Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Interpersonal differences in sociotropic and autonomous dysthymic subtypes Bieling, Peter J. 1993

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

Item Metadata


831-ubc_1993_fall_bieling_peter.pdf [ 4.77MB ]
JSON: 831-1.0086200.json
JSON-LD: 831-1.0086200-ld.json
RDF/XML (Pretty): 831-1.0086200-rdf.xml
RDF/JSON: 831-1.0086200-rdf.json
Turtle: 831-1.0086200-turtle.txt
N-Triples: 831-1.0086200-rdf-ntriples.txt
Original Record: 831-1.0086200-source.json
Full Text

Full Text

INTERPERSONAL DIFFERENCES IN SOCIOTROPIC AND AUTONOMOUSDYSTHYMIC SUBTYPESbyPETER J. BIELINGB.Sc. (Hons.) University of Victoria, 1990A THESIS SUBMITTED IN PARTIAL FULFILMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF ARTSINTHE FACULTY OF GRADUATE STUDIES(Department of Psychology)We accept this thesis as conformingto the equired standardThe University of British ColumbiaAugust 1993©Peter J. Bieling, 1993In presenting this thesis in partial fulfilment of therequirements for an advanced degree at the University of BritishColumbia, I agree that the Library shall make it freelyavailable for reference and study. I further agree thatpermission for extensive copying of this thesis for scholarlypurposes may be granted by the head of my department or by hisor her representatives. It is understood that copying orpublication of this thesis for financial gain shall not beallowed without my written permission.Department of PsychologyThe University of British ColumbiaVancouver, CanadaAugust 13, 1993AbstractThe current study reviews theoretical and research evidencewhich links the concepts of sociotropy and autonomy todepression. The concepts of sociotropy and autonomy have beenimplicated as relevant factors in pre-disposing individuals todepression and as influencing the experience of depression;however, important postulated interpersonal differences have notbeen examined empirically. The current study explorestheoretical interpersonal differences in sociotropic orautonomous dysthymic women. The motivational goals,interpersonal concerns, and other perceptions of sociotropic andautonomous individuals were assessed after an interpersonalinteraction. Subjects who were found to be dysthymic anddisplayed excessive sociotropy or autonomy were asked toparticipate in a laboratory task with a confederate who actedeither in a controlling or passive manner. Sociotropic subjectswere more motivated by interpersonal goals in the interactionthan autonomous subjects. Sociotropes were more dependent ontheir partners and felt that they had to rely on them to agreater extent for support, help, and advice; they alsoattempted to please their partner and were concerned about theirpartner's evaluation of them. These individuals felt theirpartner had evaluated them positively, was dependable, and waspleased with them. On the other hand, autonomous individualsfelt that they were being intruded upon by their partner duringthe interaction, that their partner had acted in a controllingiiimanner, and that their partner had perfectionistic standards.Perceptions of others was independent of actual behaviour ofinteraction partners. Moreover, sociotropic subjects werebetter liked by confederates than were autonomous subjects andconfederates rated sociotropics as more likeable. These resultsindicate that sociotropic individuals are more interpersonallyoriented than autonomous individuals and that these individualsnot only perceive others differently, but are perceiveddifferently by others. Implications for interpersonal models ofdepression are discussed and the possible impact of thesedifferences on the experience of depression are examined.ivTable of ContentsAbstract^Table of Contents^ ivList of Tables viAcknowledgement^ viiIntroduction 1Overview^ 1Theoretical Perspectives^ 2Measurement of Dependency/Autonomy Constructs^ 6Relationship of Dependency/Autonomy to Depression^ 7Diathesis-Stress Models of Depression andDependency/Autonomy^ 12Relationship of Dependency/Autonomy to otherPsychological Constructs 23Interpersonal Aspects of Sociotropy andAutonomy: Theoretical Perspectives^ 35Examining Interpersonal Differences inSociotropic and Autonomous Individuals 40Overview of Method^ 46Subjects 47Confederates^ 49Manipulation Checks^ 49Procedure^ 49Measures 50Classification Variables^ 50Dependent Measures^ 52Results^ 57Classification Variables^ 57Dependent Variables^ 58Discussion^ 62Overview 62Experimental Manipulation^ 63Hypotheses^ 63Future Research Questions^ 68Theoretical Implications 72Implications for Psychotherapy^ 73Limitations of the Current Study 75References^ 77Footnotes 90Appendix A^ 91Appendix B 95Appendix C^ 97Appendix D 98Appendix E^ 100Appendix F 102Appendix G^ 103Appendix H 104Appendix I^ 112List of Tablesv iTable 1. BDI Means and Standard Deviations in Sociotropicand Autonomous Groups across Passive andControlling Confederates^ 86Table 2. Warmth and Script Deviations Means and StandardDeviations in Sociotropic and Autonomous Groupsacross Passive and Controlling Confederates^ 87Table 3. Dependent Variables Means and StandardDeviations across Sociotropic andAutonomous Groups^ 88Table 4. Dependent Variables Means and StandardDeviations across Controlling and PassiveConfederate Roles^ 89viiAcknowledgementI would like to thank Lynn Alden for her guidance andsupport throughout this project. Her expertise and objectiveinsights were invaluable to me. I would like to acknowledgeDmitri Papageorgis and Jerry Wiggins for their feedback andhelpful comments.I would like to express my appreciation to Marsh Lopez-Deeand Jennifer Gill for their tireless work at odd hours. Thanksalso to Ken Meleshko for his cooperation and "seasonedexperimenter" wisdom.Thanks are also due to family who were a source ofunconditional support. I am also indebted to all those whooffered me their friendship and encouragement over countlesscups of coffee.1Overview Depression is one of the most frequently occurringpsychological disorders, with a lifetime prevalence of between3.7-6.7 percent in the general population (Robins, Helzer,Weissman, Orvaschel, Gruenberg, Burke, & Regier, 1984). Over aperiod of four years 6.7% of a sample of college students werediagnosed with depression, accounting for 80% of all psychiatricdiagnoses made in the sample (Rimmer, Halikas, & Schuckit,1982). The financial cost of depression in the United Statesalone was estimated to be 16.3 billion dollars per year(Stoudemire Frank, Hedemark, Kamlet, & Blazer, 1986). Moreimportantly there are grave human costs as well. Approximately1 out of every 200 depressed persons commit suicide. (Minkoff,Bergman, Beck, & Beck, 1973).The enormity of the problem has caused a great deal ofinterest and sparked much research. Many attempts have beenmade to link several personality factors with the experience ofdepression. The breadth and diversity of this research has madeit difficult to reach any kind of consensus, but one importanttheme does seem to emerge. The commonality of much of thiswork lies in its themes of excessive autonomy and excessivedependency and their consequences for depression (Pilkonis,1988). These themes and their connection to depression haveemerged from a variety of theoretical perspectives includingthat of developmental, personality, psychodynamic, and cognitivepsychology.2I will begin by reviewing the evolution of this excessivedependency/autonomy distinction and how it has been linked todepression. I will discuss how these issues are addressed bythe variety of psychological perspectives from which they havearisen, the research that has been generated by thisdistinction, and then turn to questions that have been leftunanswered to this point.Theoretical Perspectives Researchers from a broad variety of psychological traditionsbelieve that two basic themes, excessive autonomy and excessivedependency, are etiologically related to the experience ofdepression (Arieti, & Bemporad, 1980; Beck 1983; Beck, Epstein,& Harrison 1983; Blatt, 1974; Blatt, D'Afflitti, & Quinlan,1976; Blatt, Quinlan, Chevron, McDonald, & Zuroff, 1982; Bowlby,1969, 1977, 1980). This distinction has been made both in theform taken by the depressive episode (Blatt, 1974; Blatt et al.,1976; Blatt et al., 1982) and when examining depressionvulnerability factors in the form of a predisposing personalitysubtype (Arieti & Bemporad, 1980) or as a primary schema (Beck,1983). I shall now turn to an examination of theautonomy/dependency distinction as made by workers fromdifferent psychological perspectives.From a developmental perspective, Bowlby (1969, 1977, 1980)discussed "compulsively self-reliant" and "anxiously attached"individuals and how such individuals are prone to depression.3Bowlby believed that excessive self reliance was a defensivereaction to early frustrated, unfulfilled attachment needs andthat anxious attachment was due to unreliable caregiver responsethat results in an extreme desire for interpersonal contact andfear that this will be withdrawn. These different patterns ofearly attachment behaviour are thought to form internal "models"and expectations regarding how other people behave. Suchdysfunctional working models are hypothesized to be carriedthrough until adulthood and thus predispose an individual to theexperience of depression.In the experience of adult depression, thedependency/autonomy distinction was first made by Blatt in 1974.The first type, revolving around excessive dependency, he termedanaclitic depression. It is characterized by feelings ofhelplessness, weakness, and depletion. The second type, basedon excessive autonomy, he termed introjective depression. It isthought to be characterized by feelings of worthlessness, guilt,and a sense of having failed to live up to expectations orstandards. Blatt believed that depression was caused byimpairments in the development of object representationparticularly as it regards relationships to attachment figures.Anaclitic depression is seen as originating in the oral stage ofpsychosexual development whereas introjective depression isbelieved to result from an overly critical superego.More recently it has been hypothesized that the presence ofexcessive autonomy or dependency can also be vulnerability4factors in normal individuals who may be predisposed toexperiencing depression (Blatt et al., 1976; Blatt et al.,1982). Based on research findings which will be discussed laterin greater detail, Blatt et al. (1976) termed these factorsdependency and self-criticism, with dependency involved inanaclitic depression and self-criticism involved in introjectivedepression.Arieti and Bemporad (1980) made very similar distinctionsbased on an examination of two decades of psychotherapy withapproximately 40 depressed patients. They point out thatdepression results when an individual is "...relying to adangerous extent on external support to maintain self-esteem aswell as avenues of meaning and gratification" (p. 1360).^Theydistinguish between two types of excessive reliance, "dominantother" and "dominant goal", that can predispose one todepression. Dominant other corresponds to excessive dependencyand is described as a pattern of relating to others that beginsin childhood in which the person does not seek independentrewards but receives these through some significant other(s).Such persons are described as clingy, passive, manipulative, andanger avoidant (Arieti & Bemporad, 1980).^Dominant goalcorresponds to excessive autonomy. This pattern is alsoestablished in childhood when achievement is rewarded and theattainment of some lofty goal comes to be seen as the only wayto gain support and acceptance. Achievement of the goal hasbroad and unjustifiable meaning attached to it, and these5individuals are seclusive, arrogant, and obsessive (Arieti &Bemporad, 1980). These theoreticians argue that these patterns,established in early childhood, shape and guide a person'sexperience. Individuals thus interpret the world from theirunique perspective and their cognitive rigidity results in adifferential experience of life events. A negative event in therelationship/achievement domains interacts with the person'scognitive structure such that events that are congruent withthat person's pattern (dominant goal or dominant other) willhave more deleterious consequences than non-congruent negativeevents (Arieti & Bemporad, 1980).From a cognitive perspective, Beck (1983) describespersonality "modes" which he calls Sociotropy and Autonomy.Sociotropy is described as "the person's investment in positiveinterchange with others" (p. 272). Such individuals aredependent on social feedback for gratification and support.Autonomy is described as "the person's investment in preservingand increasing his independence, mobility, and personal rights"(p. 272). Such a person derives gratification from directinghis own activities and attaining meaningful goals. Beck writesthat individuals are very much capable of shifting from one modeto another, but that people who become depressed show apreponderance of one of the two clusters. He outlines symptomclusters which would be associated more with one mode or theother, outlines events that would precipitate depression in onetype or the other, and examines specific therapeutic foci forboth modes. We will return to examine some of these issuesafter reviewing the research that the dependency/autonomydistinction has spawned.Theoretical propositions regarding the role of excessivedependency or excessive autonomy in depression have sparked anumber of attempts to measure these constructs and examine howmeaningful they are in the experience of depression. Thisresearch can be divided into several broad categories: (1)assessment of the constructs and development of appropriatemeasures, (2) attempts to link the constructs to the experienceof depression, (3) attempts to examine vulnerability todepression by examining the effects of specific life events onindividuals with excessive dependency or autonomy, and (4)attempts to describe individual differences between depressedindividuals with excessive dependency or autonomy. We shallturn first to formal assessment of the dependency/autonomydistinction.Measurement of the Dependency/Autonomy ConstructsThere have been three major attempts to measure excessiveautonomy and excessive dependency. The first was made from apsychodynamic point of view by Blatt (Blatt et al., 1976; Blattet al., 1982) and the second from a cognitive view by Beck andhis colleagues (Beck, Epstein, Harrison, & Emery, 1983). Themost recent measure was developed by Robins and his associates(Robins, Ladd, & Luten, 1990).67Blatt and his colleagues (1976) set out to assess dependencyand autonomy by examining "a wide variety of experiences thathave been described as related to depression and to test whetherthese different experiences in non-clinical subjects appear inconsistent clusters that have continuity with observations ofclinical depression" (p. 384). Their scale, the DepressiveExperiences Questionnaire or DEQ, was rationally constructed andconsists of 66 items concerning distorted or depreciated senseof self and others, dependency, helplessness, egocentricity,fear of loss, ambivalence, difficulty in dealing with anger,self-blame, guilt, loss of autonomy, and distortions in familyrelations (Blatt et al., 1976).From a factor analysis of the scale, three factors emerged.The first was called dependency and concerned fears ofabandonment, loneliness, and helplessness. The second wascalled self-criticism and concerned feelings of guilt,hopelessness, dissatisfaction, insecurity, and a sense of havingfailed to meet expectations. The third factor was calledefficacy and concerned one's perceived resources and capacities.Although there were significant gender differences in responsesthese three factors, dependency, self-criticism, and efficacy,emerged in both the male and female sample. Test-retestreliabilities for the dependency factor range from .89-.81 andfor the self-criticism factors .83-.75 (Nietzel & Harris, 1990).Internal consistency (coefficient alpha) for dependency is .81and for self-criticism .80 (Blatt et al., 1982). To derive an8individual's dependency or self-critical score, transformationsto z-scores are made and these scores are then multiplied byfactor weights, which vary across gender. There are a number ofproblems with the DEQ and they fall within three areas.Firstly, there is some debate concerning how best to scoreresponses on the DEO for self-criticism and dependency. Anumber of variations are used (Blatt et al., 1982; Chevron,Quinlan, & Blatt, 1979; Steele, 1978). More recently a newscoring system has been devised (Welkowitz, Lish, & Bond, 1984)but it does not seem to be in common use (Riley & McCranie,1988). More fundamentally, there are concerns about the scale'svalidity, that is, whether it measures a predisposition todepression or different presentations of depression (Robins,1991). Finally, there is some doubt about whether the self-critical factor truly addresses excessive autonomy or whether itis more related to "depressive realism" (Nietzel & Harris,1990). The DEQ is the oldest and most established measure ofdependency/autonomy and as such has been used most frequently inresearch, but it appears that significant problems exist in itsformulation and psychometric properties.A more recent attempt to measure the constructs of excessiveautonomy and excessive dependency comes in the from of theSociotropy and Autonomy Scale (SAS) developed by Beck and hisassociates, from Beck's (1983) theoretical work. Items toassess the constructs were drawn from patients self-reports andclinical material (Beck, Epstein, Harrison, & Emery, 1983). The9scale was factor analyzed and yielded two factors correspondingto sociotropy and autonomy. These factors were further analyzedand the best solution for each scale indicated the presence ofthree factors in each sub-scale. Factors for the autonomy sub-scale were: (1) individualistic or autonomous achievement, (2)mobility/freedom from control by others, (3) preference forsolitude. Factors for the sociotropic sub-scale were: (1)concern about disapproval, (2) attachment/concern aboutseparation, (3) pleasing others. The internal consistency(coefficient alpha) of the sociotropy and autonomy scales was.90 and .83 respectively. Within sub-scale factors forsociotropy had consistencies between .80 and .86 and within sub-scale factors for autonomy had consistencies between .60 and.82. The correlations of the sub-scale factors with one anotherranged from .38-.58 for the sociotropy factors, and .20-.31 forthe autonomy factors. Though the scale has been frequently usedsince its inception, many items have been criticised forattempting to measure both the autonomy and sociotropyconstructs at once and the autonomy subscale has been criticizedbecause of the low inter-correlations of its three factors(Robins, 1991; Robins & Block, 1988) and because, unlikesociotropy, autonomy correlates negatively with depression(Robins, Block, & Peselow, 1989). These criticisms have led tothe development of a new measure of sociotropy and autonomy.In an attempt to rectify criticisms of both the DEQ (Blattet al., 1976) and the SAS (Beck, Epstein, Harrison, & Emery,101983), Robins, Ladd and Luten (1990) have developed asociotropy/autonomy scale, the PSI (Personal Style Inventory).These researchers were especially concerned about the problemsof the autonomy sub-scale of the SAS, outlined above, and thecontamination of DEQ items with descriptions of depressivesymptoms (Robins, 1991). The PSI was created with items fromthe SAS and DEQ, as well as the Dysfunctional Attitude Scale(Weissman & Beck, 1978) and the Inventory of InterpersonalProblems (Horowitz, Rosenberg, & Baer, 1988).^The scaleconsists of two 24-item sets, one measuring autonomy and theother sociotropy (see Appendix A).^The internal consistency(coefficient alpha) of the autonomy sub-scale is .86, andsociotropy is .88. Each sub-scale contains three factors. Forsociotropy the factors are: (1) Concern about what others think(alpha .80), (2) Dependency (alpha .72), and (3) Pleasingothers (alpha .83). For autonomy the three factors are: (1)Perfectionism/Self-criticism (alpha .70), (2) Need for control(alpha .70), and (3) Defensive Separation (alpha .80). Thefactors within each sub-scale correlate between .40 to .59 withone another. The sociotropy and autonomy sub-scales correlate.18 with one another and both subscales have very moderatecorrelations (.20 and .27) with a measure of depression (BeckDepression Inventory [BDI]; Beck, Ward, Mendelson, Mock, &Erbaugh, 1961). Thus it appears that the PSI has excellentpsychometric properties and in recent use it was successful in11linking specific symptom clusters to autonomous and sociotropicdepressed individuals (Robins & Luten, 1991).To summarize, three modes to assess the construct ofexcessive autonomy and excessive dependency are currentlyavailable: (1) the DEQ, (2) the SAS, and (3) the PSI. The DEQhas the longest history but is also a somewhat questionablemeasure, both in its validity and in its psychometricproperties. There are concerns about what it measures and howit should be scored. The SAS represents a more recent attemptto measure the constructs but is also somewhat lacking,especially in its measure of autonomy. The latest measuredeveloped, the PSI, seems to have learned the lessons from boththe DEQ and the SAS, insuring that the constructs are measuredwithout contamination (Robins, 1991). As a result, itspsychometric properties are excellent (Robins & Luten, 1990) andit currently seems to be the best measure of excessivedependency and autonomy.The Relationship of Dependency/Autonomy to Depression Now that we have examined efforts to operationalize thedependency/autonomy construct we will turn to research that hasattempted to link it to the experience of depression.A recent meta-analysis completed by Nietzel and Harris(1990) of the relationship of dependency/autonomy to depressionfound a mean effect size for the relationship between autonomyand depression to be r=.31 and for dependency and depression to12be r=.28, both being statistically significant. The data camefrom a variety of studies, using the DEQ, SAS, and lessstandardized measures (eg. Hammen et al, 1985) of dependency andautonomy and a diverse number of depression measures. Therelationship between dependency/autonomy and the most commonmeasure of depression used, the BDI (Beck et al., 1961), was .36and .33 respectively.For dependency, mean effect size for males was r=.25 and forfemales r=.28. For autonomy, mean effect size for males wasr=.38 and for females r=.33. Thus for both genders therelationships seem quite similar (Nietzel & Harris, 1990). Theauthors did not examine whether or not there were genderdifferences in levels of dependency and autonomy although thishas been suggested (Beck, 1983; Blatt, 1974; Chevron, Quinlan, &Blatt, 1978). Chevron et al. (1978) set out to test thisprediction and found females had higher levels of dependency andmales had higher levels of self-criticism. Despite this findinglittle research has looked for gender differences and in somecases no differences have been found (Smith, O'Keeffe, &Jenkins, 1988).Diathesis-Stress Models of Depression and Dependency/AutonomyPerhaps the most concerted and coherent research effortshave come in attempts to test a diathesis/stress model ofdepression using dependency and autonomy as vulnerabilityfactors to life events. A good deal of investigation has13attempted to implicate interpersonal events as leading tosubsequent increased depression in dependent persons andnegative achievement events to depressive symptomology inautonomous individuals.Researchers within the broader area examining life eventsand depression point out that the negative schemata, postulatedby Beck (1967) as instrumental in depression, may becomeactivated when individuals are confronted with personallymeaningful stressful life events (Hammen, 1988; Hammen,Ellicott, Gitlin, & Jamsion, 1989; Hammen, Marks, Mayol, &demayo, 1985; Kuiper, Olinger, & MacDonald, 1988; Kuiper,Olinger, & Air, 1989; Olinger, Kuiper, & Shaw, 1987; Robins,1990; Robins & Block, 1988, 1989; Segal, Shaw, & Vella, 1989).These researchers view Beck's cognitive model as a stress-diathesis one (Robins & Block, 1989). They believe that theeffects of stressful life events will interact with a person'sspecific vulnerability and activate depressogenic schemas. Thisview attempts to take into account the views of researchers whohave emphasized the role of stressful events in depression(Billings & Moos, 1982; Coyne & Gotlib, 1983; Keller, Lavori,Rice, Cryell, & Hirschfeld, 1986). As well, it may partiallyexplain the equivocal findings regarding the relationshipbetween cognitive variables, the negative schemata hypothesizedby Beck, and depression (Coyne, 1990; Coyne & Gotlib, 1983) byexamining what are believed to be the neglected mediators of14this relationship, stressful life events and specificvulnerabilities to them.When this approach is seen in the context ofsociotropy/autonomy, individuals who are very dependent wouldtheoretically be more vulnerable to depression when they suffera perceived loss within their interpersonal relationshipswhereas individuals high in autonomy would be more likely tosuffer from a negative event that keeps them from reaching their.important goals in achievement realms. (Hammen et al., 1989;Hammen et al., 1985; Robins, 1990; Robins & Block, 1988). Wenow examine the research that attempts to link autonomy anddependency and congruent life events to depression in bothstudent and clinical samples.The first attempt to link depression, excessive autonomy,and excessive dependency to life events was that of Hammen andher colleagues (1985). Hammen et al. used an information-processing schema approach, suggesting that persons high independency would be more attuned to and therefore more adverselyaffected by negative interpersonal relationship events whereaspersons high in self-criticism (autonomy) would be moreadversely affected by negative achievement events. The presenceof a self-critical or dependency schema was assessed on thebasis of Markus' (1977) notion that the availability of eventsin memory processes is an indication of the strength of theunderlying schema. Thus individuals where classified as self-critical or dependent based on whether a majority of past15negative events and positive events fell within interpersonalrelationship or achievement realms and the number of incidentsin which they could recall feeling helpless and dependent orcritical of themselves (Hammen et al., 1985). Subjects wereundergraduates who were followed for four months and wereinterviewed by telephone once a month to assess life events andlevel of depression.The results clearly supported the association betweenmeasures of depression and negative interpersonal relationshipevents when they occurred in persons high in dependency. Therewas some support for the relationship between depression andnegative achievement events in persons high in self-criticism.Negative achievement events had much less impact on the symptomsof those high in dependency and negative interpersonalrelationship events had less impact on depressive symptoms ofthose high in self-criticism. Hammen et al. concluded that theresults lent support to the information-processing view. Thatis, individuals have in existence schemas which make some typesof events more meaningful, salient, and more easily recalled.These dependent or self-critical schemas are thought to bestable patterns mediating the relationship between life-eventsand depression (Hammen et al., 1985). There were a number oflimitations within this study, the primary one being the way inwhich people were classified as dependent or self-critical. Themethod for group assignment was previously untried and did not16involve a standardized measure. Also, a non-depressed collegesample was used.Other researchers have also examined the hypothesis thatlife events in the interpersonal relationship or achievementdomains can specifically affect individuals with excessivedependency or autonomy and result in depression. Robins andBlock (1988) call this an interactional diathesis stress model.They view sociotropy and autonomy as personalitycharacteristics, choosing not to describe them as informationprocessing schemas as does Hammen et al. (1985, 1989). Robinsand Block (1988) studied recent life events, level ofdepression, and, using the SAS, levels of sociotropy andautonomy in undergraduates. In this study, level of depressionwas correlated with sociotropy, but not autonomy scores. Aswell, the interaction of interpersonal events and sociotropy wasthe best predictor of level of depression while the interactionbetween autonomy and achievement events had no predictive power.The authors speculate that the lack of an interaction betweenexcessive autonomy and achievement events may be due tomeasurement problems involving the SAS, discussed earlier. Animportant limitation of this study was its cross-sectionaldesign.In a laboratory analogue design, Zuroff and Mongrain(1987), tested whether a simulated achievement or interpersonalevent would have a specific impact on self-critical or dependentindividuals. Subjects were female undergraduates selected for17scoring in the highest 30% of either the self-critical ordependent factors of the DEQ. The procedure involved listeningto audiotapes of a romantic rejection or being told that one hadnot been accepted for graduate school, followed by an assessmentof introjective and anaclitic depression and a measure ofaffect, the Multiple Affect Adjective Checklist (MAACL](Zuckerman & Lubin, 1965). It was found that for dependentsubjects, the rejection resulted in more congruent (anaclitic)depression. For autonomous subjects, however, there was nodifference in the degree of introjective depression caused bythe failure or the rejection, although the failure episodecaused more introjective depression across all groups than didthe rejection episode. One limitation of this particular studywas the use of hypothetical events which called for subjects toimagine they had experienced the situation. The results,however, supported the specificity hypothesis regarding theinteraction of interpersonal events and sociotropy (dependency)but offered only partial support for the interaction of autonomy(self-criticism) and achievement events.In another study, Zuroff, Igreja, and Mongrain (1990)measured undergraduates on the Dysfunctional Attitude Scale(Weissman & Beck, 1978), the DEQ (Blatt et al., 1976) and theBDI (Beck et al., 1961). They found that dependency predictedanaclitic depression and self-criticism predicted introjectivedepression at a 12-month follow-up. They also found that, forboth dependent and self-critical groups, the worst period of18depression involved interpersonal events. It should be notedthat both of the above studies used the problematic DEQ toclassify dependent and self-critical individuals and that themeasures of anaclitic and introjective depression used weredeveloped with few psychometric considerations.The next step taken in this area of research were attemptsto measure the interaction of specific life events and thesociotropic/autonomous distinction in clinical samples ofdepressed patientsHammen and her associates attempted to extend their previousfindings by removing some of the limitations of their earlierwork (Hammen, Ellicott, Gitlin, & Jamison, 1989). This studyused patients with unipolar and bipolar depression and astandardized measure of sociotropy/autonomy, Beck's SAS.Patients symptom status was examined for six months. It wasfound that onset or exacerbation of symptoms was preceded bysignificantly more negative interpersonal relationship eventsfor sOciotropic individuals and significantly more negativeachievement events in autonomous individuals with unipolardepression. This relationship did not hold for the bipolargroup. As well, unipolar individuals who experienced a largernumber of life events congruent with their Sociotropy/Autonomyclassification were more symptomatic. Again this relationshipdid not hold for bipolar patients. These results indicate thatthe relationship between excessive dependency and autonomy,congruent life events, and depression holds for unipolar19depression and not bipolar disorder. However, Hammen et al.qualify these findings and suggest that their sample sizes mayhave been too small, or six months was an insufficient period oftime for the relationship to be evidenced in bipolar patients.Another methodological problem concerned the classification ofsociotropic and autonomous individuals. As there were noguidelines for distinguishing subtypes, this was done by asimple preponderance score of one sub-scale over another.Despite these weaknesses the results of the work were promising.The next study in this series (Hammen, Ellicott, & Gitlin,1989), sought to predict the course of a depressive episode bypostulating that the worst period of symptoms would occur afternegative life events congruent with the person informationprocessing vulnerability schema. Unipolar depressed patientswere classified as autonomous or dependent with Beck's SAS andwere followed for six months to two years. A more sophisticatedmethod was used to assign subjects to sociotropic or autonomousgroups. Subjects were called autonomous if their z-score on theautonomy scale was positive and their z-score on the sociotropyscale was negative (this was reversed for sociotropy). Thisstudy found that, for autonomous individuals, the number ofnegative achievement events best predicted level of depressivesymptoms and that the number of interpersonal events had nopredictive utility. However, for sociotropic individuals,neither the number of achievement or interpersonal events wasable to predict the level of depressive symptoms. Finally, for20both groups combined, those individuals who remained symptomatichad significantly more schema (autonomous or sociotropic)-congruent stressful events (Hammen, Ellicott, and Gitlin, 1989).Limitations in this study included the small sample size and theheterogeneity of the sample which may have allowed the presenceof concurrent personality disorders, temporal stage of eachpatient's depressive episode, and variations in psychotherapyand pharmacotherapy undergone by those in the sample. Again,however, the results were suggestive that thedependent/autonomous distinction is useful in predictingdepressive reactions to specific stressors.Robins (1990) extended his 1988 work involving theinteractional diathesis-stress model to depressed patients.These individuals were compared with inpatients diagnosed withschizophrenia who were not depressed. For the depressives, theresults showed the expected interaction between number ofnegative interpersonal events and high sociotropy on level ofdepression. The interaction for achievement events and highautonomy was not significant. The hypothesis that nointeractions would take place for the psychiatric control samplewas supported. A second experiment involving college studentsalso found support for an interaction effect of either social orachievement events and sociotropy and autonomy in dysphoricstudents but not in non-dysphoric controls. Across bothexperiments, evidence for the interaction of negativeinterpersonal events and excessive sociotropy was stronger than21evidence for the interaction of autonomy and negativeachievement events.Another study involving the assessment of specific lifeevents and excessively dependent or autonomous subtypes ondepression was carried out by Segal, Shaw, and Vella (1989).Unlike previous studies these researchers used remitteddepressives who were followed for six months. Segal et al seethe sociotropic/autonomous types as depressive personalitystyles, a view similar to that of Robins (1988, 1990). Subjectswere classified as high in need for approval or high inperformance evaluation based on a median split of theirresponses to items on the DAS (Weissman & Beck, 1978) which werepreviously found to represent the two factors of dependency andSelf-criticism (Cane, Olinger, Gotlib, & Kuiper, 1986). It wasfound that for individuals high in performance evaluation,neither interpersonal nor achievement events correlated withlevel of depression or relapse while for individuals high inneed for approval interpersonal events correlated significantlywith level of depression and relapse. These researchers believethat the findings may be accounted for by a lack of propertheoretical elaboration of the autonomous subtype, and suggestthat a new scale should be constructed to measure the dimension(Segal et al., 1989).Hammen and Goodman-Brown (1990) attempted to examine theoccurrence of depression with excessive dependency/autonomy andspecific life events in children aged 8-16 years. The sample22included children of mothers who were normal, or who hadrecurrent episodes of unipolar depression, bipolar disorder, orhad insulin dependent diabetes. Classification of"interpersonally vulnerable" or "achievement vulnerable"children was made by the same method used earlier (Hammen etal., 1985) involving recall of events in which things were goingbadly or things were going well. The results showed thatchildren who became depressed showed a greater number of schema-congruent negative life events than those who did not. Theresults were somewhat stronger for excessive dependency andnegative interpersonal events. Furthermore, the majority ofchildren who became depressed after experiencing schemacongruent events were children of mothers with unipolardepression or bipolar disorder, suggesting either the presenceof a genetic vulnerability to depression or perhaps systematicdifferences in child rearing or levels of stress. Overall,however, most children of these mothers did not becomedepressed. Limitations to this study included theunstandardized assessment of dependent or autonomous schemas anda small sample size that did not allow some importantstatistical comparisons. In general, the results added toprevious positive findings and extended the relationship betweenthe dependent/autonomous distinction, relevant life events, anddepression to a new population, children.As we have seen, in general there is more support for theinteractive effect of interpersonal events and excessive23dependency than there is for achievement events and excessiveautonomy (Hammen et al., 1985; Hammen & Goodman-Brown, 1989;Robins & Block, 1988, Robins, 1990, Segal et al., 1989; Zuroff &Mongrain, 1987, Zuroff et al., 1990).Two possibilities are put forward to explain thesedifferential findings (Robins & Block, 1988; Segal et al.,1989); the first is that only excessive dependency is avulnerability factor for depression and the second is thatmeasurement problems of the autonomy construct account for thedifferences in findings (Robins & Block, 1988; Robins, 1990).The first explanation is not consistent with the results ofHammen, Ellicott, and Gitlin (1989) who found a stronger effectfor the interaction of autonomy and negative achievement eventsas well as those studies that find trends linking depressionwith the interaction of excessive autonomy and negativeachievement events (Hammen, Ellicott, Gitlin, & Jamison, 1989;Robins, 1990). Furthermore, small sample sizes have oftenprecluded statistical analyses of differences between the effectsize of the interpersonal event-sociotropy interaction ascompared to the effect size of the achievement event-autonomyinteraction, obscuring the picture (eg. Hammen & Goodman-Brown,1990). Measurement problems, the second explanation offered toexplain the lack of findings regarding excessive autonomy andachievement events is more consonant with the finding of trendsbut not robust effect sizes. As discussed earlier, measurementand theoretical problems extant both in the SAS used by Hammen24and her associates as well as the DEQ used by Zuroff andcolleagues have been taken up by Robins and his associates andthe application of his new measure may bring clearer results forthe interactive effect of achievement events and excessiveautonomy.Relationship of Dependency/Autonomy to Other Psychological Constructs We now turn to research that has sought to link a variety ofvariables with the presence of excessive autonomy or dependency.Unlike attempts that have sought to link life-events andexcessive dependency/autonomy to depression this research hasnot evolved as clearly. This seems largely to be due to thegreat variety of variables studied, and the fact that often onlyone or two studies examine the same types of variables.Furthermore, the use of different measures of dependency andautonomy complicates inter-study comparisons. Thesedifficulties have, to a large extent, resulted in equivocalfindings that are not followed by further investigation orrefinements of methodology. Thus in many areas much work needsto be done before firm conclusions can be drawn. The researchto date on differences between excessively dependent orautonomous individuals has examined familial background (Blatt,Wein, Chevron, & Quinlan, 1979; McCranie & Bass, 1984),differences in attributional style (Brewin & Furnham, 1987;Brown & Silberschatz, 1989), different symptom presentation25(Robins, Block, & Peselow, 1989; Robins & Luten, 1991), sex roleorientation (Chevron et al., 1978; Zuroff, Moskowitz, Wielgus,Powers, & Franko, 1983), and personality correlates (Pilkonis,1988; Riley & McCranie, 1990).Research on familial differences between persons high inautonomy or dependency has come from the perspective of Blatt(1974) who has emphasized the role of a strict mother figure inthe development of anaclitic (dependent) depression and the roleof both parents as strict and demanding of achievement inintrojective depression. Research using the DEO has found thatself-criticism is negatively correlated with parental evaluationwhile dependency is not (Blatt et al., 1979). Another study byMcCranie and Bass (1984) found that individuals scoring higheron dependency are more likely to rate their mother asemphasizing strict control, being the dominant parent, andexpecting conformity. Individuals scoring high in autonomyviewed both the mother and father as emphasizing strict control,expressing inconsistent affection, and as highlightingachievement and performance. Especially noteworthy to theseresearchers was the emergence of the relationship betweenmaternal strictness and dependency, a relationship predicted byBlatt (McCranie & Bass, 1984). Thus individuals high independency were more likely to report that their parental homesemphasized passive conformity while self-critical individualsreported an emphasis on achievement. The two studies discussedhave common limitations. First, given the cross-sectional26nature of the work it is difficult to know whether child-rearingpatterns were actually different or if the presence of self-criticism or dependency influenced people's perceptions.Second, retrospective reports may have been biased by socialdesirability factors. The results were, however, encouragingfor Blatt's predictions of the etiology of anaclitic andintrojective depression.In light of theories which highlight the importance ofinternal, stable, and global attributions in depressives (eg.Abramson, Seligman, & Teasdale, 1978) investigations ofdifferences in the attributional styles of self-critical(autonomous) and dependent individuals have been conducted. Thefirst study by Brewin and Furnham (1987) predicted, based on thewritings of Blatt, that internal attributions for failure shouldonly be correlated with self-criticism. The basis for thisprediction was Blatt's contention that dependent persons feelhelpless because they believe that circumstances outside oftheir control determine what happens to them; thus theseindividuals should make external attributions. The study used asample of undergraduates who completed the Attributional StyleQuestionnaire (Peterson, Semmel, von Baeyer, Abramson, Metalsky,& Seligman, 1982) a measure of the way in which one makesattributions, the DEQ, and the BDI (Beck et al., 1961). Thisstudy found that level of depression, as well as dependency andself-criticism were positively related to making both internaland global attributions. There were no differences in27attributions made by dependent or self-critical individuals(Brewin & Furnham, 1987). The authors suggested that use of aclinically depressed sample may cause differences to emerge.This was attempted by Brown and Silberschatz (1988). They toofound that self-criticism and dependency were equally correlatedwith internal attributions. Together these results offer littlesupport for Blatt's contention that dependent persons feelhelpless because they believe that outcomes are externallycontrolled.Commonalities between attributes which have been found to besocially desirable in males and females and Blatt's descriptionsof self-critical and dependent sub-types have led researchers toexamine possible gender differences in the two constructs(Chevron et al., 1978; Zuroff et al., 1983). Based on researchwhich finds that competence and assertiveness are seen asdesirable attributes for males and warmth and expressiveness areseen as desirable attributes for females, it was postulated thatself-criticism would be associated with masculinecharacteristics and dependency would be associated with femininecharacteristics (Rosenkrantz, Vogel, Bee, Broverman, &Broverman, 1968). It was found that competency, considered adesirable male trait, was negatively related to dependency inboth males and females. Furthermore, warmth, a desirable femaletrait, was related to dependency in males and unrelated todependency in females. Warmth was negatively related to self-criticism in females. The authors believe that the results28reflect the culture's bias towards viewing masculinecharacteristics as more healthy regardless of a person's gender(Chevron et al., 1978). The unexpected correlations betweencompetency and dependency and warmth and self-criticism make theresults somewhat unclear.A second study examining the relationship ofdependency/self-criticism and sex roles was conducted by Zuroffand associates (1983). Scores on Bern's (1974) Sex RolesInventory and their relationship to dependency/self-criticism inmales and females were examined. He found that males high independency described themselves as low in masculinity and highin femininity, while females high in dependency describedthemselves as low in masculinity. There were no correlationsbetween self-criticism and masculinity/femininity in males orfemales. The two studies together present a somewhat equivocalpicture concerning the relationship of self-criticism and sexroles; however, in both studies dependency was negativelyrelated to masculinity in both males and females. Thus theconstruct of excessive dependency seems to be linked to lowlevels of masculinity while self-criticism seems to beindependent of masculinity/femininity constructs.There has also been an attempt to link different personalityprototypes to excessively dependent or autonomous individuals(Pilkonis, 1988). The prototype methodology uses a panel ofexperienced clinicians to generate descriptions of members of aparticular category. Only those descriptors that occur in a29specified quantity are retained and a second panel of judgessorts these descriptors back into the original categories.Descriptions of dependent and autonomous depressives garneredfrom the literature were categorized by 20 clinicians. Theresults indicated the presence of two prototypes within thedependent subtype: (1) excessive dependency and (2) borderlinefeatures. Within the autonomous subtype, three prototypes werefound: (1) obsessive-compulsive features, (2) defensiveseparation, and (3) lack of interpersonal sensitivity. Theauthor suggests that clinicians organize the information aboutthe global labels of dependent and autonomous into more detailedsubcategories (Pilkonis, 1988). Limitations to this study liemainly in its use of only theoretical material; the prototypesfound would need to be confirmed as valid in an actual sample.However, the relationship of clusters corresponding toobsessive-compulsive and borderline personality disorders inautonomous and dependent subtypes, respectively, seemsespecially interesting and seems to fit with clinicaldescriptions provided both by Blatt (1974) and Beck (1983).Other researchers have attempted to look for differentialrelationships between various measures of psychopathology andexcessive dependency/autonomy. Klein, Harding Taylor, andDickstein (1989) measured depressed female outpatients andnormal controls on the DEQ, the Family History ResearchDiagnostic Criteria (FH-RDC) interview (Andreasen, Endicott,Spitzer, & Winokur, 1977), the Longitudinal Interview Follow-Up30Evaluation (LIFE; Keller, Lavori, Friedman, Nielsen, Endicott,McDonald-Scott, & Andreasen, 1987), the BDI (Beck et al., 1961),the Carroll Rating Scale for depression (CRSD; Carroll,Feinberg, Smouse, Rawson, & Greden, 1981), and the SocialAdjustment Scale (SAS; Weissman & Bothwell, 1976). Patientswere followed-up after a six month period. It was found that,for those who recovered from their depression, scores on self-criticism and autonomy dropped, but not to normal levels.Higher levels of self-criticism were significantly related toloss of interest, irritability, and poorer social functioning atfollow-up, while higher levels of dependency were associatedonly with crying or tearfulness. The authors were surprised atthe lack of significant relationships between self-criticism,dependency, and almost 50 dependent variables used. The lack ofrelationships may be accounted for by the measure used to assessexcessive dependency or autonomy (the DEQ). Most of thevariables that were examined were specifically hypothesized byBlatt (1974) to be related to either introjective or anacliticdepression; it is possible that there are problems with hisarticulations of the constructs. The results are alsoinconsistent with findings that dependent and autonomoussubtypes do indeed evidence varied psychopathology and clinicalfeatures, which we will now examine.Another study assessing the relationship ofpsychopathology to excessive dependency/autonomy comes fromRiley and McCranie (1984). They measured depressed inpatients31on the DEQ, the BDI (Beck et al., 1961), the Beck HopelessnessScale (Beck, Weissman, Lester, & Trexler, 1974) which measuresspecific negative expectancies regarding self and future, theATQ (Hollon & Kendall, 1980) which measures depressivecognitions, the RAS (Rathus, 1973), which measures self reportedassertiveness, the MMPI (Hathaway & McKinley, 1940) and apsychiatric history questionnaire developed by the authors. Formales self-criticism correlated significantly with depression(BDI), hopelessness, and depressive cognitions (ATQ). Forfemales self-criticism correlated significantly with depression(BDI), hopelessness, depressive cognitions, lack ofassertiveness, and the MMPI depression scale (MMPI-D). Formales, dependency had no correlates and for females dependencycorrelated with depression (BDI), hopelessness, depressivecognitions and the MMPI depression scale. The measures werealso factor analyzed with dependency, and lack of assertivenessloading on the first factor and self-criticism, depression(BDI), depressive cognitions, and hopelessness loading on thesecond factor. There were no gender differences in the factorstructures.Scores on the MMPI indicated that self-critical males arelikely to be in acute distress, attempt to exaggerate symptoms,and report more hostility, interpersonal conflict, and psychoticsymptoms. Self-critical females were undifferentiated fromdependent females, with both subtypes reporting more acutedistress, symptom exaggeration, complaints of depression,32hostility/conflict, obsessive worry, suspiciousness, psychoticsymptoms, and introversion. Dependent men had no particularprofile except that they tended neither to exaggerate symptomsnor view themselves as being in significant distress.The authors report that their overall results support thedependent and self-critical constructs (Riley & McCranie, 1984).While results from correlational data were somewhat equivocaland contained gender differences, the factor analysis supportedthe construct validity of self-critical/dependent dimensions;self-criticism was associated with overt cognitive and affectivesigns of depression, and dependency was associated with subtlebehavioural manifestations such as passivity. The failure todistinguish a differential MMPI profile in self-critical anddependent women is consistent with the work of Klein et al.(1988), discussed earlier, which failed to find a clinicaldistinction between excessively autonomous and excessivelydependent depressed women. However, there is no explanation asto why dependency in males had no psychopathology correlates.These studies make it apparent that there are definite sexdifferences in the correlates of self-criticism and dependencyin males and females and that these differences should becarefully considered.Differential psychopathology correlates have also beenexamined by Goldberg, Segal, Vella, and Shaw (1989) who examinedthe relationship of personality disorders to excessive autonomyor dependency. Based on a factor analysis of the Dysfunctional33Attitude Scale (Weissman & Beck, 1978) by Cane, Olinger, Gotlib,and Kuiper (1986) that found two factors believed to correspondwith Beck's (1983) subtypes, depressed inpatients wereclassified as sociotropic or autonomous. The MCMI (Millon,1981), a self-report measure designed to assess personalityfunctioning consistent with Diagnostic and Statistical Manual[DSM-III] (American Psychiatric Association, 1980) was alsoadministered. Autonomous individuals were found to haveelevated scores on the negativism, anxiety, and dysthymiasubscales of the MCMI, while sociotropic individuals were foundto have elevated scores on the avoidant, dependent, anxiety, anddysthymia subscales. Beck's descriptions of the sociotropicsubtype correspond closely to their MCMI profile whichcharacterizes such persons as self-effacing, noncompetitive,constantly seeking reassurance and relying on others forguidance. The MCMI profile of autonomous individuals is lessconsistent but elevations on feelings of being misunderstood andanticipation of failure are consistent with Beck's descriptions.In general, the results offered further evidence for thevalidity of the dependency/autonomy construct.Another recent approach to validate the constructs ofexcessive dependency and autonomy in depression has been todifferentiate symptom clusters that have been postulated by Beck(1983) to occur in one subtype or another (Robins, Block, &Peselow, 1989; Robins & Luten, 1991). The symptoms postulatedby Beck (1983) to coincide with sociotropy included demands for34help, dwelling on loss of gratification, crying, concern aboutsocial attributes, responsivity to reassurance and support,lability of mood, greater reactivity to positive and negativeevents, and reports of sadness. Symptoms postulated to coincidewith autonomy included anhedonia, self-criticism, loss ofinterest in and withdrawal from other people, decreasedprobability of crying, unremitting depressed mood, lowprobability of seeking help, greater pessimism about thepossibility of being helped, attributing difficulties topersonal deficiencies, and concern about inability to function.The first attempt to link these different symptoms to sociotropyor autonomy (Robins et al., 1989) involved depressed inpatientsand outpatients who were classified as sociotropic or autonomouson the SAS and then measured on the BDI (Beck et al., 1961) andHRSD (Endicott, Nee, Cohen, Fleiss, & Sarantakos, 1981), which•emphasizes neurovegetative symptoms. The authors placed thoseitems on the HRSD and BDI that were specifically predicted byBeck (1983) to be related to sociotropy or autonomy intospecific clusters. For sociotropy these were items concerningsad mood, crying, decision-making difficulty, negative bodyimage, somatic concerns, and psychic anxiety; items concerninghopelessness or disinterest in people were removed. Forautonomy the items assessed hopelessness, perceived failure,loss of enjoyment, guilt, punishment, self-dislike, self-reproach, irritability, and guilt; an item concerning crying wasremoved. Sociotropic symptoms were significantly related to35sociotropy and autonomy symptoms were negatively related tosociotropy. Autonomous symptoms were not related to autonomybut sociotropic symptoms were negatively related to autonomy.The authors point out that problems with the autonomy subscalemay have been responsible for lack of findings with thissubtype. The results offered partial support for the excessivedependency/autonomy construct, and another study was conductedwhich used a new measure of sociotropy and autonomy.The next piece of research attempted to correct theshortcomings of the first (Robins & Luten, 1991). As such, anew measure of sociotropy/autonomy, the PSI (Robins et al,1990), was used and a more comprehensive assessment of symptomspostulated by Beck (1983) to be related to one subtype oranother was created. The sample consisted of depressedinpatients who were measured on the PSI and responded to itemsassessing the 19 clinical features hypothesized by Beck (1983)to be more strongly related to sociotropy or autonomy. Thesociotropic clinical feature composite was constituted by thefollowing items: optimism about treatment, response toreassurance, variability of mood, reactivity of mood, andfeelings of relief in regards to hospitalization. For theautonomous clinical features composite items were: loss ofinterest or pleasure, feeling like a failure, self-blame, lossof interest in people, avoidance of people, irritability, andconcern about inability to function.36The results fully supported the authors' hypotheses:sociotropic clinical features were related to sociotropy scoresbut not autonomy scores, and autonomy clinical features wererelated to autonomy scores but not sociotropy scores. Theauthors point out that these results lend good support to thevalidity of the sociotropy/autonomy construct and are in linewith the findings of research linking specific life events tothe experience of depression. They also show that the PSI is apromising measure of the sociotropy/autonomy construct.Interpersonal Aspects of Sociotropv and Autonomy: Theoretical Perspectives Having examined the research to date on the relationship ofexcessive sociotropy/autonomy and depression, what conclusionscan be drawn and what are important questions left unanswered?In their recent meta-analyses of this topic, Nietzel and Harris(1990) concluded that "dependency and achievement/autonomythemes are sufficiently related to depressive experiences...thattherapists should consider this content as a legitimate targetfor intervention with depressed clients" (p. 292). The presentexamination of the literature reinforces this notion. Thus itseems that excessive autonomy or dependency can affectvulnerability to depression, the impact which specific lifeevents have, clinical manifestations of depression, andassociated psychopathology. However, an important question leftunanswered to this point regards interpersonal correlates of37excessive dependency/autonomy (Nietzel & Harris, 1990; Robinsand Luten, 1991). That is, can depressed persons in whomautonomy or dependency dominate be differentiated in theirinterpersonal behaviour and are they perceived differentially byothers? The writings of both Blatt (1974) and Beck (1983)certainly imply that there would be interpersonal differences,both in their interpersonal behaviour and perceptions of others.The importance of interpersonal factors in depression hasbeen emphasized by a number of researchers (Barnett & Gotlib,1988; Coyne & Gotlib, 1983; Marcus & Nardone, 1992; Segal,1988). Depressed individuals are found to display longerresponse latencies in interactions and to emit less socialbehaviours (Libet & Lewinsohn, 1973). As well depressed personsare prone to more displays of self-devaluation, sadness, andhelplessness (Hokanson, Sacco, Blumberg, & Landrum, 1980).Moreover, depressed persons have been found to recall lesssocial evaluative information and to believe that othersappraise their social behaviour negatively (Loewenstein &Hokanson, 1986).Saf ran (1990a, 1990b) has conceptualized the presence of aninterpersonal schema, an abstraction based on "interactions withattachment figures that permits the individual to predictinteractions in a way that increases the probability ofmaintaining relationships with these figures" (Safran, 1990a, p.93). This concept is based on perspectives such as Bowlby's(1969) which postulate a wired-in propensity for social38relationships in infants and an ability to encode pastexperiences in order to maximize the possibility of receivingfuture sustenance from attachment figures. Thus aninterpersonal schema is learned and is most usefully construedas a program for maintaining relatedness.A dysfunctional interpersonal schema can develop if certainways of behaving were adaptive for the infant in securingrewards from attachment figures but are not adaptive forinterpersonal relatedness in the present. In other words, aperson may not attempt to relate to any given individual butrather acts in such a way as to satisfy his internal sense ofwhat is required to maintain relatedness (Safran, 1990a).Furthermore, once such a schema is in place, individuals willboth construe and construct (Strupp & Binder, 1984) their world:thus contrary information may not be incorporated into theinterpersonal schema and most likely will be misinterpreted.The interpersonal schema perspective paralells Blatt's(1974) conceptualization of the development of introjective andanaclitic depression develop. In anaclitic (dependent)depressives parental figures are thought to be unreliable inshowing affection and support, and in introjective (autonomous)depression parental figures provide rewards conditional on thechild's performance. Thus, in anaclitic depression childrenlearn that to satisfy their needs they must continuously demandattention and be dependent. In introjective depression childrenhave learned that they must earn love through achievement. Thus39both types of depression evolve due to an interpersonal schemathat emphasizes either a dependent role or an achievement rolein order to satisfy the individual's internal script forrelating with significant others. Thus such individuals can beseen as having interpersonal schemas that are dysfunctional inthe present but are rooted in early learning experiences thatwere adaptive in childhood. People's dependency or achievementfocus may be seen as an integral part of their relatednessschema, and as such should be manifested in their interpersonalbehaviours.Beck (1983) describes many attributes of the sociotropic orautonomous person that fall within the interpersonal domain.The sociotropic person is fearful of rejection, needy of others,does not take risks, asks for continuous reassurance, is eagerto give up control to others, and obtains pleasure fromreceiving. The autonomous individual is seen as less susceptibleto social feedback, reluctant to give up control to others,oblivious to the effect of his actions on other people, notempathetic, direct, decisive (to the point of being dogmatic orauthoritarian), dislikes being blocked or deterred, and dislikesexternally imposed demands or pressures. To date, no empiricalevidence for such differences exists.Beck theorizes that these attributes are pervasive andhighly salient to others. They are exhibited both in thegeneral social world and in the patient/therapist relationship(Beck, 1983). In fact, these characteristics are thought to40have a such a significant impact in the behaviour of theindividual and subsequently in the patient/therapistrelationship, that clinicians are advised to tailor therapy foreach subtype (Beck, 1983). Thus it is postulated that thetherapist should be aware of the person's status with regard tosociotropy and autonomy and to consider how this might influencethe practice and course of therapy.Autonomous individuals are thought to require acollaborative relationship in which they are allowed to setgoals, with much less focus on a warm, guided relationship.Later on in therapy, the patient's underlying rigidity should beexamined and an internal sense of freedom, not one based onachievement of all goals, should be cultivated. For sociotropicindividuals, therapy should be structured with an emphasis onhelping and guidance with less emphasis on self-determination.Explanations and clarifications by the therapist are thought toelicit a positive response (Beck, 1983). The dominant themes intherapy would appear to be "collaboration" for autonomousindividuals and "guidance" for sociotropic individuals. Thus,attesting to the importance of differences in interpersonalfactors between sociotropic and autonomous persons, Becksuggests the implementation of a differential therapeuticrelationship with each subtype.Examining Interpersonal Differences in Sociotropic andAutonomous Individuals41Given these theoretical descriptions of the way in whichsociotropy and autonomy manifest themselves within the domainsof interpersonal behaviour it would be very enlightening toempirically assess how such individuals view themselvesinterpersonally, how they perceive and react to others, and howthey are viewed by others within the context of a socialinteraction. The confirmation or disconfirmation of postulateddifferences would add considerably to our empirical knowledgebase concerning the sociotropic/autonomous distinction. Thestudy of these interpersonal behaviours will also add validationevidence for scales assessing the constructs of sociotropy andautonomy. Many items on scales such as the PSI focus on theindividuals behaviour with and in reaction to other people. Asyet there have been no studies to examine whether or not suchbehaviours and reactions actually occur in an interpersonalsituation. The PSI was developed and evaluated within acollege population, and whether it has any power in predictingactual interpersonal behaviour in such individuals is animportant empirical and practical question that I will attemptto answer.To the degree that sociotropy and autonomy are self-schema,they should bias the perception of other people. Researchinvolving individuals who are schematic on a given dimension hasshown that perception of others is subject to bias. It has beenshown that schemas operate unconsciously to gather schema-specific information about the target person as opposed to42aschematic information, that judges place greater confidence inassessments of the target person on those traits for which theyare themselves schematic, and that target persons are seen assimilar to the judge on schema-relevant traits (Catrambone &Markus, 1987; Fong and Markus, 1982; Markus et al., 1985).Furthermore, our perceptions of others have been found tosustain our own self conceptions because we seek to gaininformation from others that is consistent with our self-schemas(Swann, 1987; Swann & Read, 1981).To the degree that sociotropy and autonomy are self-schemathat affect social perception and to the degree that theorizedinterpersonal differences exist, sociotropic and autonomousindividuals should have different goals when they are engaged insocial interactions. Theory and research suggests thatsociotropic individuals may be motivated by interpersonalaspects of a social interaction while autonomous individualsshould be more motivated by their task goals. Therefore, thefirst question to be examined in the current study is whetherdifferences in sociotropy and autonomy are related to eitherinterpersonal or task goals in an interaction. Given pasttheory and research, the first hypothesis in the current studyis that the two groups will differ on measures of interpersonaland task goals. That is, sociotropic individuals will describethemselves as motivated by interpersonal goals such as havingthe interaction go smoothly and getting along with theirpartner. Autonomous individuals will describe themselves as43motivated by task goals, such as achieving their own objectivesand aims in the interaction.Second, I will examine the nature of interpersonal concernsof sociotropic and autonomous individuals within an interactivetask. Concerns about interactions with others should reflectinterpersonal issues centering on the underlying factors ofsociotropy and autonomy. Thus, within a social interactionsociotropic individuals would be more concerned with pleasingothers, the evaluation of them by others, and their dependenceon the other person. Similarly one would expect that within thecontext of a social interaction the concerns of autonomousindividuals would centre on issues of defensive-separation fromothers, maintaining control in the interaction as well asperfectionistic expectations for themselves. These concernsshould be pre-eminent across different situations. Thus thesecond hypothesis is that sociotropic individuals will scorehigher on a measure of sociotropic interaction concerns thanautonomous individuals and that autonomous individuals willscore higher on a measure of autonomous interaction concernsthan sociotropic individuals in an interactive task.Another question I wished to address concerns how theinterpersonal behaviour of another person is perceived within asocial interaction and how this might exacerbate dysphoria. Ipostulate that the self-schema of sociotropic and autonomousindividuals may bias the perception of other people ondimensions of sociotropy and autonomy. Thus persons high in44sociotropy may generally view others as more dependable, asbeing more pleased with them, and evaluating them positively.Likewise, persons high in autonomy may view others as beinggenerally overcontrolling, intruding upon them and as havingperfectionistic expectations.Such biased perceptions are likely to interact withsituational factors. Theoretical conceptions of sociotropy andautonomy suggest that such persons' interpretations andbehaviours in interpersonal situations may largely be a functionof the way in which others behave. The writings of Beck (1983)would suggest that if autonomous persons interact with acontrolling or assertive other, they respond with dislike,feelings of anger, intrusion, and over-control. A sociotropicperson on the other hand is likely to respond to such a personin a positive manner, interpreting such controlling behaviour asguiding and helpful, in essence, a "receiving of input".Conversely an autonomous person who interacts with a passive andunassertive other, is likely to interpret this behaviourpositively, given the lack of intrusion on his or her goal-directed behaviour and the presence of self-determination. Asociotropic person is likely to negatively interpret such apassive individual, given that such behaviour may elicitperceptions of lack of concern, lack of support, and dislike bythe other person. Given the importance of this situationaldimension, the current study will involve manipulating thebehaviour of a confederate to reflect either controlling or45passive behaviour in an interaction with a sociotropic orautonomous individuals. Thus the third hypothesis is that bothsituational factors and an individual's classification associotropic or autonomous will affect the perception of anotherperson. In other words, both groups will perceive a controllingpartner as being higher on the autonomy dimensions (controlling,intrusive, perfectionism) and the sociotropy dimensions(dependable, evaluating them positively, pleased with them).However, regardless of the role of the partner, sociotropicindividuals will rate their partner higher on sociotropicdimensions and autonomous individuals will rate their partnerhigher on autonomous dimensions.If sociotropic and autonomous persons perceive othersdifferently sociotropic individuals should rate their partner inan interaction as more likeable than their autonomouscounterparts. Thus the fourth hypothesis to be examined is thatsociotropic individuals will assess their partner as morelikeable than will autonomous individuals.Autonomous and sociotropic persons may also be perceiveddifferently by others within the context of a socialinteraction. Autonomous persons should appear as somewhatdisinterested in others and aggressive in pursuing their owncourse of action. The sociotropic individual should appear asdesirous of any social contact and input from others. Thus, onewould predict that persons high in sociotropy would be seen asmore social and likeable than individuals high in autonomy.46Thus the fifth hypothesis is that, on a measure of liking,sociotropic individuals will be rated as more likeable thantheir autonomous counterparts.Finally, sociotropic and autonomous individuals may showdifferent affective responses to social interactions. Suchchanges may play an important role in the maintenance ofdysphoria and will therefore be assessed in the current study.The sixth hypothesis is that, on the measures of affect, aninteraction between sociotropic and autonomous classificationand partner behaviour will occur. For autonomous individuals,interacting with the controlling partner would result infeelings of resentment and thus decrease positive affect andincrease negative affect. For sociotropic individuals, positiveaffect will increase and negative affect will decrease duringthe interaction with the controlling partner. For the passivepartner, autonomous individuals will experience an increase inpositive affect and a decrease in negative affect, given thatthey are allowed to independently make the decisions andcomplete the task in the way they would most like. Sociotropicindividuals should decrease in positive affect and increase innegative affect in response to a passive individual who does notguide and help them in the task.Overview of MethodsThe current study is a 2 x 2 between groups design. Thefirst factor (Group) is classification as sociotropic or47autonomous. The second factor (Role) is the confederate role,controlling or passive.Individuals were selected for participation based on acriterion score indicating mild dysthymia and either excessivesociotropy or autonomy. Subjects were asked to participate in asimulated "helping" task with another individual (a confederate)acting in the "helper" role. Given the predictions of Beck(1983) about the way in which such people behave with others andtheir postulated reactions, the confederates adopted either a"controlling" role in which they gave input into and influencedtheir partners decisions or adopted a "passive" role in whichthey gave little direction and allowed the subject to arrive attheir own decisions. Following the interaction, subjects wereasked to assess the interpersonal behaviour of the confederateon the factors that underlie sociotropy and the factors thatunderlie autonomy, their concerns about the interaction on thosesociotropy and autonomy factors and their focus: task orinterpersonal, liking of their partner, and the presence ofpositive and negative affect in response to the interaction.The confederates rated the likeability of the subjects in theinteraction.Subjects Subjects in this study were female undergraduates enroled infirst or second year Psychology courses, who volunteered toparticipate. Approximately 1600 individuals were screened and96 participated in the laboratory procedure. Subjects were48selected for participation if they met criteria for dysthymia(BDI > 12) and had scores on the PSI indicating excessivesociotropy or autonomy. Subjects were classified as sociotropicif they obtained a positive z-score on the sociotropy sub-scale,and a negative z-score on autonomy. Subjects were classified asautonomous if they obtained a positive z-score on the autonomyscale and a negative z-score on the sociotropy sub-scale. Thisprocedure is thought to provide the best means by which toclassify individuals and has previously been used successfullyfor this purpose (Hammen, Ellicott, & Gitlin, 1989; Robins,1990).Only women were assessed because there is some basis forbelieving that the sociotropy/autonomy construct has differentconsequences for males and females (Chevron et al., 1978, Riley& McCranie, 1984) Secondly, given the distribution ofpsychology classes, with an almost 2:1 female to male ratio, itis probable that the number of males who are sufficientlydysphoric and either sociotropic or autonomous would have beentoo small to allow for meaningful statistical comparisons.Subjects were recruited through a presentation by theexperimenter describing a study involving social interactionsand the experience of depression. Students were told that thestudy consisted of two parts, the first part involving fillingout of questionnaires and the second part involving theirparticipation in a laboratory task.49Confederates Confederates were two female research assistants, who weretrained to enact either a controlling or a passive script forthe interaction (Appendix H). The assistants pretended to beanother participant in the experiment and were present duringthe presentation of instructions and measures. The confederateswere trained to enact similar levels of warmth and scriptaccuracy in the two roles.Three subjects were excluded based on role deviations of theconfederates and these were replaced by other participants.Manipulation Checks Confederates were monitored on the number of scriptdeviations and warmth by the experimenter to insure that theroles were carried out with a similar level of accuracy andwarmth, across groups and confederates.Procedure Potential subjects who met screening criteria were asked toparticipate in the laboratory procedure at a convenient time.Upon arriving at the laboratory, they were introduced to theconfederate. At this point they completed consent forms, theBDI, and affect measures.Experimental Task. Upon completion of the questionnaires,subjects were told that the purpose of the experiment was tostudy "helping styles" and that we wished to examine differenttypes of helping behaviours (Appendix I). They were informedthat they would be taking turns being a helper for the other50person and that they would be discussing a specific problem.The confederate was always chosen as the initial "helper." Theproblem to be discussed was fitness and exercising with subjectsbeing asked to consider how they might improve their fitnesslevel. Both persons were instructed that the interaction couldproceed in whatever way they wanted with the confederate as thedesignated helper, and that they would be observed by theexperimenter during the interaction. Once the experimenter leftthe room the confederate enacted either the controlling orpassive role for the duration of the interaction. Theexperimenter monitored the behaviour of the confederatethroughout the interaction.Once the task was completed the confederate and subject wereseparated and the subject was asked to complete the affectmeasure, the interaction concerns measure, the perception ofpartner behaviour measure, and the measure of liking. Theconfederates completed the measure of liking for each subject.At the end of the study, subjects were debriefed by beingtold the nature of the study, the true identity of theconfederate, and the purpose of the experimental manipulation.Measures Classification VariablesPersonal Style Inventory. Subjects were assessed for thepresence of sociotropy/autonomy using this measure, developed byRobins et al. (1990).^The scale, illustrated in Appendix A,consists of two 24-item sets, one measuring autonomy and the51other sociotropy. Coefficient alpha for the sociotropy sub-scale is .88 and for the autonomy sub-scale is .86. Test-retestreliabilities for a period of between 5 and 13 weeks was .80 forthe sociotropy sub-scale and .76 for the autonomy sub-scale(Robins & Luten, 1991). Each sub-scale contains three factors.These factors for sociotropy are: (1) Concern what others think(alpha .80), (2) Dependency (alpha .72), (3) Pleasing others(alpha .83). For autonomy the three factors are: (1)Perfectionism/Self-criticism (alpha .70), (2) Need for control(alpha .70), (3) Defensive Separation (alpha .80). The factorswithin sub-scales correlate .40 to .59. The sociotropy andautonomy sub-scales correlate .18 with one another, indicatingthat they are largely independent dimensions. The scale'spsychometric properties and its evolution, discussed earlier, asan attempt to remedy problems with both the SAS (Beck et al.,1983) and the DEQ (Blatt et al., 1976), make it the mostdesirable measure for assessment of sociotropy/autonomy.Beck Depression Inventory. The Beck Depression Inventory (BDI)was originally developed 30 years ago (Beck, Ward, Mendelson,Mock, & Erbaugh, 1961), and remains an efficient and widely usedmeasure of severity of depressive symptoms, appearing in over500 studies (Steer, Beck, & Garrison, 1986). It is a 21-itemscale derived from clinical observations. Items were chosen toassess the intensity of depression (see Appendix B).52Correlations between clinicians' ratings of depression and theBDI have ranged between .60 and .90 (Steer et al., 1986).The BDI is frequently used to screen universityundergraduates for depression. However this practice hasreceived a great deal of criticism (Hammen, 1980; Kendall,Hollon, Beck, Hammen, & Ingram, 1987; Sacco, 1981). Thedifficulty with screening students who initially score above acertain cutoff, and then regarding these persons as "depressed"is that up to 50% will fall below that cutoff when tested againwithin even a few hours (Kendall et al., 1987). It has beenrecommended that individuals who meet a pre-specified criterionfor depression be re-tested on the day of the experiment toensure that there scores are stable (Kendall et al., 1987; Sacco1981). The current investigation selected individuals whoscored 12 or above on the BDI, the criterion for mild dysthymia(Steer et al., 1986). In order to insure stability of thesescores subjects were re-assessed on the day they completed thelaboratory procedures.Dependent Measures Interpersonal and Task Goals in the Interaction. Likert -typeitems to assess interpersonal versus task goals in theinteraction were constructed. Three items which measure thedegree of motivation by task goals (eg. To what extent were youfocused on achieving your goals in the task?) and three itemswhich measure the degree to which subjects were motivated byinterpersonal, relational goals (eg. To what extent did you wish53to maintain an agreeable relationship between you and yourpartner?) were included (Appendix C). Each set of three itemswas summed to yield two scores, interpersonal and task goals.Interaction Concerns. Subjects were asked to assess theirconcerns within the social interaction. Items to assess theseconcerns were taken from the PSI and modified to be situationspecific (Appendix D). Six Likert type items were constructedto address the three factors of sociotropy assessed by the PSI,Other's evaluation (eg. To what extent was it important thatyour partner approved of you?), Dependency (eg. To what extentdid you feel dependent on your partner during the interaction?)and Pleasing others (eg. To what extent did you try to pleaseyour partner?). Internal consistency as measured by the Alphacoefficient for this six-item scale was found to be .74 in thecurrent sample (N=96). Therefore, the six items in this scalewere summed to assess sociotropic concerns in the interaction.To assess autonomy concerns, six Likert-type items wereconstructed to address the three factors of autonomy assessed bythe PSI, Need for control (eg. To what extent did you feelcontrolled by your partner?), Defensive-Separation (eg. To whatextent did you try to keep your partner at a distance?), andPerfectionism (eg. To what extent did you feel that theinteraction didn't measure up to your expectations?)Coefficient Alpha for this six-item scale was found to be .74 in54the current sample (N=96). The six items of this scale weresummed to assess autonomous concerns in the interaction.Perception of Partner Behaviour. Subjects were asked to givetheir perceptions of their partner's behaviour on Likert-typeitems assessing the six factors which underlie the autonomy andsociotropy scales of the Personal Style Inventory (Appendix E).Items to assess these perceptions were modified to be specificto the perception of another person. Six Likert type items wereconstructed to address the three factors of sociotropy assessedby the PSI, Other's evaluation (eg. To what extent did yourpartner think positively of you?), Dependability (eg. To whatextent was your partner supportive of you in the task?) andPleasing others (eg. To what extent was your partner pleasedwith you?). The Alpha coefficient for this six-item scale wasfound to be .74 in the current sample (N=96). Therefore, thesix items of this scale were summed to assess sociotropicperceptions of the partner.To assess autonomous perceptions, six Likert type items wereconstructed to address the three factors of autonomy assessed bythe PSI, Controlling (eg. To what extent did your partner try tocontrol you too much during the task?), Intrusiveness (eg. Towhat extent did your partner intrude on you?), andPerfectionism (eg. To what extent was your partner toperfectionistic?) Coefficient Alpha for this six-item scale wasfound to be .91 in the sample. The six items of this scale were55summed to assess autonomous perceptions of the partner in theinteraction.Likeability of Partner. In order to assess the degree to whichboth confederates and subjects liked their partner, eight itemsused in past research to assess likeability (Howes & Hokanson,1979; Stephens & Hokanson, 1987), were used. Individuals areasked the degree to which they would enjoy engaging in a varietyof activities with the target individual (Appendix F). Thealpha coefficient for this scale was .91 in the current samples(N=96). . Ratings from these eight items are summed to yield alikeability score.This measure allowed the examination of differences in thedegree of liking of the confederate by autonomous andsociotropic individuals and the degree of liking of theautonomous or sociotropic subjects by the confederate.Positive and Negative Affect Scales. The Positive and NegativeAffect Scales (PANAS) were developed by Watson, Clark, andTellegen (1988) in order to efficiently measure positive andnegative affect. Positive Affect assesses the degree ofenthusiasm, activity and activation in an individual. It is astate of high energy and pleasurable feelings. Negative Affectis a dimension of subjective distress and displeasure. Itincludes states of anger, contempt, disgust, guilt, fear, andnervousness (Watson et al., 1988). The PANAS scale was designed56to measure positive and negative affect in a more reliable andvalid way than those scales developed in an informal or ad-hocway.The PANAS includes ten adjectives assessing positive affectand ten adjectives assessing negative affect (Appendix G).Internal consistency (coefficient alpha) for the Positive Affect(PA) scale ranges from .86 to .90 and for the Negative Affect(NA) scale ranges from .84-.87. Factorial validity analysesshow that the scales have two factors which account for between62.8 and 68.7 percent of their total variance. No items in thePA scale loaded on the NA factor or vice versa. Factor loadingsfor each item with their respective PA or NA factors ranged from.52 to .75 and were highly consistent. The positive affectscale correlates -.29 with the HCSL (Derogatis, Lipman, Rickels,Uhlenhuth, & Covi, 1974), a general measure of stress anddysfunction. The negative affect scale correlates .65 with theHCSL. Furthermore, the positive affect scale and negativeaffect scale correlate -.36 and .58 respectively with the BDI.Thus the scales seem reliable and valid.The authors indicate that the PANAS can be a measure ofstate affect or trait affect simply by altering the instructionsgive to subjects (Watson et al., 1988). That is subjects can beasked to what extent they feel this way at this very moment orhave felt this way in the past year, or in general. When usedas a current state measure, positive affect increasessignificantly after positive social events, such as movies or57parties (Clark & Watson, 1988), and exercising (Watson, 1988)and decreases significantly after boredom or a work relatedactivity. Negative affect increases after arguments, dailyhassles (such as missing a bus, losing something), and concernsabout relationships (Clark & Watson, 1988), as well as illness(Watson, 1988). A laboratory interaction has also been found tocause affect changes (McIntyre, Watson, Clark, & Cross, 1991).These findings suggest that the PANAS is sensitive to temporarychanges in affective states.In this study the PANAS was given to subjects before andafter their interaction, thus allowing the measurement ofaffective changes attributable to the interaction.ResultsClassification VariablesLevel of Dysphoria. In order to ensure similar levels ofdysphoria in the selected sociotropic and autonomous groups forboth the controlling and passive confederate roles, a one-way,between groups analysis of variance (ANOVA) was performed. Thefour groups (Sociotropic-Controlling X=15.61, N=22, Sociotropic-Passive X=14.46, N=24 Autonomous-Controlling X=16.27, N=24,Autonomous-Passive X=16.13, N=22) did not differ significantlyon level of depression, 2 > .05 (See Table 1).Manipulation Checks. A 2 (sociotropic vs. autonomous) x 2(controlling vs. passive role) between groups multivariateanalysis of variance (MANOVA) was carried out on manipulation58check measures, warmth, and script deviations. There were nosignificant differences between the groups or roles on warmth ornumber of script deviations, all Fs < 1, R>.05. Means andstandard deviations for these measures are shown in Table 2.This result indicates that the roles were represented with equalaccuracy and that confederates did not differ on the dimensionof warmth in the two roles.Dependent Variables To examine the dependent variables, a 2 x 2 between groupsMANOVA was performed with sociotropic or autonomousclassification as the first factor (group) and confederatebehaviour (role) as the second factor. The dependent variableswere: (1) interpersonal goals (2) task goals (3) sociotropicconcerns (4) autonomous concerns (5) sociotropic partnerperceptions (6) autonomous partner perceptions (7) confederateliking (8) partner liking.To ensure homogeneity of group variances, Bartlett's-Box Fprocedure was employed. Wilks's lambda served as the overallcriterion of significance in the analysis, followed byunivariate tests in the case of overall significance.Overview. A significant overall effect on the eight dependentvariables was found for the Group Factor, R < .001 and anoverall significant effect was found for the Role factor, 2 <.001. There were no significant overall interaction effects.59Interpersonal and Task Goals in the Interaction. On the itemsmeasuring interpersonal goals, univariate tests showed asignificant effect for the Group factor, F(1,92)=17.02, g < .01.The mean of the Sociotropic group was significantly higher onthis measure (Table 3). This result indicates that sociotropicindividuals were more motivated by interpersonal goals thattheir autonomous counterparts.On items measuring task goals, no significant effects werefound for either the Group or Role factor. This indicates thatthe two groups did not differ in the extent to which they weremotivated by task goals and that task goals were invariantacross the two confederate roles.Interaction Concerns. On the items measuring sociotropicconcerns in the interaction, a significant main effect for theGroup factor was found, E(1,92)=9.02, R<.01. The mean on thismeasure was higher in the Sociotropic group (Table 3). Thisindicates that sociotropic individuals were more concerned aboutreceiving a positive evaluation, were more dependent on theirpartner, and had tried harder to please their partner.On items measuring autonomous concerns, a main effect wasfound for the Role factor, F=6.65, R<.05. The mean for thismeasure was higher for the Controlling role, across the twogroups (Table 4). Thus, the controlling partner resulted inboth groups feeling more controlled, and defensive, and alsoresulted in the setting of more perfectionistic standards by60subjects. Additionally, there was a trend toward significancefor the Group factor on the autonomy measure, F=3.13, R<.09.The mean for this measure (Table 3) was higher in the autonomousgroup. Thus there was some indication that regardless of role,autonomous individuals felt more controlled, and defensive andthat they were more perfectionistic than their sociotropiccounterparts.Perception of Partner Behaviour. On the items assessingsociotropic perceptions, a significant main effect was found forthe Group factor, F(1,92)=8.31, R<.01. The mean for thismeasure was significantly elevated in the Sociotropic group(Table 3). This indicates that, regardless of the confederaterole, sociotropic subjects perceived their partner as evaluatingthem more positively, as being pleased with them, and as beingmore dependable.On the items measuring perceptions of autonomy, a maineffect was found for Role, F(1,92)=35.10, R<.001. The mean forthe Controlling role was significantly higher on this measure(Table 4). This indicates that the Controlling confederate wasaccurately seen as acting in a more controlling manner, enactingmore intrusive behaviour, and as having more perfectionisticexpectations. A main effect was also found for the Groupfactor, F(1,92)=4.48, 2<.05. The mean for this measure (Table3) was significantly higher in the autonomous group. Thisresult indicates that, regardless of the confederates'61controlling or passive behaviour, autonomous individuals saw theconfederate as more controlling, and intrusive and as havingmore perfectionistic expectations.Partner Liking. On the measures which assessed degree ofconfederate likeability, a significant main effect emerged forboth Role and Group factors. Univariate tests revealed asignificant effect for the Group factor on confederatelikeability, F(1,92)=7.24, R<.01. The mean for this measure washigher in the sociotropic group (Table 3). This resultindicates that, regardless of the confederate's behaviour,sociotropic individuals liked their interaction partner morethan did autonomous individuals. A significant effect for theRole factor also emerged, F(1,92)=6.30, R<.05. The mean forthis measure was significantly higher in Role 2, the passiverole (Table 4). Thus, the passive confederate was liked betterby both groups than was the controlling confederate.On the measure of subject likeability, univariate testsrevealed a significant effect for the Group factor,F(1,92)=11.68, R<O1. The mean of the sociotropic group (Table3) was significantly higher on this measure. This indicatesthat sociotropic individuals were better liked by theconfederates than their autonomous counterparts.Affect Measures. For the measure assessing Positive Affect a 2(sociotropic vs. autonomous groups) x 2 (controlling vs. passive62confederate) repeated measures MANOVA was carried out with Pre-interaction Positive Affect and Post-interaction positive affectas the dependent measures. No significant effects emerged forbetween- or within-subject factors, indicating that no positiveaffect changes occurred as a result of the interaction.For the measure assessing Negative Affect a 2 (sociotropicvs. autonomous groups) x 2 (controlling vs. passive confederate)repeated measures MANOVA was carried out with Pre-interactionNegative Affect and Post-interaction Negative affect as thedependent measures. No significant effects emerged for betweenor within subject factors, indicating that no negative affectchanges occurred as a result of the interaction.'DISCUSSIONOverviewTo summarize the present findings, sociotropic subjects weremore motivated by interpersonal goals in the interaction thanautonomous subjects, although the two groups did not differ intheir task goals. Sociotropic individuals were more dependenton their partners and felt that they had to rely on them to agreater extent for support, help, and advice. These individualsalso evidenced greater concern about behaving in such a way asto please their partner and were more concerned about theirpartner's evaluation of them than were autonomous subjects.Sociotropic individuals felt their partner had evaluated themmore positively, was more dependable, and was more pleased withthem than did the autonomous individuals. On the other hand,63autonomous individuals felt that they were being intruded uponby their partner during the interaction, that their partner hadacted in a controlling manner, and that their partner hadperfectionistic standards. Finally, sociotropic subjects werebetter liked by confederates than were autonomous subjects andsociotropics rated the confederates as more likeable.Experimental Manipulation Sociotropic and autonomous individuals participating in thetask perceived the confederate as enacting different behavioursin the two roles. The controlling role resulted in subjects ofboth groups rating the confederate as enacting more controllingand intrusive behaviours and as having perfectionisticstandards. Moreover, both groups of subjects tended to feeldefensive and controlled by their partner in that role.Additionally, both groups tended to like the passive partner toa greater extent. Ratings by the experimenter demonstrated thatboth roles were carried out with equal accuracy and levels ofwarmth. Thus it appears that the manipulation of roles wasachieved and that differences were in the planned direction.Hypotheses The first hypothesis regarding interpersonal vs. task goalsduring the interaction was partially supported, with a maineffect for the group factor emerging on interpersonal goals.Sociotropic individuals were more motivated by interpersonalgoals than were autonomous subjects. These individuals were64interested in the interaction proceeding smoothly and in gettingalong well with their partner. This was of much less interestto the autonomous subjects. This is consistent with theory thatpredicts greater motivation to maintain interpersonalrelationships in sociotropic individuals (Beck, 1983).It is less clear why autonomous subjects did not havegreater interest in task goals than sociotropic subjects. It ispossible that this particular task (eg. creating a fitness plan)did not activate the theorized need for achievement that theseindividuals are thought to possess. Alternatively, it maysimply be that autonomous individuals are no more concernedabout achieving their goals than their sociotropic counterparts,though they may be less focused on the interpersonal aspects ofa situation. The latter explanation would certainly requirefurther empirical inquiry through the use of other, perhapssomewhat more achievement oriented, tasks. Such an explanationwould appear to be consistent with research on stress-diathesismodels which finds a greater relationship for interpersonal lossand sociotropy than it does for achievement events and autonomy(Hammen et al., 1985; Hammen & Goodman-Brown, 1989; Robins &Block, 1988, Robins, 1990, Segal et al., 1989; Zuroff &Mongrain, 1987, Zuroff et al., 1990). It may be that autonomousindividuals are not necessarily or exclusively schematic forachievement events, but more concerned with interpersonal themesof power, individuation, and mastery (Blatt & Zuroff, 1992).Perhaps autonomous individuals are schematic for and more prone65to become depressed in response to situations in which theirsense of mastery or individuation is threatened. Such events,though certainly distinct from the interpersonal rejection whichhas been found to negatively impact sociotropic individuals, maynonetheless contain important interpersonal components.These findings may therefore have important implications fordiathesis-stress models of depression, highlighting the need forprecise operational definitions of stressful events and the needfor further elucidation of the factors which underlie theconstrual of interpersonal events as negative. In light of thecurrent findings, it may well be that sociotropic and autonomousindividuals attach different meanings to similar interpersonalevents.The second hypothesis regarding interaction concerns forsociotropic and autonomous groups was.partially confirmed with amain effect for the group factor on sociotropic concerns and atrend toward significance on autonomy concerns. Sociotropicindividuals were more dependent on their partners, relying onthem to a greater extent for support, help, and advice.Additionally, sociotropic individuals were more concerned aboutbehaving in such a way as to please their partner than were theautonomous subjects and were more concerned about whether theyhad been positively evaluated by their partner. There was atrend suggesting that autonomous individuals were more concernedabout being controlled in the interaction, defensively66separating themselves from their partner, and settingperfectionistic standards.A significant main effect for partner role on interactionconcerns also emerged. Not surprisingly, both groups ofsubjects felt more controlled and defensive when the confederatewas enacting the controlling role. This finding indicates thatthe role was perceived differently and accurately by subjects.These results show that to some extent sociotropic andautonomous individuals do have different concerns when they areinteracting with others, and that these differences are in thehypothesized direction. Sociotropic individuals are indeed moredependent on others, try harder to please them, and areconcerned with how they will be evaluated. This finding isconsistent with research findings that dependent, non-depressedwomen are more likely to use compromise in resolving conflictswith boyfriends (Blatt & Zuroff, 1992).The third hypothesis regarding perception of the partnerbehaviour was supported. A main effect was found for both theGroup and Role factors on the measures of partner perception.An overall main effect for role indicated that both sociotropicand autonomous subjects perceived their partner as beingmarkedly different in the two roles. In the controlling rolethe partner was seen as enacting more controlling behaviours, asacting intrusively, and as having high, perfectionisticstandards. This result further indicates that the two roles67were perceived differently and that the partner's behaviour wasconstrued accurately by the subjects.An overall significant effect also emerged for the groupfactor on the perception of partner measures. The partner wasperceived differently by the two groups across the two disparateroles. Regardless of the partner's passive or controllingbehaviour, sociotropic individuals believed that their partnerhad evaluated them more positively, was more pleased with them,and had been more dependable than did the autonomousindividuals. Autonomous subjects felt that the partner hadacted in a controlling manner, had behaved intrusively, and hadperfectionistic standards.Again, despite the salient differences in the situation,these perceptions of the partner were consistent across roles inthe two groups. These significant main effects suggest thatsociotropic and autonomous individuals perceive othersdifferently, as predicted, across different situations. Thoughthese situational differences may be perceived, it appears thatthe perception of another person is nonetheless modified by anindividual's being sociotropic or autonomous.The fourth hypothesis regarding likeability of theconfederate partner was supported. A significant main effectfor the group factor occurred. Regardless of role, sociotropicindividuals judged their partner as more likeable than didautonomous individuals. This finding is consistent with thegreater interpersonal focus of sociotropic individuals, and the68fact that these individuals were more interested in pleasingtheir partner. On the other hand autonomous individuals feltdefensive, that their partner had intruded upon them and thattheir partner was perfectionistic. It is not surprising thenthat the two groups rated their partner's likeabilitydifferently. Again, the likeability of the confederate wasstable across the large, perceived situational difference, andwould suggest that sociotropic individuals respond toindividuals more positively across a variety of situations.There was also a main effect for the role factor on thedependent measures. Regardless of whether individuals weresociotropic or autonomous, the passive partner was rated as morelikeable than the controlling partner. This result againsuggests that the manipulated roles were different and that thisdifference was perceived by subjects.The fifth hypothesis regarding likeability of subjects wassupported. Sociotropic subjects were better liked byconfederates than were autonomous subjects. This finding isconsistent with previous results which found that self-critical(DEQ) females participating in an interaction were judged asless likeable by peer raters (Zuroff et al., 1983). Thisprevious study used non-participant judges to rate likeabilityof the target. It is now clear that this difference inlikeability is also perceived by an actual participant in theinteraction. This result is also interesting in that it appearsthat sociotropic and autonomous individuals may be aware of the69impact they are having on others. As previously discussed,sociotropic individuals felt that their partner was more pleasedwith them and had evaluated them more positively than didautonomous individuals. This perception appears to have beenquite correct. Sociotropic individuals were indeed judged asmore likeable by their partners.The sixth hypothesis regarding affective changes across theinteraction was not supported. Indeed, no significant findingsemerged on the affective measures. It appears that theinteraction did not alter the affective state of the subjects.This is of significance to interpersonal models of depressionwhich postulate that interpersonal interactions result inincreased negative affect in depressed individuals, therebyexacerbating depression (Coyne, 1976). It is possible that thenature of the interaction was not of a quality that such changeswould occur or that the interaction was not of a sufficientduration to effect such changes.Future Research Ouestions The current findings leave some unresolved issues whichmight be profitably addressed by future research. The findingthat sociotropic individuals are more likeable than autonomousindividuals poses several as yet unanswered questions. First,what specifically do autonomous and sociotropic persons do thatmakes them more or less likeable? It would be illuminating toexamine specific behaviours which might result in autonomous70individuals being judged as less likeable than sociotropicpersons. At this point it is unclear what behaviouraldimensions might most efficaciously be assessed and whatinstruments might be appropriate in such research.Second, it has been suggested that the behaviour of bothsociotropic and autonomous individuals eventually results ininterpersonal rejection (Beck, 1983) a view which is consistentwith interpersonal models of depression (Coyne, 1976).^Thecurrent results suggest that autonomous individuals are indeedless liked than sociotropic individuals after a fairly briefinteraction. The process by which these persons might berejected by others appears to be quite salient and immediate.Upon meeting someone for the first time they act in a way thatmakes them less likeable than their sociotropic counterparts.Further, it appears that autonomous individuals are less likelyto try to please their partner and because of their lack ofinterpersonal focus, may be not be distressed by their lack oflikeability.It appears that, upon an initial meeting and subsequentbrief interaction, sociotropic individuals are liked to agreater extent than their autonomous counterparts. The processwhich may ultimately result in the rejection of sociotropicindividuals is not apparent in these results. Such a result mayhave a number of implications. First, it may suggest thatsociotropic individuals may not particularly provokeinterpersonal rejection from others as has been previously71suggested. However, this would not imply that these individualsare not keenly sensitive to rejection when it does occur. Infact, the interpersonal focus that these individuals endorsewould seem to sensitize them to rejection. It may be that theyare no more likely to suffer such a loss, other things beingequal, than any other individual, but that they are acutelyattuned to such rejection when it does occur. However, other asyet untested possibilities exist.It may be that the behaviour of sociotropic individualsleads ultimately to rejection after the passage of time.Sociotropic individuals indicate that they are more dependent onothers and that they try to please them. Behaviour that isinitially affiliative may eventually be perceived as excessivelyneedy and dependent thereby leading to rejection.Third, it may be that sociotropic individuals would be lesslikeable than another group of non-dysphoric individuals, apossibility not examined in the current study. Thussociotropics may only be likeable relative to their autonomouscounterparts, and be more rejected than another comparisongroup. To address this possibility, one would need a third,non-dysthymic comparison group. However, this poses thequestion of which group of individuals would provide the mostmeaningful comparison in terms of likeability. Threepossibilities exist: 1) non-dysthymic sociotropic or autonomousindividuals, 2) non-dysthymic individuals low on both sociotropyand autonomy or 3) non-dysthymics high on both scales. Although72it is unclear what additional comparison group is mostappropriate, the lack of such a comparison group makes anyconclusion speculative. The data do however allow thetheoretically consistent conclusion that autonomous individualsare less liked than sociotropic individuals.Theoretical Implications The present findings have a number of implications fortheoretical views of the excessive dependency and autonomy.Blatt (1974) believed that excessive dependency resulted in ananaclitic depression which was characterized by feelings ofhelplessness and weakness. Introjective depression, caused byexcessive autonomy, was thought to result in excessive concernabout achieving goals and feelings of worthlessness. Arieti andBemporad (1980) created similar constructs, distinguishingbetween excessive reliance on a "dominant other" or "dominantgoal" and postulating that congruent negative life events resultin depression. The present findings suggest that sociotropy mayindeed be related to feelings of helplessness, dependency andexcessive reliance on others. However, autonomy does notnecessarily appear to be related to a greater preoccupation withachievement goals, a finding which is inconsistent with thetheoretical views of Blatt and Arieti and Bemporad. From thepresent findings it appears instead that autonomy is related togreater concerns about intrusion and over-control and that this73construct may be somewhat more interpersonal in nature thanpreviously thought.The present results are quite consistent with Beck'stheoretical views of sociotropy and autonomy. Beck (1983)defined sociotropy as "the person's investment in positiveinterchange with others" (p. 272) and autonomy as "the persons'sinvestment in preserving and increasing his independence,mobility, and personal rights." (p.272). The current findingssupport the view that sociotropics are more concerned withcreating a positive interpersonal interchange and suggest thatautonomous persons are indeed concerned with infringement oftheir freedom and independence in the interpersonal realm.Implications for PsychotherapyOverall, sociotropics appear to be more interpersonallyoriented and more affiliative in interactions. This orientationresults in differential perception of others as well asperceptible differences in behaviour. Sociotropic individualsare better liked and try harder to please others than theirautonomous counterparts. The latter group is less concernedabout their interpersonal impact and perceive others' behaviourin a more negative, defensive fashion. These differences,apparent after a brief laboratory interaction, would most likelybe apparent in a variety of interpersonal settings, quitepossibly including psychotherapy. Consistent with Beck'stheorizing, sociotropes may indeed prefer a warmer, more74interpersonal focus for therapy and may play a very active partin creating such a therapeutic relationship.Autonomous individuals may be less likely to be interestedin the formation of a close interpersonal alliance and may, infact, be acting in such a way as to prevent the establishment ofsuch a relationship. Thus the current results could imply adifferential response to psychotherapy in the two depressivesubtypes. Recent work (Peselow, Robins, Sanfilipo, Block, &Fieve, 1992) has shown that individuals who are high in autonomyshow much greater responsiveness to anti-depressant medicationthan do persons high in sociotropy (74% vs. 39%). Perhapspatients high in sociotropy would be more likely to respondpositively to therapy involving a significant interpersonalcomponent. Such therapy might focus largely on interpersonalissues, including relationships with significant others, whichmay play a very significant role in the depression of thesociotropic individual. However, it is also clear from thecurrent study that autonomous individuals display distinctiveand maladaptive interpersonal behaviours which might be examinedand modified in therapy. Such interpersonal behaviours were notthe target of therapy in the Peselow et al. study; it ispossible that the greater response to pharmacotherapy in theautonomous group was simply by default. That is, given theabsence of a comparative therapy we know only the relativeeffectiveness of pharmacotherapy in each group. Greater gains75might be possible in both groups via the application of othertherapeutic techniques.Limitations of the Current StudyThe current study suggests that there are significantinterpersonal differences between sociotropic and autonomousindividuals and that these differences exist across salientsituational variability. There are however a number of cautionsto be heeded in interpreting the data.First, the design of the study does not allow causalinferences to be made. It may well be that group differencesare the result of an unassessed variable that systematicallydiffers in sociotropic and autonomous individuals. Althoughgroups were equated on the important variable of level ofdepression, they may have differed in other ways that were notmeasured.Second, these results may not generalize to persons withclinical levels of depression. Whether individuals who areclinically depressed and differentiated on sociotropy andautonomy would show similar perception and behaviouraldifferences in interpersonal interactions is an empiricalquestion.Third, these findings may not generalize from the structuredlaboratory situation provided in the current study to dailyencounters. Though fairly strong differences emerged in a briefinteraction, whether these differences would emerge in the daily76interactions of sociotropic and autonomous individuals mayprovide an interesting question for future research.Finally, only one gender was assessed in the current study.Given that sociotropy and autonomy may have differing correlatesin each gender (Chevron et al., 1978; Zuroff et al., 1983), thepresent pattern of findings may not be replicated with malesamples.Despite these limitations, the current results do suggestthat there are important interpersonal differences insociotropic and autonomous dysthymics. Further studies mightexamine whether this pattern of results is replicated in a mixedgender sample of clinically depressed individuals. The mostinteresting question may be what implications theseinterpersonal differences have for the conduct and outcome ofpsychotherapy.77References Abramson, L. Y., Seligman, M. E. P., & Teasdale, J. D. (1978).Learned Helplessness in Humans: Critique and Reformulation.Journal of Abnormal Psychology, 87(1), 49-74.American Psychiatric Association (1980). Diagnostic andStatistical Manual of Mental Disorders (3rd ed.). WashingtonDC: APA.Andreasen, N. C., Endicott, J., Spitzer, R. L., & Winokur, G.(1977). The Family History Method Using Diagnostic Criteria.Archives of General Psychiatry, 34, 1229-1235.Arieti, S., & Bemporad, J. R. (1980). •The PsychologicalOrganization of Depression. American Journal of Psychiatry,137(11), 1360-1365.Barnett, P. A., & Gotlib, I. H. (1988). PsychosocialFunctioning and Depression: Distinguishing Among Antecedents,Concomitant, and Consequences. Psychological Bulletin,104(1), 97-126.Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J.(1961). An Inventory for Measuring Depression. Archives of General Psychiatry, 4, 561-571.Beck, A. T. (1967). Depression: Clinical, Experimental, andTheoretical Assumptions. New York: Harper & Row.Beck, A. T., Weissman, A., Lester, D., & Trexler, L. (1974).Measurement of Pessimism: The Hopelessness Scale. Journal ofConsulting and Clinical Psychology, 42, 861-865.Beck, A. T. (1983). Cognitive Therapy of Depression: NewPerspectives. In P. J. Clayton & J. E. Barnett (Eds.),Treatment of Depression: Old Controversies and New Approaches(pp. 265-290). New York: Raven Press.Beck, A. T.. Epstein, N., & Harrison, R. (1983). Cognitions,Attitudes and Personality Dimensions in Depression. BritishJournal os Cognitive Psychotherapy, 1, 1-16.Beck, A. T., Epstein, N., Harrison R. P., & Emery, G. (1983).Development of the Sociotropy-Autonomy Scale: A measure ofPersonality factors in Psychopathology. UnpublishedManuscript. University of Pennsylvania, Philadelphia.Bern, S. L. (1974). The Measurement of Psychological Androgyny.Journal of Consulting and Clinical Psychology, 22, 155-162.78Billings, A., & Moos, R. (1982). Psychosocial Theory andResearch on Depression: An Integrative Framework and Review.Clinical Psychology Review, 2, 213-237.Blatt, S. J. (1974). Levels of Object Representation inAnaclitic and Introjective Depression. Psychoanalytic Study ofthe Child, 29, 107-157.Blatt, S. J., D'Affliti, J. P., & Quinlan, D. M. (1976).Experiences of Depression in Young Adults. Journal of Abnormal Psychology., 85, 383-389Blatt, S. J., Wein, S. J., Chevron, E., & Quinlan, D. M. (1979).Parental Representations and Depression in Normal YoungAdults. Journal of Abnormal Psychology, 88(4), 388-397.Blatt, S. J., Quinlan, D. M., Chevron, E. S., McDonald, C., &Zuroff, D. (1982). Dependency and Self-Criticism:Psychological Dimensions of Depression. Journal of Consultingand Clinical Psychology., 50(1), 113-124.Blatt, S. J., & Zuroff, D. C. (1992). Interpersonal Relatednessand Self-Definition: Two Prototypes for Depression. Clinical Psychology Review, 12, 527-562.Bowlby, J. (1969). Attachment and Loss: Volume 1. Attachment.New York: Basic Books.Bowlby, J. (1977). The making and Breaking of Affectional Bonds:I. Aetiology and Psychopathology in Light of AttachmentTheory. British Journal of Psychiatry, 130, 201-210.Bowlby, J. (1980). Attachment and Loss: Volume 3. Loss: Sadness and Depression. New York: Basic Books.Brewin, C. R., & Furnham, A. (1987). Dependency, Self-Criticism,and Depressive Attributional Style. British Journal of Clinical Psychology, 26, 225-226.Brown, J. D., & Silberschatz, G. (1989). Dependency, Self-Criticism, and Depressive Attributional Style. Journal of Abnormal Psychology, 98(2), 187-188.Cane, D. B., Olinger, L. J., Gotlib, I. H., & Kuiper, N. A.(1986). Factor Structure of the Dysfunctional Attitude Scalein a Student Population. Journal of Clinical Psychology,42(2), 307-309.Carroll, G. J., Feinberg, M., Smouse, P. E., Rawson, S. G., &Greden, J. F. (1981). The Carroll Rating Scale for Depression:I. Development, Reliability, and Validation. British Journal of Psychiatry, 138, 194-200.79Catrambone, R., & Markus, H. (1987). The Role of Self-Schemas inGoing Beyond the Information Given. Social Cognition, 5(4),49-368.Chevron, E. S., Quinlan, D. M., & Blatt, S. J. (1978). Sex Rolesand Gender Differences in the Experience of Depression.Journal of Abnormal Psychology, 88, 680-683.Clark, L. A., & Watson, D. (1988). Mood and the Mundane:Relations Between Daily Life Events and Self-Reported Mood.Journal of Personality and Social Psychology,  54(2), 296-308.Coyne, J. C. (1976). Toward an Interactional Description ofDepression. Psychiatry, 39, 28-40.Coyne, J. C., & Gotlib, I. H. (1983). The Role of Cognition inDepression: A Critical Appraisal. Psychological Bulletin,4(3), 472-505.Coyne, J. C. (1990). Concepts for Understanding Marriage anDeveloping Techniques of Marital Therapy: Cognition UberAlles? Journal of Family Psychology, 4(2), 185-194.Derogatis, L. R., Lipman, R. S., Rickels, K., Uhlenhuth, E. H.,& Covi, L. (1974). The Hopkins Symptom Checklist (HSCL): ASelf-Report Symptom Inventory. Behavioral Science, 19, 1-15.Endicott, J., Nee, J., Cohen, J., Fleiss, J. L., & Sarantakos,S. (1981). Hamilton Depression Rating Scale Extracted fromSchedule for Affective Disorders and Schizophrenia and SADS-C.Archives of General Psychiatry, 38, 837-844.Fong, G. T., & Markus, H. (1982). Self-Schemas and JudgementsAbout Others. Social Cognition, 1(3), 191-204.Goldberg, J. O., Segal, Z. V., Vella, D. D., & Shaw, B. F.(1989). Depressive Personality: Millon Clinical MultiaxialInventory Profiles of Sociotropic and Autonomous Subtypes.Journal of Personality Disorders, 3(3), 193-198.Hammen, C. (1980). Depression in College Students: Beyond theBeck Depression Inventory. Journal of Consulting and ClinicalPsychology, 4g(1), 126-128.Hammen, C., Marks, T., Mayol, A., & demayo, R. (1985).Depressive Self-Schemas, Life-Stress, and Vulnerability toDepression. Journal of Abnormal Psychology, 94(3), 308-319.Hammen, C. (1988). Depression and Cognitions about PersonalStressful Life Events. In L. B. Alloy (ed.), Cognitive Processes in Depression (pp.77-108). New York: GuildfordPress.80Hammen, C., Ellicott, A., & Gitlin, M. (1989). Vulnerability toSpecific Life Events and Prediction of Course of Disorder inUnipolar Depressed Patients. Canadian Journal of Behavioural Science, 21(4), 377-388.Hammen, C., Ellicott, A., Gitlin, M., & Jamison, K. R. (1989).Sociotropy/Autonomy and Vulnerability to Specific Life Eventsin Patients with Unipolar Depression or Bipolar Disorders.Journal of Abnormal Psychology, 98(2), 154-160.Hammen, C. & Goodman-Brown, T. (1990). Self-Schemas andVulnerability to Specific-Life Stress in Children at Risk forDepression. Cognitive Therapy and Research, 14(2), 215-227.Hathaway, S. R., & McKinley, J. C. (1940). A MultiphasicPersonality Inventory (Minnesota): Construction of theSchedule. Journal of Psychology, 10, 249-254.Hokanson, J. E., Sacco, W. P., Blumberg, S. R., & Landrum, G. C.(1980). Interpersonal Behaviour of Depressive Individuals in aMixed-Motive Game. Journal of Abnormal Psychology, 89(3), 320-332.Hollon, S., & Kendall, P. C. (1980). Cognitive Self-Statementsin Depression: Development of an Automatic ThoughtsQuestionnaire. Cognitive Therapy and Research, 4, 383-396.Horowitz, L. M., Rosenberg, S. E., Baer, B. A., Ureno, G., &Villasenor, V. S., (1988). Inventory of InterpersonalProblems: Psychometric Properties and Clinical Applications.Journal of Consulting and Clinical Psychology, 56(6), 885-892.Howes, M. J., & Hokanson, J. E. (1979). Conversational andSocial Responses to Depressive Interpersonal Behaviour.Journal of Abnormal Psychology, 88, 625-634.Keller, M. B., Lavori, P. W., Rice, J., Coyell, W., &Hirschfeld, R. M. A. (1986). The Persistent Risk of Chronicityin recurrent Episodes of Non-Bipolar Major DepressiveDisorder: A Prospective Follow-Up. American Journal of Psychiatry, 143, 24-28.Keller, M. B., Lavori, P. W., Friedman, B., Nielsen, E.,Endicott, J., McDonald-Scott, P., & Andreasen, N. C. (1987).The Longitudinal Interval Follow-Up Evaluation: AComprehensive Method of Assessing Outcome in LongitudinalProspective Studies. Archives of General Psychiatry, 44, 540-548.Kendall, P. C., Hollon, S. D., Beck, A. T., Hammen, C., &Ingram, R. E. (1987). Issues and Recommendations Regarding the81Use of the Beck Depression Inventory. Cognitive Therapy andResearch, 11(3), 289-299.Klein, D. N., Harding, K., Taylor, E. B., & Dickstein, S.(1988). Dependency and Self-Criticism in Depression:Evaluation in a Clinical Population. Journal of Abnormal Psychology, 97(4), 399-404.Kuiper, N. A., Olinger, L. J., & MacDonald, M. R. (1988).Vulnerability and Episodic Cognitions in a Self-WorthContingency Model of Depression. In L. B. Alloy (ed.),Cognitive Processes in Depression (pp.289-309). New York:Guildford Press.Kuiper, N. A., Olinger, L. J., & Air, P. A. (1989). Stressfulevents, Dysfunctional Attitudes, Coping Styles, andDepression. Personality and Individual Differences, 10(2),229-237.Leary, T. (1957). Interpersonal Diagnosis of Personality. NewYork: Ronald Press.Libet, J., & Lewinsohn (1973). The Concept of Social Skill WithSpecial Reference to the Behaviour of Depressed Persons.Journal of Consulting and Clinical Psychology,  40, 304-312.Loewenstein, D. A., & Hokanson, J. E. (1986). The Processing ofSocial Information by Mildly and Moderately Dysphoric CollegeStudents. Cognitive Therapy and Research, 10(4), 447-460.Marcus, D. K. & Nardone, M. E. (1992). Depression andInterpersonal Rejection. Clinical Psychology Review, 12, 433-449.Markus, H. (1977). Self-Schemata and Processing of InformationAbout the Self. Journal of Personality and Social Psychology,35(1), 63-78.Markus, H., Smith, J., & Moreland, R. L. (1985). The Role of theSelf-Concept in the Perception of Others. Journal ofPersonality and Social Psychology, 49(6), 1494-1512.McCranie, E. C., & Bass, J. D. (1984). Childhood FamilyAntecedents of Dependency and Self-Criticism: Implicationsfor Depression. Journal of Abnormal Psychology, 93(1), 3-8.McIntyre, C. W., Watson, D., Clark, L. A., & Cross, S. R.(1991). The Effect of Induced Social Interaction on Positiveand Negative Affect. Bulletin of the Psychonomic Society,29(1), 67-70.82Millon, T. (1981). Millon Clinical Multiaxial Inventory Manual (3rd ed.). Minneapolis, MN: National Computer Systems.Minkoff, K., Bergman, E., Beck, A. T., & Beck, R. (1973).Hopelessness, Depression, and Attempted Suicide. AmericanJournal of Psychiatry, 1301 455-459.Nietzel, M. T., & Harris, M. J. (1990). Relationship ofDependency and Achievement/Autonomy to Depression. Clinical Psychology Review, 10, 279-297.Olinger, L. J., Kuiper, N. A., & Shaw, B. F. (1987).Dysfunctional Attitudes and Stressful Life Events: AnInteractive Model of Depression. Cognitive Therapy andResearch, 11(1), 25-40.Peselow, E. D., Robins, C. J., Sanfilipo, M. P., Block, P., &Fieve, R. R. (1992). Sociotropy and Autonomy: Relationship toAntidepressant Drug Treatment Response to and Endogenous-Nonendogenous Dichotomy. Journal of Abnormal Psychology,101(3), 479-486.Peterson, C., Semmel, A., von Baeyer, C., Abramson, L. Y.,Metalsky, G. I., & Seligman, M. E. P. (1982). TheAttributional Style Questionnaire. Cognitive Therapy andResearch, 6, 287-300.Pilkonis, P. (1988). Personality Prototypes Among Depressives:Themes of Dependency and Autonomy. Journal of PersonalityDisorders, 2(2), 144-152.Rathus, S. A. (1973). A 30-item Schedule for Assessing AssertiveBehaviour. Behaviour Therapy, 4, 398-406.Riley, W. T., & McCranie, E. W. (1990). The DepressiveExperiences Questionnaire: Validity and PsychologicalCorrelates in a Clinical Sample. Journal of PersonalityAssessment, 54(3&4), 523-533.Rimer, J., Halikas, J., & Schuckit, M., (1982). Prevalence andIncidence of Psychiatric Illness in College Students: A FourYear Prospective Study. Journal of American College Health,30(5), 207-211.Robins, C. J., & Block, P. (1988). Personal Vulnerability, LifeEvents, and Depressive Symptoms: A Test of the SpecificInteractional Model. Journal of Personality and Social Psychology, 54(5), 847-852.Robins, C. J., & Block, P. (1989). Cognitive Theories ofDepression Viewed from a Diathesis-Stress Perspective:83Evaluations of the Models of Beck and of Abramson, Seligman,and Teasdale. Cognitive Therapy and Research, 13(4), 297-313.Robins, C. J., Block, P., & Peselow, E. D. (1989). Relations ofSociotropic and Autonomous Personality Characteristics toSpecific Symptoms in Depressed Patients. Journal of Abnormal Psychology, 98(1), 86-88.Robins, C. J. (1990). Congruence of Personality and LifeEvents in Depression. Journal of Abnormal Psychology,99(4), 393-397.Robins, C. J., Ladd, J., & Luten, A. G. (1990). Development ofthe Personal Style Inventory: A Questionnaire to AssessSociotropic and Autonomous Personality Characteristics.Unpublished Manuscript, Duke University Medical Centre,Department of Psychiatry.Robins, C. J. (1991). Personal Communication.Robins, C. J., & Luten, A. G. (1991). Sociotropy and Autonomy:Differential Patterns of Clinical Presentation in UnipolarDepression. Journal of Abnormal Psychology, 100(1), 74-77.Robins, L., Helzer, J., Weissman, M., Orvaschel,H., Gruenberg,E., Burke, J., & Regier, D. (1984). Psychiatric Diagnoses inThree Sites. Archives of General Psychiatry, 41, 949-958.Rosenkrantz, P., Vogel, S., Bee, H., Broverman, I., & Broverman,D. (1968). Sex-Role Stereotypes and Self-Concepts in CollegeStudents. Journal of Consulting and Clinical Psychology, 32,287-295.Sacco, W. P. (1981). Invalid Use of the Beck DepressionInventory to Identify Depressed College-Student Subjects: AMethodological Comment. Cognitive Therapy and Research, 5(2),143-147.Safran J. D. (1990a). Towards a Refinement of CognitiveTherapy in Light of Interpersonal Theory: I. Theory. Clinical Psychology Review, 10, 87-105.Saf ran J. D. (1990b). Towards a Refinement of CognitiveTherapy in Light of Interpersonal Theory: II. Practice.Clinical Psychology Review, 10, 107-121.Segal, Z. V. (1988). Appraisal of the Self-Schema Construct inCognitive Models of Depression. Psychological Bulletin,03(2), 147-162.Segal, Z. V., Shaw, B. F., & Vella, D. D. (1989). Life Stressand Depression: A Test of the Congruency Hypothesis of Life84Event Content and Depressive Subtype. Canadian Journal ofBehavioural Science, 21(4), 389-400.Smith, T. W., O'Keefe, J. L., & Jenkins, M. (1988). Dependencyand Self-Criticism: Correlates of Depression or Moderators ofStressful Events? Journal of Personality Disorders, 2(2), 160-169.Steele, R. E. (1978). Relationship of Race, Sex, Social Class,and Social Mobility to Depression in Normal Adults. Journal ofSocial Psychology, 104, 37-47.Steer, R. A., Beck, A. T., & Garrison, B. (1986). Applicationsof the Beck Depression Inventory. In N. Sartorius & T. A. Baneds.), Assessment of Depression. Heidelberg: Springer-Verlag.Stephens, R. S., & Hokanson, J. E. (1987). Responses toDepressed Interpersonal Behaviour: Mixed Reactions in aHelping Role. Journal of Personality and Social Psychology,52(6), 1274-1282.Stoudemire, A., Frank, R., Hedemark, N., Kamlet, M., Blazer, D.(1986). The Economic Burden of Depression. General Hospital Psychiatry, 8(6), 387-394.Strupp, H. H., & Binder, J. L. (1984). Psychotherapy in a NewKey: A Guide to Time Limited Dynamic Psychotherapy. New York:Basic Books.Swann, W. B. (1987). Identity Negotiation: Where Two Roads Meet.Journal of Personality and Social Psychology, 53(6), 1038-1051.Swann, W. B., & Read, S. J. (1981). Self-Verification Processes:How We Sustain Our Self-Conceptions. Journal of Experimental Social Psychology, 17, 351-372.Watson, D. (1988). Intraindividual and Interindividual Analysesof Positive and Negative Affect: Their Relation to HealthComplaints, Perceived Stress, and Daily Activities. Journal of Personality and Social Psychology, 54(6), 1020-1030.Watson, D., Clark, L. A., & Tellegen, A. (1988). Development andValidation of Brief Measures of Positive and Negative Affect:The PANAS Scales. Journal of Personality and Social Psychology, 54(6), 1063-1070.Weissman, M. M., & Bothwell, S. (1976). Assessment of SocialAdjustment by Patient Self-Report. Archives of General Psychiatry, 33, 1111-1115.85Weissman, A. N., & Beck, A. T. (1978). Development andValidation of the Dysfunctional Attitude Scale. Paperpresented at the annual meeting of the Association for theAdvancement of Behaviour Therapy, Chicago.Welkowitz, J., Lish, J. D., & Bond, R. N. (1985). The DepressiveExperiences Questionnaire: Revision and Validation. Journal ofPersonality Assessment, 91, 89-94.Zuckerman, M., & Lubin B. (1965). Manual for the Multiple Affect Adjective Checklist. San Diego, CA: Educational andIndustrial Testing Service.Zuroff, D. C., Moskowitz, D. S., Wielgus, M. S., Powers, T. A.,& Franko D. L. (1983). Construct Validation of the Dependencyand Self-Criticism Scales of the Depressive ExperiencesQuestionnaire. Journal of Research in Personality, 17, 226-241.Zuroff, D. C., & Mongrain, M. (1987). Dependency and Self-Criticism: Vulnerability Factors for Depressive AffectiveStates. Journal of Abnormal Psychology, 96(1), 14-22.Zuroff, D. C., Igreja, I., & Mongrain, M. (1990).Dysfunctional Attitudes, Dependency, and Self-Criticism asPredictors of Depressive Mood States: A 12-Month LongitudinalStudy. Cognitive Therapy and Research, 14(3), 315-326.Table 1. BDI Means and Standard Deviations in Sociotropic and AutonomousGroups across Passive and Controlling Confederates.Sociotropic AutonomousControlling RolePassive Role15.61 (4.00)14.45 (3.95)16.26 (5.20)16.13 (4.83)Table 2. Warmth and Script Deviations Means and Standard Deviations in Sociotropic and AutonomousGroups across Passive and Controlling Confederates.Sociotropic AutonomousWarmthControlling Role 3.70 (0.47) 3.69 (0.62)Passive Role^. 3.75 (0.68) 3.69 (0.56)Script DeviationsControlling Role 2.48 (0.79) 2.27 (0.67)Passive Role 2.38 (0.97) 2.09 (0.60)47 49Table 3. Dependent Variables Means and Standard DeviationsAcross Sociotropic and Autonomous Groups.Sociotropic AutonomousInterpersonal Goals 5.78 (0.68) 4.98 (1.12) **Task Goals . 4.33 (1.13) 4.29 (1.22)Sociotropic Concerns 25.12 (5.65) 21.24 (6.84) **Autonomous Concerns 18.79 (6.54) 21.16 (6.48)Sociotropic Perceptions 30.34 (4.87) 27.35 (5.10) **Autonomous Perceptions 15.12 (7.91) 18.67 (9.06) *Confederate Liking 36.62 (7.10) 32.59 (7.30) **Subject Liking 32.32 (9.63) 25.43 (9.93) **47 49* p < .05^** p < .01Table 4. Dependent Variables Means and Standard DeviationsAcross Controlling and Passive Confederate Roles.Controlling PassiveInterpersonal Goals 5.24 (1.02) 5.51 (1.00)Task Goals 4.24 (1.14) 4.37 (1.21)Sociotropic Concerns 22.88 (6.19) 23.43 (6.96)Autonomous Concerns 21.65 (7.39) 18.27 (5.15) *Sociotropic Perceptions 28.29 (5.22) 29.36 (5.14)Autonomous Perceptions 21.35 (8.45) 12.34 (6.18) **Confederate Liking 32.71 (7.11) 36.48 (7.37) *Subject Liking 28.96 (11.31) 28.63 (9.32)N 49 47* p < .05^**p < .0190Footnotes1 An additional analysis on affective measures was performedutilizing pre-interaction affect as a covariate. An ANCOVAcarried out on post-interaction positive affect utilizing pre-interaction positive affect as a covariate showed no significantmain or interaction effects, p>.05. An ANCOVA carried out onpost-interaction negative affect utilizing pre-interactionnegative affect as a covariate showed no significant main orinteraction effects, 2>.05.Appendix AHere are a number of statements about personal characteristics. Please read each onecarefully, and indicate whether you agree or disagree, and to what extent, by circlinga number.91StronglyDisagree Disagree SlightlyDisagreeSlightlyAgree Agree StronglyAgree1 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 6.1 2 3 4 5 61 2 . 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5I. I often put other people'sneeds before my own.2. I tend to keep otherpeople at a distance.3. I find it difficult to beseparated from people I love.4. I am easily bothered by otherpeople making demands of me.S. I am very.sensitive to theeffects I have on the feelingsof other people.6. I don't like relying onothers for help.7. I am very sensitive tocriticism by others.S. It bothers me when I feelthat I am only average andordinary.,.. I worry a lot about hurtingor offending other people.10. When I'm feeling blue, I don'tlike to be offered sympathy.11. It is hard for me to breakoff a relationship even ifit is making me unhappy.12. In relationships, peopleare often too demanding ofone another.13. I am easily persuaded byothers.14. I usually view my performance^1^2^3^4^5^6as either a complete successor a complete failure.-2-92StronglyDisagree Disagree SlightlyDisagree SlightlyAgree Agree StronglyAgree1 2 3 4 5 61 2 3 4 5 61 2 3 4 s 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 615. I try to please other peopletoo much.16. I don't like people to invademy privacy.17. I find it difficult if Ihave to be alone all day.18. It is hard for me to takeinstructions from people whohave authority over me.19. I often feel responsible forsolving other people'sproblems.20. I often handle big decisionswithout telling anyone elseabout them.21. It is very hard for me toget over the feeling of losswhen a relationship has ended.22. It is hard for me to havesomeone dependent on me.23. It is very important to meto be liked or admired byothers.24. I feel badly about myselfwhen I an not activelyaccomplishing tfiings.25.1 feel I have to be niceto other people.26. It is hard for me to expressadmiration or affection. .27. I like to be certain thatthere is somebody close Ican contact in case somethingunpleasant happens to me.28. It is difficult for me tomake a long-term committmentto a relationship.-3-93StronglyDisagree Disagree SlightlyDisagree SlightlyAgree AgreeStronglyAgree1 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 629. I an too apologetic toother people.30. It is hard for me to openup and talk about my feelingsand other personal things.31. I an very concerned withhow people react to me.32. I have a hard time forgivingmyself when I feel I haven'tworked up to my potential.33.I get very uncomfortable^.when I'm not sure whether .or not someone likes me.34.When making a big decision, Iusually feel that advice fromothers is intrusive.35.It is hard for me to say.^"no" to other people'srequests.36.I resent it Oben people tryto direct my behavior oractivities.37. I become upset when somethinghappens to me and there'snobody around to talk to.38.Personal questions from othersusually feel like an invasionof my privacy.39. I am most comfortable whenI know my behavior is whatothers expect of me.40. I am very upset when otherpeople or circumstancesinterfere with my plans.41. I often let people takeadvantage of me.-4-94StronglyDisagree DisagreeSlightlyDisagreeSlightlyAgree AgreeStronglyAgree1 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 61 2 3 4 5 642. I rarely trust the adviceof others when making abig decision.43. I become very upset when afriend breaks a date orforgets to call me asplanned.44. I become upset more than mostpeople I know when limitsare placed on my personalindependence and freedom.45. I judge myself based on howI Chink others feel about me.46. I become upset when otherstry to influence my thinkingon a problem.47. It is hard for me to letpeople know when I amangry with them.48. I feel controlled when othershave a say in my plans.10 I do not feel sad.I feel sad.0 I don't feel I am any worse thananybody else.2 I am sad all the time and I can't snap out of it. I am critical of myself for my weaknessesor mistakes.3 I am so sad or unhappy that I can't stand it. 2 I blame myself all the time for my faults.3 I blame myself for everything bad2^0 I am not particularly discouraged about thefuture.that happens.2aI feel discouraged about the future.I feel I have nothing to look forward to.I feel that the future is hopeless and thatI don't have any thoughts of killing myself.I have thoughts of killing myself, but Iwould not carry them out.things cannot improve.2 I would like to kill myself.3^0 I do not feel like a failure.3 I would kill myself if I had the chance.I feel I have failed more than the 10average person. 0 I don't cry any more than usual.2 As I look back on my life, all I can see is I cry more now than I used to.a lot of failures. 2 I cry all the time now.3 I feel I am a complete failure as a person. 3 I used to be able to cry, but now I can't cryeven though I want to.4o I get as much satisfaction out of things as Iused to. 11 I am no more irritated now than l ever am.I don't enjoy things the way I used to. I get annoyed or irritated more easily than2 I don't get real satisfaction out of anything I used to.anymore.2 I feel irritated all the time now.3 I am dissatisfied or bored with everything. 3 I don't get irritated at all by the things thatused to irritate me.5 I don't feel particularly guilty.I feel guilty a good part of the time. 12 0 I have not lost interest in other people.2 I feel quite guilty most of the time. I am less interested in other people than3 I feel guilty all of the time. I used to be.2 I have lost most of my interest in11 I don't feel I am being punished. other people.1 I feel I may be punished.3 I have lost all of my interest in other people.2 I expect to be punished.3 I feel I am being punished. 13 0 I make decisions about as well asI ever could.7 I don't feel disappointed in myself. I put off making decisions more thanI used to.I am disappointed in myself. 2 I have greater difficulty in making2 I am disgusted with myself. decisions than before.3 I hate myself. 3 I can't make decisions at all anymore. Subtotal Page 1 CONTINUED ON BACKAppendix BDate-Name:^  Marital Status: Age:^ SexOccupation:  Education This questionnaire consists of 21 groups of statements. After reading each group of statements carefully,circle the number (0, 1, 2 or 3) next to the one statement in each group which best describes the way youhave been feeling the past week, including today. If several statements within a group seem to apply equallywell, circle each one. Be sure to read an the statements in each group before making your choice.yrihTHE PSYCHOLOGICAL CORPORATIONVIVHARCOLJRT BRACE JOVANOVICK INC.Copyright C 1978 by Aaron T. Beck. All rights reserved. Printed in the U.S.A.NOTICE: It is against the law to photocopy or otherwise reproducethis questionnaire without the publisher's written permission. 9-0183599514 o I don't feel I look any worse than I used to. 18^o I haven't lost much weight, if any, lately.I am worried that I am looking old orunattractive.I have lost more than 5 pounds.I have lost more than 10 pounds.2 I feel that there are permanent changesin my appearance that make me lookunattractive.I have lost more than 15 pounds.I believe that I look ugly. I am purposely trying to lose weight byeating less. Yes^No15 o I can work about as well as before.It takes an extra effort to get started atdoing something.20^a I am no more worried about my healththan usual.2 I have to push myself very hard to doanything.I am worried about physical problemssuch as aches and pains; or upsetstomach; or constipation.I can't do any work at all. I am very worried about physicalproblems and it's hard to think ofmuch else.18 o I can sleep as well as usual.I don't sleep as well as I used to.I am so worried about my physicalproblems that I cannot think aboutanything else.2 I wake up 1-2 hours earlier than usualand find it hard to get back to sleep.I wake up several hours earlier than Iused to and cannot get back to sleep. 21-o I have not noticed any recent changein my interest in sex.I am less interested in sex than I usedto be.^-17 o I don't get more tired than usual. 2 I am much less interested in sex now.1 I get tired more easily than I used to. 3 I have lost interest in sex completely.2 I get tired from doing almost anything.3 I am too tired to do anything.18 My appetite is no worse than usual.My appetite is not as good as it used to be.2 My appetite is much worse now.3 I have no appetite at all anymore.Subtotal Page 2Subtotal Page 1'Ibtal ScoreTPC 0528.001^ 181920 BCDE961097Appendix CGOAL FOCUS 1.) To what extent were you focused on achieving your goals inthe task?2.) To what extent did you want to obtain your own objectives inthe task?3.) To what extent were you concerned about the goals that youwanted to achieve?INTERPERSONAL FOCUS1.) To what extent did you want everything to go smoothlybetween you and your partner?2.) To what extent were you concerned about maintaining a goodunderstanding between you and your partner?3.) To what extent did you wish to maintain an agreeablerelationship between you and your partner?98Appendix DInteraction ConcernsSOCIOTROPY CONCERNS Other's Evaluation1.) To what extent was it important that your partner approvedof you?2.) To what extent were you concerned about what your partnerthought of you?Dependency3.) To what extent did you rely on your partner for help?4.) To what extent were you dependent on your partner during thetask?Pleasing Others995.) To what extent did you try to please your partner?6.) To what extent did you feel you had to be nice to your partner?AUTONOMY CONCERNS Need for Control1.) To what extent were you upset when your partner interferedwith your plans?2.) To what extent did you feel controlled when your partnertried to influence you?Defensive Separation3.) To what extent did you try to keep your partner at adistance?4.) To what extent were you comfortable relying on your partnerfor help?Perfectionism1005.) To what extent did you feel that the interaction didn't liveup to your expectations?6.) To what extent did you feel that the interaction fell shortof your standards?101Appendix EPerception of Partner's BehaviourSOCIOTROPY PERCEPTIONS Other's Evaluation1.) To what extent did your partner have a negative reaction toyou?2.) To what extent did your partner think positively of you?Dependency3.) To what extent was your partner supportive of you in thetask?4.) To what extent could you rely on your partner to providesupport to you in making the decisions?Pleasing Others5.) To what extent was your partner pleased with you?1026.) To what extent did your partner seem pleased with your behaviour?AUTONOMOUS PERCEPTIONSControlling1.) To what extent did your partner try to control you too muchduring the task?2.) To what extent did your partner try to influence you toomuch?Intrusiveness3.) To what extent did your partner intrude on you?4.) To what extent was your partner too forward during theconversation?Perfectionism5.) To what extent did you think your partner's standards weretoo high?6.) To what extent was your partner too perfectionistic?103104Appendix FWe would like to get some idea of how you feel about yourpartner. Answer the following questions according to the scalebelow. Put the number that best describes the way you feel inthe blank next to the question.1^2^3^45 6 7definitely not^probably not^probably yesdefinitely yesWould you like to meet this person again? Would you like to spend more time with her in theWould you like to work with this person in theWould you like to sit next to her on a three hourfuture?future?busride?^ Would you invite this person to visit you?^ Would you like to have this person as a friend?^ Would you ask this person for advice?^ Would you consider sharing an apartment with thisperson or having her^ for a roommate?105Appendix GThis scale consists of a number of words that describe differentfeelings and emotions. Read each item and then mark theappropriate answer in the space next to that word. Indicate towhat extent you feel this way right now, that is at the presentmoment. Use the following scale to record your answers.1^ 2^3^45very slightly^a little^moderately quite a bitextremelyor not at all^interested ^irritable^distressed ^alert^excited ^ashamed^upset ^inspired^strong ^nervous^guilty ^determined^scared ^attentive^hostile ^jittery^enthusiastic ^activeproud ^afraid106Appendix HROLE 1-CONTROLLINGWell, since I'm supposed to be helping you with exercising moremaybe I should take control of things.I guess we should figure out what kinds of things you should bedoing and when you're going to do them.Why don't we start by choosing the kinds of exercise you'll wantto do.We need to pick some activities that will build physicalstrength and also, something aerobic for cardiovascular fitness,ok?Since we've got this schedule, I'll write this down for you.Let's start with aerobic stuff. Now, that kind of exercise canbe, like a team sport, or an individual activity like aerobicsor dancing. The key, of course, is to increase your heart ratefor at least a half an hour or so.Since you're going to school and your schedule is already prettyhectic you probably wont have time for a serious team sport. I107really think aerobics would be ideal, they're good for you and alot fun too. Oh, its best to start with low impact kind.[If subjects say they don't like aerobics say: "Well, give thema try and if you don't like it try something like aquafit, thestuff they do in water."]So let me write that down. [note at bottom of schedule][Pause and Think]Now, your also going to really want to include something toincrease your muscle tone.The best exercise for strength and toning is really weighttraining and, again, you can do this on your own so itsconvenient. I guess there are many different kinds of weighttraining but free weights are probably best.[If subjects say they don't know anything about weight trainingsay: "Most gyms have instructors that will show you what todo."][If subjects say they dislike weight training say: "If you walkwith some of those weights that they have for your ankles andwrists that might also do the trick."]108I'll also write down free-weight training. [note at the bottomof the schedule]That should do it, unless you have something you'd like to add.[Pause and Think]All right. So I think you should do those two activities, andyou should really do them at least 3 times a week.Now... I suppose you can do aerobics and weights separately, butI think its probably best if you do them one after the other.Its probably cheapest and most practical for you to work out oncampus so you should look into that.[Pause and Think]OK, I think that the best time for you to exercise would beduring lunch time. It gives you a nice break in the day andwill give you some energy for the afternoon.So lets see, [examine schedule]Its always nice to have the weekends free so I think a Monday,Wednesday, Friday schedule makes the most sense.109It should take about an hour so I'll fill in from noon until oneon those three days.[If subjects have classes pick the hour closest to lunch forthem]All right, I think that if you do these two activities on aregular basis you'll be in great shape.So... I'd say we're done.110ROLE 2- PASSIVEOK, since I'm supposed to be helping you I think its importantthat you control things and tell me what kind of exercise you'dlike to do.I think Canada Fitness says that people need two types ofexercise, one that increases cardiovascular fitness and anotherthat will help to build strength. Would you like to work on oneor both or something else?[let subject direct to start area, and always return to italics.If they say "one" say "cardiovascular or strength?", if both say"Would you like to start with cardiovascular or strength?"]I think cardiovascular exercise is something that would increaseyour heart rate for at least a half an hour or so, it can bedone individually or in some kind of team sport. Are there anyaerobic activities that you would like to try?[wait for subject response, and then say:]Maybe you'd like to write this down or should I?[when subject gives aerobic activities say:]111That sounds pretty good.Fine.Great.[If there are three or more activities say: "That will be toughto fit in to your schedule, don't you think?"]Is there anything else you might like to do, or is that it?[wait for additional responses and then say:]I guess strength exercise, develops muscle tone and usuallyinvolves weight training. Do you think you'd like to trysomething like that?[wait for subject response]That sounds pretty good.Fine.Great.112[If there are three or more activities say: "That will be toughto fit in to your schedule, don't you think?"]What about [something aerobic] or [something for strength]?[say one or both depending on previous choice, or skip if allbases covered]Anything else?[wait and then say:]Now. [pen drop] You may want to think about where you're goingto work out. There are plenty of fitness clubs and communitycentres around and there are always the campus facilities. ButI'll leave that up to you to decide on your own.I guess they say its a good idea to be active on a regularbasis. Do you have certain days that might work better thanothers?[wait for days, if none forthcoming skip to times.]And any specific times?[wait for times, if none forthcoming say: "maybe you could checkout a few places and find a time that suits you."]113[allow subject to create schedule and offer support, includingnodding. then say:]Would you like to fill out the schedule or should I?[while completing schedule say:]OK, that looks good.I think that would be fine.[When activities are scheduled say:]Do you think this will help get you started?[if no: say "Actually going that first day is always thetoughest. But once you start it gets easier."][wait for response: allow other activities to be added orchanged]OK, that's great. Do you think maybe we're done?114Appendix I What we are interested in studying is how people interact whenthey are trying to help each other with a problem. We believethat people use many different styles of helping behaviour andwe're interested in finding out what those styles are and howeffective the various strategies are.So today we'd like you to take turns helping each other with aspecific problem. Now, discussing problems in your own life maybe a bit too personal so we are going to supply a problem foryou to work on. We'd like you to consider how this issueaffects your life while you are talking.Do you have any questions so far?The problem that we'd like you to discuss is exercising andfitness. Many people feel that they could improve themselves inthis area and are unhappy with their level of activity. So tryto think about how this issue impacts on your life and what youcould do to be more fit.While you two are talking I'll be behind the one-way mirror.When you think you're finished with this first problem just letme know and I'll come back in. I'll let you know when I getback there and then you can begin.


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