i r s t option available seems relevant. It may be that the individuals who u t i l i z e d i t were subsequently less l i k e l y to use repression because they had less need to (Sherwood, 14 Note 1 , p. 2 6 ) . " It appears then, that although the evidence indicative of a stress reducing role for attributive projection i s by no means overwhelming, Holmes' ( 1 9 7 8 ) negative conclusion i s unwarranted. In two studies (Bennett and Holmes, 1 9 7 5 ; Burish and Houston, 1 9 7 9 ) use of projection was followed by anxiety reduction, and in three studies (Stevens and Reitz, 1 9 7 0 ; Zemore and Greenough, 1 9 7 3 ; Heilbrun, 1 9 7 8 ) the use of projection was related to a decrease in defensive behaviour. It i s clear that in the current context of disputed conclusions, what, i s required i s a further test of the pro-jection hypotheses. However, given the problems of interpre-tation associated with previous tests, care must be taken to establish clear, defensible c r i t e r i a which w i l l allow for the unambiguous inference of the mechanisms in question. The requisite methodological c r i t e r i a have been established by Halpern ( 1 9 7 7 ) and Sherwood (Note 1 ) . Halpern ( 1 9 7 7 ) has argued that any adequate investi-gation of classical projection must be based on a definition of projection derived from Freudian theory, that only in this way can appropriate test conditions be established. Halpern arrives at the following three conditions: 1) a potentially threatening sexual or aggressive impulse must be aroused in subjects; 2 ) one must discriminate between those higher def-ensive subjects who are l i k e l y to project class i c a l l y and those lower defensive subjects who are not l i k e l y to project ''5 classically; and 3) a suitable target must be provided—in the case of classical projection a disliked and dissimilar other (Halpern, 1977). Sherwood (Note 1) has supplemented these conditions so that attributive as well as classical projection may be exam-ined. In addition, he has presented c r i t e r i a which allow one to distinguish projection from other systematic biases in person perception. F i r s t of a l l , Sherwood has suggested that lower defensive subjects may be expected to project attribu-tively rather than not at a l l , since they are more l i k e l y to self-ascribe the t r a i t or impulse in question. Second, a suitable target for persons who project attributively would be a liked and similar other. As was mentioned earlier, i t i s necessary to take both positive and negative halo biases into consideration before inferring the operation of a projective mechanism. According to Sherwood (Note 1), " i f the attributions of an experimental group are to provide evidence for a particular type of defensive projections, they must deviate from this baseline pattern (halo effect) in a predictable way. In the case of attributive projection, the pure halo effect should be markedly attenuated, primarily because of an attenuation in the positive halo bias, ... In the case of classical projection, on the other hand, the pure halo effect should be markedly exaggerated 16 in both the positive and negative directions (Sherwood, Note 1, pp. 5-6)." Thus, attributive projection may be demonstrated by evidence of relatively more attribution of a negative t r a i t or impulse to a favourable target person. Classical projection may be demonstrated both by increased attribution of a negative t r a i t or impulse to an unfavourable target person and by relatively decreased attribution of a negative t r a i t to a favourable target person. However, the question of whether or not projection serves a defensive function s t i l l remains. This problem can be assayed in two ways: 1) by determining i f there i s a re-duction i n affect associated with psychological distress (e.g. anxiety) following projection; and 2) determining i f there i s a reduction i n defensive behaviours or percep-tions following projection. By incorporating the conditions and c r i t e r i a requisite for a valid evaluation (Halpern, 1977; Sherwood, Note 1), this study purposes to test the four following hypotheses: 1) i f lower defensive subjects who self-ascribe a negatively valued t r a i t are provided with a favourably perceived (sim-i l a r ) target person, attributive projection w i l l occur;[ 2) i f higher defensive subjects who deny possession of a negatively valued t r a i t are provided with an unfavourably perceived (dissimilar) target person, classical projection w i l l occur; 3) i f subjects are allowed to project attribu-tively, they w i l l manifest a decrease in anxiety; and U) i f subjects are allowed to project clas s i c a l l y , they w i l l man-ife s t a decrease i n anxiety. 18 Method Subjects. Subjects for this study were undergraduate students enrolled in Winter Session courses in Psychology and/or English at the University of British Columbia, Subjects were s o l i c i -ted directly from the classroom by either the experimenter himself or by means a prepared statement read by the i n d i -vidual classroom instructors. Following the request for par-ticipation, volunteers were asked either to remain after class to sign up for an appointment or to do so by telephone. Aside from knowlege of the English language and the appropriate gen-der, there were no prerequisite qualifications for participa-tion. A total of 84 volunteers completed a l l aspects of the study. Materials. The materials used in this project included three sets of stimuli and five different measures, one independent and four dependent. The f i r s t set of stimuli was a series of eighteen male homosexually-oriented slides. The original photographs from which these slides were madeCconsisted of proofs and published items supplied by Centurion Press of Hollywood. The content of the slides ranged from pinup style nude males through to males performing f e l l a t i o upon each other and poses suggestive of anal intercourse. The second set of stimuli consisted of two brief de-scriptions of two f i c t i o n a l individuals. One of them was designed to create a favourable impression and reads as 19 follows: This person i s a twenty-one year old student at the University of British Columbia. He has just completed his third year of studies in modern European history. During the school year, he makes extensive use of the aquatic centre and plays intramural basketball. The second description depicts a less savoury character and i s designed to create an unfavourable impression. It reads, This person i s a twenty-seven year old man who i s currently unemployed. He has served time in Oakalla (a local corrections f a c i l i t y ) after being convicted on charges of assault and petty theft. As i s evident from the descriptions above, no reference i s made to the sexual preferences of the target individuals. The third set of stimuli include one male and two female pinups. The male and one of the female photographs were taken from magazines whose models are assumed to repre-sent a normative conception of physical attractiveness. The second female photograph was taken from the "readers* wives" section of a sex-oriented magazine and was judged by the experimenter to reflect a relatively lower level of physical attractiveness. The measures used in this study included a homosexual defensiveness questionnaire (HDQ), an affect adjective check-l i s t (AACL), an attribution checklist (AC), an attractiveness/ desirability scale (ADS), and a sexual arousal scale (SAS). The. HDQ (see Appendix 1) i s a questionnaire modelled after Halpern's (1977) Sexual Defensiveness Scale. The HDQ consists of six items taken from the Attitudes Toward Homosexuality-Scale (Dunbar, Brown and Amoroso, Note k) t six items generated by the investigator, and eight f i l l e r items taken from the Sexual Liberalism-Conservatism and the Sex Concern-Guilt scales (Dunbar, Brown and Amoroso, Note 4). The HDQ has a six point Likert scale format. Overall, the scale measures attitudes towards homosexuality per se, physical contact be-tween males i n general, male nudity, and toleration of homo-sexuals in social situations. Negative attitudes vis a vis these questions were assumed to constitute homosexual defen-siveness in this study and i t was presumed that persons ad-judged to be higher defensive on the basis of their stated attitudes would deny sexual arousal to homosexually oriented stimuli. The AACL (Zuckerman, I 9 6 0 ) i s a scale used to measure anxiety at the time of testing (see Appendix 2 ) . It consists of twenty-one anxiety related adjectives and sixty-eight f i l -l er items. Ten of the items are scored negatively (failure to check the item results i n the item being scored) and ele-ven are scored positively. The AC i s a t r a i t rating scale assembled by the inves-tigator. A l l but two of the twenty-two items are for f i l l e r purposes and were chosen a r b i t r a r i l y from the Personality Research Form Scales. The two items which were relevant to this study were the t r a i t s • " l u s t f u l " and "anxious." The test, which follows the six point Likert scale format, was intended to e l i c i t projection by the subjects. The SAS (see Appendix 3) i s a single item, six point Likert scale designed to obtain a simple self-report measure of sexual arousal. The ADS (see Appendix k) i s a set of three, six point Likert scale items designed to obtain a measure of the perceived attractiveness and desirability of the models in the three pinup style pictures. Procedure. I n i t i a l l y , a l l volunteers were asked to complete an informed consent form and the homosexual defensiveness questionnaire before making an appointment for an individual session i n which the study would be completed. Between ses-sions, the f i r s t subgample of HDQ's were scored and the med-ian determined. On the basis of these scores, subjects were separated into two groups: higher defensive (scored above the median) and lower defensive (scored below the median). After this f i r s t set of data (consisting of 55 completed HDQ's) had been collected, the procedure was changed so that a l l data, including the HDQ, were collected in a single i n d i -vidual session After completing the HDQ, subjects from both the higher and lower defensive groups were assigned in equal num-bers to one of three conditions. Assignations were made according to an agenda which ensured a counterbalanced order 22 of presentation within each set of stimuli and dependent measures. In the "experimental projection" condition, subjects were asked to view each of the stimulus slides for about ten seconds. Subjects viewed the slides through a small, port-able slide viewer which they operated themselves. After view-ing a l l of the.slides, they were asked to complete the sexual arousal scale. Next, the following instructions were read to the subject: In this next part of the study, I am going to read you a very brief description of an individual. As I'm reading i t , I would like you to form as complete and vivid a mental picture of this individual as you can. The description of one of the target persons was then read to the subject, was repeated, and the subject asked to f i l l out an attribution checklist for the person just described. This procedure was then repeated for the second target person. The order of target person presentation was alternated. Following this, subjects were asked to complete the affect adjective checklist. Then, they were given the attractiveness/desira-b i l i t y scale. After reading i t s instructions, subjects were asked to rate each of the three models according to the scale. The pictures were shown in counterbalance order. At this point, subjects were informed that the experiment was over and they were invited to sign up to have a summary of the 23 results mailed to them. A l l of the subjects' questions'were answered, and the subjects were asked to refrain from discus-sing the project u n t i l i t had been completed. In the "no-projection control" condition, subjects f o l -lowed the same procedure as was outlined above except that they were not read the target persons' descriptions and did not complete attribution checklists for those individuals. In the "no-threat control" condition, subjects followed the same procedure as in the experimental condition except that they did not view the stimulus slides and did not com-plete the sexual arousal scale. There were two c r i t e r i a for excluding the data of in d i -vidual subjects. F i r s t l y , any subject who was found to per-ceive the target person favourability inappropriately was ex-cluded. Secondly, any subject who rated their sexual orien-tation as being either bisexual or homosexual was excluded from the data analysis. 2k Results The results of this study disconfirmed each of the four major hypotheses which were proposed. Although some evidence for projection, and a defensive function for projection, did emerge, the attributive pattern was such that i t did not f a l l neatly into either of the two frameworks which had been posit-ed. And from this fact, i t follows that the stress reduction which was supposed to be attendant on the operation of the two mechanisms of projection, could not occur. Before examining each of the main hypotheses in turn, the implicit hypothesis that subjects who scored higher on the homosexual defensiveness questionnaire would tend to deny sex-ual arousal, and subjects who scored lower on the HDQ would self-ascribe sexual arousal was tested. Higher defensive subjects were those who obtained a score of kO or greater, and lower defensive subjects were those whose score was less than 40. Using a t-test to compare the sexual arousal scale scores of higher defensive subjects with those of lower defensive sub-jects, a significant.difference indicative of greater self-ascription by lower defensive subjects was obtained, t(5*f) = 3.225, £ < .05. In order to determine whether or not classical and attributive projection had been e l i c i t e d , attribution check-l i s t data were subjected to a series of manipulations. But before testing the hypotheses, an analysis of variance was performed on the data to discover whether or not there were any significant order effects with regard; to which target per-son was presented f i r s t . No significant differences were ob-p tained (see Table 1). Two separate two-way fixed effects analyses of variance were performed for the t r a i t s " l u s t f u l " and "anxious." Factor one was experimental condition (projection or no-threat con-trol) and factor two was target person (favourable or unfav-ourable). Both of these analyses revealed a main target per-son effect (lu s t f u l , F (1, .108) = 4.91, p< , 0 5 ; anxious, F (1, 108) = 19.43, £ <.05) but no experimental condition or interaction effects (see Table 2), Thus, across both the pro-jection and the no-threat control conditions, subjects rated the unfavourable target person as being more lu s t f u l and more anxious than the favourable target person. This finding sup-ports the notion that halo effect attributions would be oper-ating in this situation. The next two analyses (again based on data for the trai t s l u s t f u l and anxious) were concerned solely with sub-jects in the projection condition and examined the effects due to subjects' level of defensiveness (factor one) and tar-get favourability (factor two). The ANOVA based on the "lust-f u l " data yielded no significant effects, while from the "an-xious" data, the only significant effect to emerge was, again p For the purpose of order effect analyses, the data for two subjects was randomly eliminated so that the c e l l sizes would be equal. A similar procedure was performed for the ADS data order analysis. 26 Table 1 Summary of Order Effect Analyses for Attribution Checklist Data Data Base Source MS df F 1. High Defensive S's Target Favourability .1.5 1 1.03 Trait Anxious Order 4.2 1 2.87 Interaction .2 1 ..23 2. High Defensive S's Target Favourability a Trait Lustful Order Interaction 3. Lower Defensive S's Target Favourability 8.2 1 4.09' Trait Anxious Order 2.7 1 1.33 Interaction .7 1 .33 4. Lower Defensive S's Target Favourability 2.1 1 1.08 Trait Lustful Order .1 1 .02 Interaction 3.4 1 1.80 Note. for Mean Squares Within, df = 20. C r i t i c a l value for F was 4.35 a No ANOVA was calculated in this instance as three of the four means were equal and would have resulted in F = 0. 2 7 Table 2 Summary of Analyses of Variance for A t t r i b u t i o n Checklist Data i n which Experimental Condition was Factor One and Target Fav-o u r a b i l i t y was Factor Two. Data Base Source HS df F 1. T r a i t L u s t f u l Condition 1 . 7 5 1 1 . 4 2 Target 6 . 0 4 1 4.9V Interaction . 1 4 1 . 1 1 2 . T r a i t Anxious Condition . 0 1 1 . 0 1 Target 31.08 1 1 9 . 4 3 Interaction . 0 1 1 . 0 1 * P < . 0 5 target favourability, F (1, 26) = 9.69, £ <.05 (see Table 3). Thus, at this level of analysis, the only significant pattern of attribution to emerge appears to be halo based. For the third set of analyses of variance (see Table 4), the data were organized to assay interactions between experimental condition and subjects' level of defensiveness. Using data from the unfavourable target person, two ANOVA's were carried out, one for each of the two t r a i t s . No signif-icant effects were detected. Parallel analyses based on fav-ourable target person data also failed to yield significant effects. There was no evidence for any differential patterns of attribution regardless of experimental condition or sub-jects' level of defensiveness. The evidence for a non-hypothesized form of projection emerged from the fin a l set of analyses of attribution check-l i s t data. Two ANOVA's were done; one was based on the un-favourable target person, and the other on the favourable target person. The f i r s t factor was level of defensiveness and the second factor was ratings on each of the two.traits. In the former analysis, no significant effects resulted (see Table 5). However, the favourable target person data dis-closed a significant main effect for subjects', level of defen-siveness (F (1, 52) = 9.77, £<.05), though not for factor two or the interaction. Further analysis revealed significantly more attribution of the t r a i t " l u s t f u l " to the favourable target person by higher defensive than by lower defensive 29 Table 3 Summary of Analyses of Variance for Attribution Checklist Data in which Level of Defensiveness i s Factor One and Target Fav-ourability i s Factor Two. Data Base • Source MS df F 1. Trait Lustful Target Favourability 2.16 1 1.74 Defensiveness 2.16 1 1.74 Interaction 3.02 1 2,43 2. Trait Anxious Target Favourability 16.07 1 9.69* Defensiveness 4.57 1 2.75 Interaction 1.79 1 1.08 * p<.05 Table 4 3 0 Summary of Analyses of Variance for Attribution Checklist Data in which Experimental Condition i s Factor One and Level of De-fensiveness i s Factor Two. Data Base Source MS df F 1i Unfavourable Target Condition , 1 . 4 5 1 1 . 0 3 Trait Lustful Defensiveness .16 1 . 11 Interaction . 4 5 1 . 3 2 2 . Unfavourable Target Condition a Trait Anxious Defensiveness Interaction 3 . Favourable Target Condition . 4 5 1 . 4 3 Trait Lustful Defensiveness 3 . 0 2 1 2 . 9 0 Interaction 2 . 1 6 1 , 2.08 4 . Favourable Target Condition . 0 2 1 . 0 1 Trait Anxious Defensiveness 1 . 4 5 1 1 . 0 5 Interaction 5.16 1 3 . 7 5 No ANOVA was calculated sets of identical means in this instance which would have as there resulted were in F two = 0 . c r i t i c a l value for F was 4 . 0 3 \ Table 5 31 Summary of Analyses of Variance for Attribution Checklist Data in v/hich Level of Def ensiveness i s Factor One and Trait Ratings are Factor Two. Data Base Source MS df F 1. Favourable Target Defensiveness 1 1 . 1 6 1 9 . 7 7 * Trait. . 0 2 1 . 0 2 Interaction . 0 2 1 . . 0 2 2 . Unfavourable Target Defensiveness . 0 7 1 .Oif Trait 7 . 1 4 1 3 . 8 4 Interaction . 2 9 1 . 1 6 * P < . 0 5 32 subjects (t (26) = 2,if29» p.<.05); this was also the case for the t r a i t anxious (t (26) = 2.071, p<.05). However, at this point i t i s important to r e c a l l the ANOVA already cited (see Table k) which revealed no significant effects when the no-threat control condition was included in the analysis. Thus, though the absolute differences between higher and lower defen-sive subjects in attribution to the favourable target person are significant when only the extremes are compared, the changes relative to their respective control groups are non-significant. The data relating to the defensive function of projec-tion consists of anxiety scores as measured by the AACL and defensive compensation scores, a composite of the ADS re-sponses. These two measures w i l l be discussed separately. F i r s t of a l l , a two-way fixed effects analysis of variance was performed on the AACL data. Factor one in this analysis was experimental condition (projection, no-projection control, no-threat control) and factor two was level of defen-siveness (high or low). None of the effects were significant. A one way analysis of variance u t i l i z i n g data from low-er defensive subjects only, revealed significant differences between the treatment conditions, F (2, 39) = 3.695, p<.05. Comparisons between the groups indicated that this effect was due to significant differences between the projection and the no-threat control conditions (t (26) = 2.52, p<.05), between the no-projection control and the no-threat control conditions (t (26) = 2.45, P.<.05), but not between the projection and 33 the no-projection control conditions (t (26) = .29, £> .05)# Both of the groups of lower defensive subjects who were ex-posed to the stimulus slides were significantly more anxious than the group of lower defensive subjects who were not threatened, regardless of whether or not they were given the opportunity to project. A second one-way analysis of variance was performed using the data obtained from higher defensive subjects. In this case, the test s t a t i s t i c did not reach the c r i t i c a l val-ue. Subsequent planned comparisons, however, revealed an interesting result. Analogously to the analysis of projection by higher defensive subjects to a favourable target person, in this instance, although neither the projection (t-(26) = 1.207, p_>.05) nor the no-projection control conditions (t (26) = .841» P>.05) differed significantly from the no-threat control condition, they did differ significantly from each other (t (26) = 2.391, £ <.05). In fact, the subjects in the projection condition were less anxious after being given the chance to project (x = 6.285) than were the un-threatened control subjects (X = 7.857). Before moving on to the analysis of the defensive compensation scores, there remains one more AACL based find-ing to report. A comparison of higher and lower defensive subjects in the no-threat control condition revealed a large, though non-significant, difference between the two groups (t (26) = 2.013, £>.05). 34 As was mentioned earlier, the defensive compensation score i s a composite of ADS responses. It was obtained by summing the subjects' ratings for the two female models and then subtracting from this total the rating for the male model. After this reconstruction had occured, these scores were f i r s t examined for order effects. Separate one-way analyses of variance were performed for higher and lower defensive sub-jects. No significant effects were obtained in either case (see Table 6). A two-way analysis of variance was performed on the defensive compensation data. The f i r s t factor in this anal-ysis was experimental condition (projection, no-projection control, no-threat control) and the second factor was subjects' level of defensiveness (high or low). The results were as follows: both factor one (F (2, 78) = 3.692, p<.05) and factor two (F ( 1 , 78) = 11.965, £<.05) were significant, while there was no significant interaction effect (see Table 7). Additional comparisons (see Table 7) suggest that the significant main effect for experimental condition i s due primarily to differences between the scores obtained from the no-threat control condition and both of the other conditions. The results do not indicate a decrease in defensive compen-sation for subjects who are allowed to project. The signif-icant main effect for defensiveness i s due to substantially higher defensive compensation scores for higher defensive subjects across a l l conditions (X = 4.90) than for lower defensive subjects across a l l conditions (If = 3.40). Table 6 Summary of Order Effect Analyses of Variance for ADS Data. Data Base Source MS df F 1 . High Defensive S's Order .50 2 .06 2. Low Defensive S's Order 3.58 2 .48 , Note. A l l F values non-significant. 36 Table 7 Summary of Analyses of Defensive Compensation Data • ANOVA Data Base Source MS df F-t , 1. A l l Scores Condition 14.58 2 3.69* Defensiveness 47.25 1 11.96* Interaction 2.25 2 .57 2. High Defensive S's Condition 4.67 2 1.28 3. Low Defensive S's Condition 12.16 2 2.86 t-test Data Base Comparison df t 1. Low Defensive S's no-projection vs no-threat 26 2.58* 2. Low Defensive S's projection vs no-•threat 26 1.76 3. High Defensive S's no-pro j ec tion«vs no-threat 26 1.09 4. High Defensive S's projection vs no-•threat 26 1.42 * p-<.05 37 Discussion An unexpected outcome of this study was the finding that even though each of the four major hypotheses were dis-confirmed, an unanticipated pattern of projection did occur and appeared to serve an anxiety reducing function. This latter projection consisted of characteristics which had been defined as elements of both classical and attributive projec-tion. As a result of these findings, several new conclusions about the phenomenon of projection are worth considering. In addition, the results of this study argue forcefully for the necessity of defining clearly the conditions required for the e l i c i t a t i o n of the phenomenon in question, and for the inclu-sion in an empirical test, conditions which control for the influence of related phenomena. The most d i f f i c u l t of the conclusions to draw i s that projection has been e l i c i t e d at a l l . Were i t not for the fact that higher defensive subjects given a chance to project manifested a lower level of anxiety following projection, this conclusion might not have been reached at a l l . Given that this did in fact occur, i t colours slightly the examination of the projection data. The projection data revealed differences between higher and lower defensive subjects i n their attributions towards favourable target persons, with higher defensive subjects attributing higher levels of the negative t r a i t s . Comparison with their respective higher and lower defensive no-threat control counterparts indicates that this significant differ-ence was achieved through relatively increased attribution by higher defensive subjects and relatively decreased attribution by lower defensive subjects. In other words, for higher de-fensive subjects, there was an attenuation of the halo bias while for lower defensive subjects, the halo bias was exagger-ated. These facts leave the door open to draw one of several alternative conclusions. One would be simply to say that pro-jection did not occur since overall the projection condition did not differ from the no-threat control condition. Another would be to say that only lower defensive subjects projected (albeit negatively) since in absolute terms their scores dif-fered most substantially from their control group, A third possibility i s that i t was the higher defensive subjects who projected. This last one i s the conclusion to which I sub-scribe, largely as a result of the facts mentioned earlier, namely, that i t was these same higher defensive subjects whose anxiety level was lowest, lower even than that of the no-threat control group. To hold to this conclusion has a number of implica-tions. F i r s t of a l l , i t underscores the prerequisite c r i -terion identified by Halpern (1977) that only higher defensive subjects may be expected to project. Second, i t contradicts both Halpern's notion that projection w i l l occur onto an unfavourable target, and Sherwood's (Note 1) prediction that i t i s the lower defensive subjects who w i l l project onto 39 favourable targets. The projection by higher defensive sub-jects appears to have occured only onto the favourable target person. , The second and third conclusions of this study relate to the stress reducing or defensive function of projection. As the reader w i l l have gathered from the discussion above, i t i s clear that the use of projection by higher defensive subjects may serve an anxiety reducing function. However, besides this reduction in anxiety, there was no strong evi-dence for a reduction in other defensive behaviour such as defensive compensation. This was the case for both higher and lower defensive subjects. While i t i s true that following pro-jection higher defensive subjects were not significantly more defensive than non-threatened higher defensive control sub-jects, given the overall pattern of results this would not appear to argue for decreased defensiveness. This i s so for two reasons. F i r s t , higher defensive subjects in the projec-tion condition actually scored higher than higher defensive subjects in the no-projection control condition. If projec-tion had reduced the need for defensive compensation, i t would follow that the observed pattern would have been reversed. Second, given the generally high defensive compensation scores \ of a l l higher defensive subjects, including those in the no-threat control condition, the inference of a ceiling effect for these scores does not appear unwarranted. As for the lower defensive subjects, both of the groups which were exposed to the stimulus slides obtained defensive compensation scores significantly higher than lower defensive no-threat control subjects. Since i t has already been concluded that lower defensive subjects did not project onto either of the target persons, this discovery i s not surprising. Among the most interesting of the findings to emerge from this study was the consistent difference in responding between higher and lower defensive subjects. This feature surfaced within each of the measures used in this study, and suggests that the importance of this factor has been under-estimated in.previous research in the f i e l d . That these two groups would differ in their scores, on the sexual arousal scale was anticipated; that these differences would persist across other measures only now seems a log i c a l , i f unexpected, outcome. Given that both the HDQ and the ADS were constructed to measure aspects of the concept "defensiveness," i t follows that i f both scales are valid, persons adjudged to be defen-sive on the one may be expected to be adjudged defensive on the other. A similar consistency would also apply to subjects identified as lower defensive. For differences in anxiety scores, the argument i s less straightforward. If we may assume that higher defensiveness i s not an innate characteristic, i t would appear reasonable to assume also that the acquisition of the characteristic (or response mode) was adopted or learned because i t served some functional purpose. In addition, from the results of this 41 study we know that higher defensiveness i s associated with denial of unacceptable impulses (lower SAS scores) and the use of projection (increased attribution of negative charac-. t e r i s t i c s to favourable or similar others).. We also know that the use of projection leads to a reduction i n anxiety, a re-duction from a level of anxiety which i s probably not ascrib-able to the effects of the stimulus slides alone, since higher defensive subjects in the no-threat control condition were themselves more anxious than were lower defensive subjects in the same condition. From this set of assumptions and results, ' i t follows that the characteristic or response mode labelled "defensiveness" represents a functional method of coping with higher levels of anxiety. If this reasoning i s correct, i t would follow that persons already adjudged to be defensive w i l l also manifest relatively high levels of anxiety. The above conclusion i s not, of course, the only one which may be drawn from the data. The most obvious alterna-tive i s that both defensiveness and anxiety are due to some unspecified third variable. Another po s s i b i l i t y i s that anxi-ety i s i t s e l f a function of defensiveness. while both of these alternatives are feasible, they do not follow the "principle of parsimony." The one requires that we posit an additional variable, the nature of which i s entirely unknown. To accede to the other would require the formulation of addi-tional presuppositions. This study also highlights the importance of clearly ^ 2 defined response c r i t e r i a and appropriate comparison groups. Had this study incorporated only the conditions established by Halpern (1977) , neither projection not anxiety reduction would have been el i c i t e d . If the dichotomy between higher and lower defensive subjects had been neglected, no positive con-clusions would have resulted. Inclusion of both sets of c r i -t e ria made i t possible to ascertain that projection i s a unitary phenomenon, that i s , i t appears to be a single mech-anism used by a single specifiable population and i s directed towards a predictable class of persons. In the same vein, i t i s easy now to recognize the value of controlling for halo based effects. Although the specific conclusion with regard to projection would have been the same, the magnitude of the observed effect would have been unnecessarily exaggerated. More importantly, however, i s the fact that i t would in that case have been impossible to make . any general statements regarding the higher defensive subjects per se; and the extent to which this factor plays a role in defensive behaviour would have been overlooked. At this juncture, i t i s necessary to consider the possi-b i l i t y that demand characteristics of the experimental situa-tion may have influenced the outcome of the study. Prior to their participation in this project, subjects were informed that i t s purpose was to examine the way in which viewing ex-p l i c i t sexual materials may effect the impressions we form of other persons. The question now becomes, given this 43 information and the ever-increasing knowlege about the method and materials of the experiment i t s e l f , to.what extent i s i t l i k e l y that any subject w i l l be able to accurately deduce the specific hypotheses under investigation. On the basis of information gleaned from subjects during post-experimental discussions, and the relatively complex nature of the hypo-theses themselves, i t i s probably safe to suggest that few i f any accurate impressions were formed. What did emerge i s that subjects held widely varying opinions on the task, though most of them believed i t was mainly concerned with attitudes towards homosexuals. Since there was such divergence in the guesses of the subjects concerning the true nature of the study, i t i s impossible to assess the nature and extent of the influence exerted by demand characteristics. However, given this var-i a b i l i t y , i t i s unlikely that any resultant bias was con-sistent. Additional features of the experimental situation which require examination at this time are the stimulus items used in this study. Did these materials evoke the responses for which purpose they were chosen? The data indicate that the desired outcomes were obtained in each case. That the stimulus slides were successful i n raising anxiety i s evi-denced by the significantly lower AACL scores amongst no-threat control subjects. Similar data from the attractiveness /desirability scales underscores the success of this manipu-lation. Differential perceptions of the two target persons was confirmed by the attribution patterns of subjects in the no-threat control condition. These subjects consistently attributed higher levels of both the trai t s " l u s t f u l " and "anxious" to the target person designated unfavourable. The extent to which each of the stimuli were successful in bring-ing out the predicted responses thus lends credence to the validity of the dependent results. In summary, the results of this study represent evi-dence for a self-serving bias in person perception, namely, defensive projection. Contrary to prediction, the bias f o l -lowed neither of the hypothesized patterns. Rather, i t con-sisted of the attribution of a negative characteristic to a similar (favourable) other by a higher defensive subject. The use of this mechanism was observed to follow the viewing of anxiety arousing material and was found to be useful in reducing this anxiety. The data from this study also i n d i -cate that i t i s possible to discriminate higher from lower defensive individuals on the basis of their responses to a homosexual defensiveness questionnaire. Differences between these groups persist across a number of measures. Following from these conclusions, there are several new questions which may be posed for empirical resolution. For example, among a group of higher anxious individuals, i s there a greater proportion of persons who would also be judged higher defensive than there would be among lower anxious individuals? Is higher defensiveness as measured 45 by the HDQ associated with defensiveness with regards to say aggression, or perhaps to "conservative" values generally, or i s i t a more idiosyncratic descriptor? Answers to these questions may make i t possible to answer more valuative queries such as whether or not the use of defensive projection i s an adaptive means of dealing with anxiety. 46 Reference Notes 1. Sherwood, G. G. Self-serving biases in person perception: a review. Manuscript submitted for publication, 1980. 2. Holmes, D. S. Attribution (projection) of one's undesir-able t r a i t s : self-defence or person perception? Unpub-lished manuscript, 1976. 3 . Hellman, R. & Houston, B. K. A study of the stress-reducing effect of attributive projection. Unpublished manuscript, 1970. 4. Dunbar, J., Brown, M. & Amoroso, D. M. Sexual Liberalism-Conservatism Scale, Sex Concern-Guilt Scale, Attitudes Toward Homosexuality Scale. Unpublished tests, n.d 47 References Bennett, D. H.. & Holmes, D. S. Influence of denial and projec-tion on anxiety associated with threat to self esteem. Journal of Personality and Social Psychology, 1975, 32, 915-921. Bramel, D. Dissonance theory approach to defensive projection. Journal of Abnormal and Social Psychology, 1962, 64, 121-129. Bramel, D. Target selection for defensive projection. Journal of Abnormal and Social Psychology, 1963, 66, 318-324. Burish, T. G. & Houston, B. K. Causal projection, similarity projection, and coping with threat to self-esteem. Journal of Personality, 1979, 47, 57-60. Burish, T. G., Houston, B. K. & Bloom, L. I. Effectiveness of complementary projection in reducing stress. Journal of C l i n i c a l Psychology, 1978, 34, 200-206. Edlow, D. W. & Kiesler, C. A. Ease of denial and defensive pro-jection. Journal of Experimental Social Psychology, 1966, 2, 56-59. Epstein, R. & Baron, R. M. Cognitive dissonance and projected h o s t i l i t y towards outgroups. Journal of Social Psychology, 1969, 79, 171-182. Halpern, J. Projection: a test of the psychoanalytic hypo-thesis. Journal of Abnormal psychology, 1977, 86, 536-542. Heilbrun, A. B. Projective and regressive styles of processing aversive information. Journal of Consulting and C l i n i c a l  Psychology, 1978, 46, 156-164. 4 8 H o l m e s , D . S . D i m e n s i o n s o f p r o j e c t i o n . P s y c h o l o g i c a l  B u l l e t i n , 1968, 69, 2 4 8 - 2 6 8 . H o l m e s , D . S . P r o j e c t i o n a s a m e c h a n i s m o f d e f e n s e . P s y c h o -l o g i c a l B u l l e t i n , 1978, 85, 677-688* H o l m e s , D . S . & H o u s t o n , B . K. T h e d e f e n s i v e f u n c t i o n o f p r o -j e c t i o n . J o u r n a l o f P e r s o n a l i t y a n d S o c i a l P s y c h o l o g y , 1971, 2 0 , 208-213. M i n t z , E . A n e x a m p l e o f a s s i m i l a t i v e p r o j e c t i o n . J o u r n a l o f  A b n o r m a l a n d S o c i a l P s y c h o l o g y , 1956, 52, 270 -280 . M u r r a y , H . A . F o r e w o r d . I n H . H . A n d e r s o n & G . L . A n d e r s o n ( E d s . ) , A n i n t r o d u c t i o n t o p r o j e c t i v e t e c h n i q u e s . E n g l e -w o o d C l i f f s : P r e n t i c e - H a l l , 1 9 5 1 . M u r s t e i n , B , & P r y e r , R . T h e c o n c e p t o f p r o j e c t i o n : a r e v i e w . P s y c h o l o g i c a l B u l l e t i n , 1959, 56, 353-374. S e c o r d , P . F . , B a c k m a n , C . W. & E a c h u s , H . T . E f f e c t s o f i m -b a l a n c e i n t h e s e l f - c o n c e p t o n t h e p e r c e p t i o n o f p e r s o n s . J o u r n a l o f A b n o r m a l a n d S o c i a l P s y c h o l o g y , 1964, 68, 442-446. S h e r w o o d , G . G . C l a s s i c a l a n d a t t r i b u t i v e p r o j e c t i o n . J o u r n a l  o f A b n o r m a l P s y c h o l o g y , 1979, 8 8 , 635 -640. S t e v e n s , H . A . & R e i t z , W. E . A n e x p e r i m e n t a l i n v e s t i g a t i o n o f p r o j e c t i o n a s a d e f e n s e m e c h a n i s m . J o u r n a l o f C l i n i c a l  P s y c h o l o g y , 1970, 2 6 , 152 -154. Z e m o r e , R . & G r e e n o u g h , T . R e d u c t i o n o f e g o t h r e a t f o l l o w i n g a t t r i b u t i v e p r o j e c t i o n . P r o c e e d i n g s o f t h e 8 1 s t A n n u a l  C o n v e n t i o n o f t h e A m e r i c a n P s y c h o l o g i c a l A s s o c i a t i o n , 1973, 8, 343-344. 49 Zuckerman, M. The development of an affect adjective check l i s t for the measurement of anxiety. Journal of Consulting  Psychology, 1960, 2if, 457-462. Appendix I Homosexual Defensiveness Questionnaire 51 Instructions. Following, i s a series of statements concerning several aspects of human sexuality. Below each statement are the numbers one to six, number one meaning strong disagreement with the statement and number six indicating strong agreement. Please indicate your level of agreement by c i r c l i n g the number which corresponds to your opinions. 1. Premarital sexual relations often equip persons for more stable and happier marriages. 1 2 3 4 5 6 disagree agree 2. Things wouldn't be as bad as they are i f people exercised more control over their sexual impulses. 1 2 3 4 5 6 disagree agree 3. Homosexuality i s a rotten perversion. 1 . 2 3 4 5 6 disagree agree 4. Homosexuality ought to be suppressed. 1 2 3 4 5 6 disagree agree 5. I feel uncomfortable when a male friend embraces me. 1 2 3 4 5 6 disagree agree 6. Male nudity in the movies i s disgusting. 1 2 3 4 5 6 disagree agree 7. Sometimes I do sexual things which later I feel guilty about. 1 2 3 4 5 6 disagree agree 8. I would never be seen in public with a known homosexual . 1 2 3 4 5 6 disagree agree 9. Men should be encouraged to show more affection to each other. 1 2 3 4 5 6 disagree . agree 10. Religious groups should not attempt to impose their standards of sexual behaviour on others. 1 2 3 4 5 6 disagree agree 1 1 . I would invite known homosexuals to a party in my home. 1 2 3 4 5 6 disagree agree 12. Pornography corrupts the moral fibre of society. 1 2 3 4 5 6 disagree agree 13. I have never thought-of having sex with another man. 1 . 2 3 4 5 6 disagree agree ' , 52 14. Sex i s one of the biggest problems in our society. 1 2 3 4 5 6 disagree agree 15. I feel uncomfortable being undressed in public changing rooms and showers. 1 2 3 4 5 6 disagree agree 16. Homosexual marriages should be o f f i c i a l l y recognized. 1 2 3 4 5 6 disagree agree 17. When I am sad, I enjoy having a friend put his arm around me. 1 2 3 4 5 6 disagree agree 18. Oral-genital activity i s a dirty and disgusting practice. 1 2 3 4 5 6 disagree agree 19. If I found out that my friend was a homosexual, I would end our friendship. 1 2 3 4 5 6 disagree agree 20. Please rate your own sexual orientation on the following scale. 1 2 3 4 5 6 7 completely bisexual completely heterosexual homosexual in thought in thought and behaviour and behaviour. Appendix II Affect Adjective Check List 5k Listed below are a number of words which refer to mood. Please place a check mark on the line behind those words which apply to you right now. Be sure to indicate only the words which apply to your mood right at this moment. active frightened polite afraid furious rejected agreeable glad sad alive gloomy safe alone good secure amiable good-natured shaky angry happy steady awful healthy stormy bitter hopeless strong blue inspired suffering calm interested sunk cheerful i r r i t a t e d sympathetic clean joyful tame contented kindly tender cooperative l i v e l y tense cruel lonely terrible desperate lost t e r r i f i e d destroyed loving thoughtful didagreeable low tormented discontented lucky understanding discouraged mad unhappy disgusted mean unsociable enraged merry upset enthusiastic miserable vexed fearful nervous whole fine offended wilted f i t outraged w i l l f u l forlorn panicky worrying free peaceful young friendly pleasant. Appendix III Sexual Arousal Scale 56 Please indicate on the scale below how sexually aroused you are feeling right now. Circle the number which best corres-ponds to your level of arousal. 1 2 3 k 5 6 not at a l l very strongly aroused aroused Appendix IV Attractiveness/Desirability Scale 58 Please indicate on the scale below how attractive and desir-able you perceive the models to be right now. Circle the num-ber which corresponds to how attractive you think they are. Male Model not at a l l attractive Female Model I 1 not at a l l attractive 5 6 very attractive 5 6 very attractive Female Model I I 1 1 not at a l l attractive 3 5 6 very attractive 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items