S O C I O T R O P Y A N D A U T O N O M Y A N D T H E I N T E R P E R S O N A L M O D E L O F

D E P R E S S I O N : A N I N T E G R A T I O N

by

Peter J . Biel ing

B .Sc . (Hons.) University of Victor ia, 1990

M.A. University of Brit ish Co lumbia , 1993

A T H E S I S S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F

T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F

D O C T O R O F P H I L O S O P H Y

IN

T H E F A C U L T Y O F G R A D U A T E S T U D I E S

( D E P A R T M E N T O F P S Y C H O L O G Y )

W e accept this thesis as conforming

to the required standard

T H E U N I V E R S I T Y O F BRIT ISH C O L U M B I A

June 1997

© Peter J . B ie l ing, 1997

In presenting this thesis in partial fulfilment of the requirements for an advanced

degree at the University of British Columbia, I agree that the Library shall make it

freely available for reference and study. I further agree that permission for extensive

copying of this thesis for scholarly purposes may be granted by the head of my

department or by his or her representatives. It is understood that copying or

publication of this thesis for financial gain shall not be allowed without my written

permission.

Department

The University of British Columbia Vancouver, Canada

DE-6 (2/88)

Abstract

Resea rche rs and theorists have suggested that two personal i ty styles may serve

as pathways for the deve lopment of depress ion. One personal i ty style, sociotropy,

involves intense needs for posi t ive interchange with others, whereas the other style,

autonomy, involves an excess i ve need for self-control and independence. T h e s e

personal i ty styles were invest igated in the context of C o y n e ' s (1976) interpersonal

model of depress ion , which sugges ts that depressed persons are rejected by others.

R e s e a r c h on this model has been equivoca l , and it is poss ib le that these two

personal i ty d imens ions result in meaningful interpersonal di f ferences within depressed

persons. Dep ressed outpatients, (N=41) and non-depressed controls (N=41) were

a s s e s s e d on sociotropy and autonomy and then participated in a brief task in which a

research assistant he lped them plan adapt ive life changes. T h e s e interactions were

rated subjectively and uti l izing behavioura l coding. Resul ts indicated that a

combinat ion of depress ion and autonomy were particularly l ikely to lead to rejection

and less posit ive interpersonal behaviours. Sociotropy was related to percept ions of

interpersonal deference, whereas depress ion was assoc ia ted with self-orientation in

the task. T h e s e results suggest that both depress ion and personal i ty impact rejection

and interpersonal behaviours in soc ia l interactions. Overa l l , this study represents a

signif icant step toward greater specif ici ty in the interpersonal model of depress ion, and

cl inical implicat ions of these f indings are descr ibed.

i i i

Table of Contents

Abstract ii

Table of Contents iii

List of Tables vii

List of Appendices viii

Acknowledgments ix

Introduction . 1

Theoretical Evolution of Sociotropy and Autonomy 2

Construct Assessment 5

Research Evidence 8

Diathesis-Stress Models of Depression 8

Relationship of Sociotropy and Autonomy to Other Variables 10

Parental Relationships. 11

Attributional Research. 12

Gender Roles. 12

Differential Symptomatology. 13

Therapy Outcome. 14

Psychological Correlates. 15

i v

Interpersonal Models of Depression 17

Depression and Interpersonal Rejection 19

Interpersonal Concomitants of Sociotropy and Autonomy 26

Research Evidence 28

The Present Study 31

Hypotheses 34

Method 35

Overview 35

Participants 35

Depressed Sample 35

Inclusion Criteria for Depressives. 37

Control Sample 37

Inclusion Criteria for Control Group. 39

Experimenters 39

Confederates 39

Procedure 40

Experimental Task 41

Diagnostic and Classificatory Measures 41

V

Structured Clinical Interview for DSM-III-R 41

Beck Depression Inventory 42

Personal Style Inventory 42

Dependent Measures 43

Desire for Future Interactions 43

Confederate Perceptions of Participants 44

Participant Behaviour 45

Results 46

Preliminary Analyses 46

Participant Characteristics 46

Confederate Checks 47

Participant Gender 47

Description of Independent Variables 48

Main Analyses .....49

Overview 49

Desire for Future Interactions 50

Confederate Perceptions of Participant 52

Interpersonal Deference. 52

Self-Orientation. 53

Participant Behaviours 54

v i

Discussion , 57

Overview 57

Rejection, depression, and personality 58

Depression, Personality, and Others' Perceptions 59

Depression, Personality, and Behaviour 60

Implications for the Interpersonal Model 62

Caveats 66

Future Directions 67

Overall Summary 68

Footnotes 69

References • 70

Appendices • • 81

v i i

List of Tables

Tab le 1: Recruitment Sou rces 36

Tab le 2: Samp le Character is t ics 38

Tab le 3: M e a n s and standard deviat ions of sociotropy and

autonomy scores in d e p r e s s e d and control groups 48

Tab le 4: Corre lat ions between des i re for future interactions (DFI),

sociotropy, autonomy, and depress ion 50

Tab le 5: Hierarchical regress ion ana l yses predicting rejection from

depress ion , sociotropy, autonomy, and interactions 51

Tab le 6: Part ial correlat ions between interpersonal deference, sociotropy,

autonomy, and depress ion 52

Tab le 7: Hierarchical regress ion ana l yses predict ing interpersonal deference

from depress ion, sociotropy, autonomy, and interactions 53

Tab le 8: Corre lat ions between self-orientation, sociotropy, autonomy, and

depress ion 53

Tab le 9: Hierarchical regress ion ana l yses predict ing self-orientation from

depress ion , sociotropy, autonomy, and interactions 54

Tab le 10: Corre lat ions between rated behaviours and sociotropy, autonomy,

and depress ion 55

Tab le 11: Hierarchical regress ion ana l yses predicting posit ive behaviors

from depress ion, sociotropy, autonomy, and interactions 56

List of Appendices

Appendix A: Task Script 81

Appendix B: Pleasant Events Lists 84

Appendix C: Beck Depression Inventory 85

Appendix D: Personal Style Inventory 86

Appendix E: Confederate Perceptions of Participants 90

Appendix F: Descriptive Statistics of Dependent Variables 92

i x

Acknowledgments There are many people who have helped in different capac i t ies to bring this

project to a conc lus ion . S o m e have contributed in a profess ional way, others in a

personal one, and most have he lped in both ways. First, I w ish to thank my advisor,

Lynn A lden for her contributions: reading (and re-reading) my work, giving me feedback

and advice, teaching me c learer writing and thinking, providing support whenever the

waiting list ran low, and turning significant statistical tests into a meaningful story. I

a lso wish to thank the members of my committee, Darr in Lehman and Dimitri

Papageorg is for their different perspect ives on critical i ssues of the project.

Then there are the people without whom the data cou ld not have been gathered,

who had to show up e a c h and every day. Thank you to S h a w n Reyno lds and Mar iana

Brussoni , who cheerful ly p lanned pleasant events with the multi tudes. No less

important, Beren ica Ve jvoda, and Tracy Tanchuk, who ably coded the v ideotapes

without ever nodding off. Thanks for the reliabil it ies.

1

Introduct ion

Scho la rs from a broad variety of psycholog ica l perspect ives have examined

personal i ty patterns that may precipitate and mediate depress ion. Despi te d iverse

backgrounds, the work of these theorists and researchers converges on two sets of

personal i ty features that are thought to serve as pathways for the development of

depress ion. The first pathway, termed sociotropy, involves intense needs for posit ive

interchange with others, dependency , and he lp lessness . The second pathway, termed

autonomy, involves an excess i ve need for achievement, self-control, and independence

from others (Arieti & Bemporad , 1980; Beck 1983; Blatt, 1974; Bowlby, 1977). T h e s e

personal i ty features are theor ized to have distinct interpersonal components (eg. Beck,

1983; Blatt & Zuroff, 1992) that may be meaningful for both the onset and maintenance

of depress ion.

C o y n e (1976a) p roposed an interpersonal model of depress ion which focuses on

other's responses to depress ives , and how such responses might exacerbate

depress ion. However, empir ical support for Coyne ' s model has been equivoca l . One

poss ib le explanat ion for these confl ict ing f indings is that research in this a rea has not

taken into account meaningful interpersonal heterogeneity among dep ressed persons

(Marcus & Nardone, 1992). Indeed, previous research has sugges ted that sociotropy

and autonomy affect the soc ia l impact of individuals on others (Biel ing & A lden , 1996).

The current research examines the interpersonal impact of these personal i ty features in

interactions of dep ressed individuals. Toward this end, the theoretical evolut ion of

2

these personal i ty features will be summar ized , fol lowed by a review of the empir ical

literature that examines the consequences of these features for depress ion.

Theoretical Evolution of Sociotropy and) Autonomy

T h e involvement of sociotropy and autonomy in depress ion has been descr ibed

by at least four groups of theorists, each from a different psychologica l perspect ive.

Al though there are dif ferences in each of the four approaches, their commonal i ty is

more striking. T h e labels for these personal i ty features vary considerably from one

approach to another. Autonomy has been labeled self-crit icism, dominant goal

orientation, and self-definit ion. Sociotropy has been labeled dependency, dominant

other orientat ion, and interpersonal re la tedness (Blatt & Zuroff, 1992). R e s e a r c h

indicates that, despi te dif ferences in terminology, sociotropy and dependency , as well

as autonomy and self-cri t icism, are over lapping if not ana logous constructs (Alden &

Bie l ing, 1996; B laney & Kutcher, 1991; Nietzel & Harris, 1990). The current work will

use the terms sociotropy and autonomy to refer to these personali ty features except

when referring to spec i f ic studies or speci f ic theoretical approaches.

Wr i t ing from a developmental perspect ive, John Bowlby (1969, 1977, 1980)

descr ibed "compuls ive ly self-reliant" and "anxiously at tached" individuals and how such

individuals are prone to depress ion. Bowlby be l ieved that excess ive self re l iance was

a defens ive react ion to early frustrated attachment needs and that anxious attachment

was a react ion to unrel iable caregiver response. T h e s e different patterns of ear ly

attachment behav iour were thought to result in internal "models" and expectat ions

3

regarding how other people behave. Such dysfunct ional work ing models were

hypothes ized to be carr ied through until adul thood and to pred ispose an individual to

the exper ience of depress ion .

Wri t ing from a psychodynamic perspect ive, S idney Blatt (1974) dist inguished

two subtypes of depress ion . The first, anacl i t ic depress ion , w a s character ized by

feel ings of he lp lessness , weakness , and deplet ion. S u c h individuals were theor ized to

have a strong need to be cared for by others and to va lue c lose socia l contact beyond

all e lse. They were sa id to be fearful of los ing the support of others, and thus to have

trouble express ing dissat isfact ion or anger. Blatt termed the second subtype

introjective depress ion and character ized it by feel ings of wor th lessness, guilt, and a

sense of hav ing fa i led to live up to expectat ions or s tandards. S u c h individuals were

descr ibed as highly competit ive, demanding, and crit ical of others because of their

intense compet i t iveness. Anacl i t ic depress ion was thought to deve lop in an individual

who was excess ive ly dependent whereas introjective depress ion was thought to

develop b e c a u s e of excess ive self-crit icism.

Arieti and Bemporad (1980) made a similar bipartite distinction based on an

examinat ion of two decades of psychotherapy with approximately forty depressed

patients. They pointed out that depress ion results when an individual rel ies

excess ive ly on external sources of support to maintain se l f -es teem and gratification

(Arieti & Bemporad , 1980) They dist inguished between two types of excess ive

rel iance, "dominant other" and "dominant goa l . " Dominant other corresponds to

excess ive dependency and was descr ibed as a pattern of relating to others that begins

4

in chi ldhood. Here, the person seeks rewards through some signif icant other(s) who is

rel ied upon to bestow meaning, gratification and sel f -esteem (Arieti & Bemporad,

1980). S u c h persons are descr ibed as cl ingy, pass ive, manipulat ive, and anger

avoidant. The other form of excess ive rel iance was termed dominant goal and

cor responds to excess i ve autonomy. Th is pattern was sa id to be estab l ished in

chi ldhood if parents p laced an excess ive emphas is on achievement . Eventual ly, "the

individual se lects some fantast ic goal for himself which he then pursues fanatically,

apparently for its own s a k e " (Arieti & Bemporad, 1980; p 1361). Ach ievement of this

goal has broad and unjustif iable meaning attached to it. In addit ion these individuals

are reclusive, arrogant, and obsess ive . Arieti and Bemporad argued that these

patterns result in interpretations of speci f ic life events that may lead to depress ion.

Wri t ing from a cognit ive perspect ive, Beck (1983) descr ibed personali ty

"modes, " which he ca l led sociotropy and autonomy. Sociot ropy (social dependency)

was descr ibed as "the person 's investment in posit ive interchange with others" (p. 272).

Su ch individuals were descr ibed as dependent on soc ia l feedback for gratification and

support. Autonomy w a s descr ibed as "the person's investment in preserving and

increasing his independence, mobility, and personal r ights" (p. 272). Such persons

derive gratif ication from direct ing their own activities and attaining meaningful goals.

Beck outl ined symptom clusters which would be more frequently assoc ia ted with one

mode or the other, desc r ibed events that would precipitate depress ion in one type or

the other, and examined speci f ic therapeut ic foci for both modes.

5

Theoret ica l work regarding sociotropy and autonomy in depress ion has sparked

numerous invest igat ions examin ing the role of these personal i ty features in depress ive

disorders. Th is research can be d iv ided into several broad categor ies: (1) assessmen t

of the personal i ty features and development of s tandard ized measures , (2) attempts to

investigate vulnerabil i ty to depress ion by examining the speci f ic interactions between

negat ive life events and sociotropy or autonomy, and (3) attempts to identify individual

di f ferences between depressed individuals who are sociotropic or autonomous. E a c h

of these three a reas will be reviewed, beginning with the formal assessmen t of these

personal i ty features.

Construct A s s e s s m e n t

There have been three major attempts to formally measure sociotropy and

autonomy. The first was made from a psychodynamic perspect ive by Blatt (Blatt,

D'Affliti, & Quin lan, 1976) and the second attempt from a cognit ive perspect ive by Beck

and his co l leagues (Beck, Epste in , Harr ison, & Emery, 1983). T h e most recent

measure was deve loped by Rob ins and his assoc ia tes as a response to psychometr ic

problems evident in the two previous measures (Robins, Ladd , Welkowi tz , B laney,

Diaz, & Kutcher, 1994).

Blatt and his co l leagues (1976) set out to a s s e s s dependency and self-cri t icism

by examin ing a wide variety of exper iences related to depress ion. Thei r sca le , the

Depress ive Exper iences Quest ionnai re or D E Q , was rationally constructed and

a s s e s s e s distorted or depreciated s e n s e of self, dependency, he lp lessness ,

6

egocentricity, fear of loss, ambiva lence, difficulty in deal ing with anger, sel f -b lame,

guilt, loss of autonomy, and distortions in family relat ions (Blatt et al . , 1976). The sca le

was found to have three underlying factors. The first was ca l led dependency and

included fears of abandonment, lonel iness, and he lp lessness . The second was ca l led

self-cr i t ic ism and concerned feel ings of guilt, hope lessness , life dissat isfact ion,

insecurity, and a s e n s e of having fai led to meet expectat ions. The third factor was

ca l led eff icacy and concerned one's perce ived resources and capaci t ies. The D E Q is

the oldest and most establ ished measure of these personal i ty features and as such has

been used most frequently in research. However, it appears that signif icant problems

exist in the sca le ' s psychometr ic propert ies; the D E Q has been cri t icized for its lack of

factor purity, particularly in cl inical samples . Moreover, there are quest ions concern ing

the content validity of D E Q items and the method used for sca le scor ing (Vigl ione,

C lemmey, & Cammenzu l i , 1990; Welkowi tz , L ish , & Bond , 1985).

A more recent attempt to measure the personal i ty features of sociotropy and

autonomy comes in the form of the Sociot ropy and Autonomy S c a l e (SAS) which was

deve loped by Beck and his assoc ia tes (Beck, Epste in , Harr ison, & Emery, 1983). Items

to a s s e s s the personal i ty features were drawn from patients' self-reports and cl inical

records. T h e items were factor ana lyzed and y ie lded two factors corresponding to

sociotropy and autonomy. However, the S A S has a lso been cri t icized on psychometr ic

grounds. T h e autonomy sca le in part icular is v iewed as limited because its items load

on both autonomy and sociotropy, the low inter-correlations between sub-factors on the

7

autonomy sca le , and its negat ive correlation with depress ion (Robins, Block, &

Pese low, 1989; Rob ins & Block, 1988).

In an attempt to rectify cri t icisms of both the D E Q and the S A S , Rob ins and his

co l leagues deve loped a sociotropy / autonomy sca le , the PSI (Personal Style

Inventory; Rob ins , et a l . , 1994). The PSI was created with items from the S A S and

D E Q , as wel l a s the Dysfunct ional Attitude S c a l e and the Inventory of Interpersonal

Prob lems (Horowitz, Rosenbe rg , & Baer, 1988). The sca le cons is ts of two 24-item

sets, one measur ing autonomy and the other sociotropy. Unl ike its predecessors , the

PSI demonstrates excel lent reliability (i.e. high internal cons is tency and factor purity)

and validity (i.e. appropr iate patterns of convergent and discriminant correlat ions with

other measures) .

T o summar ize , three instruments are presently used to a s s e s s sociotropy and

autonomy. T h e first evo lved from Blatt's psychodynamic perspect ive and the second

from Beck 's cogni t ive perspect ive. The third measure evo lved as a response to

psychometr ic prob lems inherent with the two other avai lab le measures . Interestingly,

results from research uti l izing these different measures are often treated as if the three

sca les , and the constructs they are des igned to measure , are ana logous (e.g. Blatt &

Zuroff, 1992; Nietze l & Harr is, 1990). Such an approach s e e m s warranted in light of

recent research ev idence which has found that the D E Q , S A S , and PSI appear to

measure quite s imi lar personal i ty features (Alden & Bie l ing, 1996; B laney & Kutcher,

1991). T h e s e f indings offer some justification for treating studies that employ these

different measu res a s a coherent body of empir ical literature. Next, a summary of

8

research investigating these personal i ty features and their role in depress ion will be

presented.

Resea rch Ev idence

A meta-analys is completed by Nietzel and Harr is (1990) examin ing the

relat ionship between sociotropy, autonomy, and measures of depress ion found highly

signif icant mean effect s i zes in the studies sampled. For sociotropy, the mean effect

s i ze related to measures of depress ion (in terms of r) was computed to be .28, p_ <

.0001. For autonomy, the mean effect s ize was .31, p_ < .0001. Stud ies investigating

the role of these personal i ty features are quite diverse, but have general ly involved

either an evaluat ion of d ia thes is-s t ress models of depress ion or the investigation of

di f ferences between sociotropic and autonomous individuals. E a c h of these areas will

be examined in subsequent sect ions.

D ia thes is -St ress Mode ls of Depress ion

Pe rhaps the most concer ted and coherent research efforts in this a rea have

come in attempts tb link sociotropy and autonomy, negat ive life events, and depress ion.

Resea rche rs examin ing life events and depress ion have sugges ted that speci f ic

depressogen ic schemata may become act ivated when indiv iduals are confronted with

personal ly meaningful stressful life events (Hammen, Ellicott, & Git l in, 1992; Hammen,

Ellicott, Git l in, & Jam ison , 1989; Hammen, Marks, Mayo l , & deMayo , 1985; Rob ins ,

1990; Rob ins & Block, 1988; S e g a l , Shaw, & Ve l la , 1989). Moreover , this type of

diathesis-s t ress model incorporates the v iews of researchers who have emphas ized the

role of stressful events in depress ion (Bi l l ings & Moos , 1982; C o y n e & Gotl ib, 1983;

Kel ler, Lavor i , R ice , Cryel l , & Hirschfe ld, 1986). Accord ing to this d iathesis-st ress

model , individuals who are sociot ropic would be vulnerable to depress ion when they

suffer a perce ived loss within the interpersonal realm whereas individuals high in

autonomy would be vulnerable if they exper ience an impediment to goal attainment in

ach ievement realms (Beck, 1983; Hammen et a l . , 1989; Hammen et a l . , 1985; Robins,

1990; Rob ins & Block, 1988).

Stud ies in this a rea have used a variety of methods including cross-sect iona l ,

longitudinal, and laboratory ana logue methods. They a lso differ in terms of the types of

populat ions studied; some have used co l lege students whereas others have used

cl inical ly dep ressed individuals. O n e recent study examined both symptom change and

psychophys io log ica l responses to different types of st ressors (Al len, de L'Horne, &

Trinder, 1996). Despi te these different des igns, populat ions, and measures , the results

of this work are highly consistent. Most studies have been able to show a signif icant

interaction of negat ive interpersonal events and sociotropy to predict symptoms of

depress ion . However, these s a m e studies do not support the interaction of negat ive

ach ievement events and autonomy to predict depress ion (Al len et a l . , 1996; Hammen

et a l . , 1985; Hammen et al . , 1992; Rob ins & Block, 1988, Rob ins , 1990; Zuroff &

Mongra in , 1987). It has been sugges ted that autonomous individuals are vulnerable to

a broader range of negat ive events than previously thought (Blatt & Zuroff, 1992;

Nietzel & Harr is, 1990), but this notion has not yet been empir ical ly examined. Th is as

10

yet unreso lved i ssue concern ing autonomy highlights the need to understand what

these personal i ty features measure and to identify their psycholog ica l correlates.

Re la t ionsh ip of Sociotropy and Autonomy to Other Var iab les

R e s e a r c h e r s have sought to link a variety of var iables to sociotropy and

autonomy or to dist inguish individuals who differ on these personal i ty features. Unl ike

research that has sought to evaluate the d iathesis-st ress model , this work has not

evo lved as clearly. Th is seems largely due to the great variety of var iables studied, the

somet imes tenuous theoretical grounding of the work, and the fact that research is

often not programmatic, result ing in isolated studies that are not repl icated. T h e s e

difficulties have led to equivocal f indings that have not been fol lowed by further

investigat ion or ref inements of methodology. R e s e a r c h on di f ferences between

sociotropic or au tonomous individuals has examined famil ial background (Blatt, W e i n ,

Chevron , & Qu in lan , 1979; McCran ie & B a s s , 1984; S a d e h , Rub in , & Berman, 1993;

Whif fen & Sassev i l l e , 1991), di f ferences in attributional style (Brewin & Furnham, 1987;

Brown & S i lberschatz , 1989), different symptom presentat ion (Robins, Block, &

Pese low, 1989; Rob ins & Luten, 1991), treatment outcome (Blatt, 1992; Pese low,

Rob ins , Sanf i l ipo, Block, & F ieve, 1992), and personal i ty correlates (Cappe l iez , 1993;

Mongra in , 1993; Pi lkonis , 1988; Ri ley & McCran ie , 1990; Zuroff, 1994).

11

Parenta l Rela t ionsh ips.

R e s e a r c h on famil ial di f ferences has evo lved from the psychodynamic

perspect ive of Blatt (1974), who emphas ized the role of object relat ions and

attachments to signif icant others in the etiology of sociotropy and autonomy. Blatt

highlighted the role of a strict mother f igure in the development of anacl i t ic (dependent)

depress ion and descr ibed both parents as strict and demand ing of achievement in

introjective (self-crit ical) depress ion. However, empir ical invest igat ions have proven to

be inconclusive. It wou ld appear that the mother p lays a larger role in the development

of dependency , a conc lus ion which is congruent with Blatt 's descr ipt ions (McCran ie &

Bass , 1984; S a d e h , et a l . , 1993). However, these studies conflict in their conc lus ions

i

regarding what types of maternal behaviour result in dependency . It is more apparent

that self-cri t icism is related to parental str ictness and an emphas is on achievement

(Blatt, et a l . , 1979; M c C r a n i e & B a s s , 1984; S a d e h , et a l . , 1993; Whi f fen & Sassev i l le ,

1991). However, s tudies in this a rea suffer from common methodological problems.

First, this research is c ross-sect iona l and thus b iased recol lect ions or reporting of i i

information about parents cannot be ruled out. S e c o n d , measu res examining these

parental var iab les are typical ly not s tandard ized; each study appears to use newly

constructed measures . Interestingly however, these studies do demonstrate significant

assoc ia t ions between family background and development of spec i f ic personali ty

styles.

12

Attributional R e s e a r c h .

Both theory and research highlight the importance of internal, stable, and global

attributions in depress i ves (Abramson, Se l igman, & Teasda le , 1978). Studies have

a lso been conducted to compare the attributional style of sociot ropic and autonomous

individuals (Brewin & Furnham, 1987; Brown & Si lberschatz, 1988). Resu l ts of both

studies suggested that level of depress ion , sociotropy, and autonomy were posit ively

related to both internal and global attributions. T h e s e assoc ia t ions lend some support

to the notion that sociotropy and autonomy may serve as vulnerabil i ty factors by

inf luencing attributions. However , the results do not support the predict ion that

sociotropy and autonomy are assoc ia ted with unique, differential attributional styles.

G e n d e r Ro les .

Commonal i t ies between male and female gender role attributes and Blatt's

descr ipt ions of self-cr i t ic ism and dependency have led researchers to examine the

relat ionship between these personal i ty features and gender role attributes (Chevron,

Quin lan, & Blatt, 1978; Zuroff, Moskowi tz , Wie lgus , Powers , & Franko, 1983). T h e s e

studies demonstrated that competency, cons idered a des i rab le male trait, was

negatively related to dependency in both men and women. Furthermore, warmth, a

des i rab le female trait, w a s related to dependency in men and unrelated to dependency

in women (Chevron et a l , 1978). Dependency has a lso been assoc ia ted with low

levels of mascul ini ty and high levels of femininity (Zuroff et a l . , 1983). Thus , the

13

construct of dependency seems to be l inked to low levels of mascul ini ty and warmth

whereas self-cr i t ic ism seems to be independent of gender role attributes.

Differential Symptomatology.

Another recent approach to val idate the constructs of sociotropy and autonomy

has been to differentiate symptoms postulated by Beck (1983) to occur in one subtype

or another (Robins, Block, & Pese low, 1989; Rob ins & Luten, 1991). Rob ins and his

co l leagues (1989) studied dep ressed inpatients and outpatients who were c lass i f ied as

sociotropic or autonomous on the S A S . B a s e d on theoretical descr ipt ions of

sociotropic and autonomous symptom patterns (Beck, 1983), the authors p laced items

from the BDI (Beck et a l . , 1961) and Hamil ton Rat ing S c a l e for Depress ion ( H R S D ;

Endicott, Nee , C o h e n , F le iss , & Sarantakos , 1981) into two clusters. Soc io t rop ic

symptoms were signif icantly related to sociotropy and negatively related to autonomy.

Au tonomous symptoms were negat ively related to sociotropy, but they were not

posit ively related to autonomy as had been predicted. The authors pointed out that

prob lems with the autonomy subsca le of the S A S may have been respons ib le for lack

of f indings with this subtype, a notion wh ich was fol lowed up in the s e c o n d study.

T h e s e c o n d study in this ser ies uti l ized the PSI (Robins et a l . , 1994), and a more

comprehens ive assessmen t of symptoms postulated by Beck (1983) to be related to

sociotropy or autonomy (Robins & Luten, 1991). T h e s e researchers found a signif icant

relat ionship between sociotropic cl in ical features and sociotropy scores , as wel l as a

relat ionship between autonomy cl inical features and autonomy scores . T h e authors

14

pointed out that these results lend good support to the validity of the sociotropy /

autonomy constructs and are in line with the f indings of research that has l inked

speci f ic life events to the exper ience of depress ion .

Therapy Outcome.

R e s e a r c h e r s have a lso examined the relat ionship of sociotropy and autonomy to

the outcome of psychotherapy. Blatt and his co l leagues have reported a ser ies of

studies examin ing psychotherapy outcome and personal i ty factors, and another group

of workers examined response to medicat ion in these personal i ty styles (Blatt, 1995;

Blatt, Zuroff, Qu in lan , & Pi lkonis, 1996; P e s e l o w et a l . , 1992). Uti l izing data col lected

in the N IMH Treatment of Depress ion Col laborat ive R e s e a r c h Program, Blatt examined

the role of perfect ionism (self-criticism) on both the treatment a l l iance and ultimate

outcome in therapy (Blatt, 1995; Blatt et a l . , 1996). Per fect ionism was negatively

assoc ia ted with therapeut ic response in brief treatment for depress ion (Blatt 1995).

Moreover, subsequent analys is suggested that the quality of the therapeutic a l l iance in

perfectionist ic individuals was a signif icant predictor of improvement (Blatt et a l . , 1996).

These f indings have led Blatt to conc lude that therapists need to be attentive to

interpersonal aspec ts of the individual, and not al low d iagnos is to dominate dec is ions

about how therapy is conducted.

Another study which examined these personal i ty factors and treatment response

was performed by P e s e l o w and co l leagues (Pese low et a l . , 1992). T h e s e authors

examined response to ant idepressant medicat ion in a large sample of outpatients who

15

were either sociotropic, autonomous, high on both these d imensions, or low on both

d imensions. They identif ied a differential response rate in the autonomous vs. the

sociotropic groups: 7 4 % of the patients in the autonomous group responded to

medicat ion whe reas only 3 9 % of patients in the sociotropic group responded to

ant idepressants. Interestingly, the sociotropy / autonomy distinction was a better

predictor of r esponse to medicat ion than the endogenous / non-endogenous distinction,

previously thought to be predict ive of response to ant i -depressants (Pese low et a l . ,

1992). T h e s e studies of personal i ty and treatment, a l though d iverse in their

approaches, do sugges t that sociotropy and autonomy have signif icant cl inical

implications. Moreover , this research highlights the potential role of interpersonal

factors in therapy with these personal i ty types (Blatt, 1995; Blatt et a l . , 1996).

Psycho log ica l Corre lates.

A quest ion left unanswered to this point concerns sociotropy and autonomy

themselves: what, speci f ical ly, are these constructs all about? O n e pathway toward

understanding the mean ing of these personali ty features is to examine their

relat ionship to es tab l i shed psycholog ica l measures. Th is research has examined

numerous var iab les, with greater or lesser theoretical justif ication for doing so. A s a

direct result, conc lus ions are often c louded by inconsistent or contradictory results from

one study to the next. Nonethe less , this research provides va luab le insight into these

personali ty features and their meaning in depress ion.

16

S o m e researchers have examined the assoc iat ion between sociotropy,

autonomy, and es tab l ished measures of psychopathology. R i ley and M c C r a n i e (1990)

found that self-cri t icism was assoc ia ted with overt cognit ive and affective s igns of

depress ion, and that dependency was assoc ia ted with subt le behavioural

manifestat ions such as passivity. Go ldberg and his co l leagues examined the

relat ionship of personal i ty d isorders to sociotropy and autonomy (Goldberg, Sega l ,

Ve l la , & Shaw, 1989). Au tonomous individuals were found to have e levated scores on

the negat iv ism, anxiety, and dysthymia subsca les of the M C M I (Mil lon, 1981) whereas

sociotropic individuals were found to have elevated scores on the avoidant, dependent,

anxiety, and dysthymia subsca les . Another group of researchers has attempted to

relate sociotropy and autonomy to personal i ty disorders descr ibed in DSM-III-R

(Ouimette, Kle in, Anderson , R i so , & L izardi , 1994). Sociot ropy w a s most c losely

assoc ia ted with histrionic, dependent , and avoidant d iagnoses ; however, autonomy

showed a more complex pattern of assoc ia t ions with a number of cont inua, including,

sch izo id , obsess ive-compu ls ive , pass ive-aggress ive , and ant isoc ia l features. T h e s e

studies suggest that sociotropy and autonomy do show relat ionships with var ious

measures of psychopathology, but are not redundant with such measures . Moreover,

they reinforce the complexity of the autonomy construct.

Sociot ropy and autonomy may a lso be related to measures which have evolved

from long-standing tradit ions within personal i ty psychology. Thus , these consructs

have been examined in the context of trait measures of the "big f ive" personal i ty

var iab les (i.e. neurot ic ism, extraversion, agreeab leness , consc ien t iousness , and

openness to exper ience; Cappe l i ez , 1993; Mongra in 1993; Zuroff, 1994). Sociotropy

was posit ively correlated with neurot ic ism, agreeab leness , extraversion, and lack of

openness . Autonomy had posi t ive correlat ions with consc ient iousness, neurot ic ism,

lack of extraversion, and lack of ag reeab leness . Thus, in terms of the "big f ive" factors

of personal i ty, sociotropic indiv iduals are anxious, vulnerable, and warm, whereas

autonomous individuals are dep ressed , pass ive , lacking in posit ive emot ion, and

conscient ious.

Taken together, studies examin ing the relat ionship of sociotropy and autonomy

to estab l ished measures of psychopatho logy and personali ty suggest that sociotropic

individuals can be descr ibed as dependent , anxious, pass ive, warm, agreeab le , and

non-asser t ive. Au tonomous indiv iduals tend toward hostility, passivity, negat ive affect,

irritability, and insensitivity to others. Certa in ly one aspect of these descr iptors which is

very striking is the degree to which these character ist ics fall into the interpersonal

domain. Indeed, these empir ical f indings are entirely congruent with theoret ical

descr ipt ions of sociotropy and autonomy which highlight interpersonal behaviours in

such individuals. In the next sect ion exist ing interpersonal models of depress ion are

outl ined and a potential integration of sociotropy and autonomy is descr ibed.

Interpersonal Models of Depression

The role of interpersonal factors in depress ion has rece ived cons iderab le

theoretical and empir ical attention. At least three different models concern ing the

interpersonal behaviour of dep ressed individuals have been descr ibed. T h e first

18

explicit model that l inked interpersonal factors with depress ion evo lved from the

behavioura l perspect ive of Lewinsohn and his co l leagues. T h e main tenet of this

model is that dep ressed individuals d isplay a lack of soc ia l ski l ls both prior to and

during their depress ive ep isode (Libet & Lewinsohn, 1973). Another perspect ive on

interpersonal factors in depress ion comes from Beck 's cognit ive theory of depress ion .

Th is theory sugges ts that depress ion is c a u s e d by negat ive schemata which b ias and

distort a n individual 's percept ions of external events. Depress ives are thus subject to a

variety of errors in thinking which lead them to negat ive evaluat ions of themse lves and

their re lat ionships (Beck, 1983).

A third and more comprehens ive interpersonal model of depress ion focuses on

the response of others to the depress ive, as wel l as the consequences of others'

responses for the depress ive . C o y n e (1976) descr ibed a cycl ical p rocess in which the

behaviour of the depress ive elicits, over time, rejection from others. Coyne ' s model

postulates that the dep ressed person is motivated to seek out reassurance and

sympathy from others. T h o s e in the environment of the individual respond to such

concerns directly. That is, they are support ive, helpful, and reassur ing. However, any

support ive behav iour is simply met with further demands for reassurance. Th is

communicates to the other that his or her response has not succeeded in the intended

effect of assuag ing the depressed person's request for help. Wha t fol lows is a ser ies

of amb iguous communicat ions. The non-depressed other, now frustrated by the lack of

s u c c e s s of prev ious reassur ing behaviour and the continuing demands of the

depress ive , may begin communicat ing at two, very different, levels. They may express

support ive content but do so in a rejecting tone, or s ignal rejection with non-verbal

cues . The depress ive , accurately perce iv ing the equivocal communicat ions of the

other, becomes even more eager to test the non-depressed person and d isp lays more

reassurance-seek ing to test the relat ionship. At this point, it becomes highly likely that

the other will d i sengage from the dep ressed person. Th is confirms the depress ives '

susp ic ion about the other's non-genuine communicat ion, and p lunges them into greater

d ist ress at the loss of an important soc ia l relat ionship. Such a rejection completes the

feedback loop to the individual who now f inds him or herself in a "depress ive spi ra l "

(Coyne, 1976a, p. 29).

Depress ion and Interpersonal Reject ion

Unl ike behavioural models which postulate a s imple skil l deficit in depress ion,

Coyne ' s model involves a progression of events that are perhaps more difficult to

capture with standard experimental methodology. Resea rche rs have largely focused

on the component of the model that is most easi ly tested; that depressed persons elicit

rejection from others. However, a comprehens ive review of studies examining rejection

of depress ives was unable to reach definit ive conc lus ions. Al though a rejection effect

is somet imes found, many other t imes it is not (Marcus & Nardone, 1992). A meta­

ana lys is wh ich examined the literature on the interactional theory conc luded that the

depression-re ject ion correlation was .26 ac ross all variet ies of studies, though varying

from .18 to .61 depend ing on methodological factors (Segr in & Dil lard, 1992). T h e s e

20

reviewers conc luded that methodological factors p layed an important role in

determining whether or not depress ion is assoc ia ted with rejection.

T h e strongest f indings for rejection are made when long term interactions or

relat ionships are s tudied (Marcus & Nardone, 1992; Segr in & Dil lard, 1992). However,

some of this work c a n be cr i t ic ized on methodological grounds. First, studies of long

term interactions usual ly involve married coup les with a dep ressed member. C o u p l e s

identified for s u c h studies are likely to have an extensive previous history of negat ive

interactions and a deteriorated relationship. Moreover, longitudinal studies such as

these do not capture other very important events, including d ivorces and separat ions in

couples with a d e p r e s s e d member. Final ly, these studies usual ly do not measure

rejection, per se , but rather satisfact ion in the relat ionship (Marcus & Nardone, 1992).

Another parad igm investigating long term interactions has avo ided some of

these methodologica l confounds in a novel manner. Jo iner and his co l leagues have

studied co l lege roommates, who are unacquainted at the outset of the studies and who

are fol lowed longitudinal ly over the course of the academic year (Joiner & Metalsky,

1995; Joiner, A l fano, & Metalsky, 1992). Th is avoids the problems assoc ia ted with a

history of diff iculties, and at the same time al lows interpersonal patterns to deve lop

naturalist ically over time. Th is research has found that depress ion , and more

specif ical ly, reassu rance-seek ing by depressed individuals, predicts rejection by

roommates. In addit ion, dep ressed individuals induced more negative moods in their

roommates (Joiner & Metalsky, 1995; Joiner, et a l . , 1992). T h e s e studies offer strong

21

support for Coyne ' s model regarding rejection, and they a lso support the contention

that reassurance-seek ing may play a role in the depress ive interpersonal cyc le.

A final paradigm evaluat ing the interpersonal model involves measur ing rejection

of depressed persons after interactions with a stranger. Usual ly , these studies involve

having dep ressed targets interact with non-depressed part icipants for a brief period

and then a s s e s s i n g rejection of the targets. Th is paradigm has rece ived some crit icism

for examining C o y n e ' s model on a too simplist ic level (Doerf ler & Chap l in , 1985).

T h e s e authors point out that the model is both cycl ical and temporal ; the rejection

process may take t ime to deve lop and is most likely to emerge in ongoing relat ionships.

However, C o y n e (1985) has argued that a short-term parad igm does not suffer from

the b iases of a prev ious negat ive relat ionship history, and that this approach may

capture the incept ion of an emerg ing socia l phenomenon.

The first study to examine rejection in this paradigm w a s carr ied out by Coyne

(1976b). In that study, dep ressed female outpatients, non -depressed outpatients, and

community volunteers were recruited to speak to female undergraduates on the

telephone. Both targets and participants were told that the experiment was a study of

the acquaintance p rocess and that they would be convers ing for twenty minutes.

Part ic ipants were told to d i scuss whatever they wished, as long as this maintained

anonymity. C o y n e measured three participant responses to the conversat ion: induction

of negat ive mood, evaluat ion of targets on bipolar adject ives (e.g., good-bad), and

wi l l ingness to engage in future interactions. Al l three measu res conf i rmed Coyne ' s

hypotheses. Part ic ipants were more hostile, angry, and dep ressed after speak ing to

22

the depressed patients, and eva luated the depressed patients more negat ively than the

community volunteers. Moreover , part icipants were less wil l ing to engage in future

activit ies with depressed indiv iduals, specif ical ly, ask ing them for adv ice and sitting

next to them on a bus trip (Coyne, 1976b). Overal l , the results from this study offered

strong support for the interpersonal model .

S ince the publ icat ion of this initial work, a number of researchers have

attempted repl icat ions and modif icat ions of this study (Borden & Baum, 1987; Dobson ,

1989; Got l ib & Rob inson , 1982; K ing & Heller, 1984; McN ie l , Arkowitz, & Pri tchard,

1987; Paddock & Nowicki 1986; Segr in , 1993; Strack & Coyne , 1983). Different

populat ions have been used to test the model ; some researchers have cont inued to

use depressed patients (e.g. K ing & Hel ler, 1984; McN ie l , et a l . , 1987) whereas others

have used dysphor ic student targets to a s s e s s levels of rejection by others (e.g.

Borden & Baum, 1987; Dobson , 1989; Strack & Coyne , 1983). Methodolog ica l

variat ions have included use of face to face conversat ions (e.g., Borden & Baum, 1987,

McN ie l et a l . , 1987) and te lephone conversat ions (e.g. K ing & Hel ler, 1984). Final ly,

some studies have a s s e s s e d others' percept ions and actual behav iours thought to be

related to depress ion (Gotl ib & Rob inson , 1982; Paddock & Nowicki , 1986). Despi te

the numerous methodological changes and refinements, these studies have all

invest igated the bas ic rejection tenet of Coyne ' s model.

Wha t conc lus ions can be drawn from the results of these s tud ies? The most

important issue, at least in terms of theoret ical implications for Coyne ' s model , is

whether or not dep ressed indiv iduals elicit rejection in brief interactions with strangers.

23

Only two of the ten studies reviewed here found a rejection effect (Coyne, 1976b;

Strack & C o y n e , 1983). T h e s e two studies used different populat ions; one examined

dep ressed outpatients and the other dysphor ic undergraduates. In one study,

part icipants and targets interacted face to face. In the other study, the interaction took

p lace over the te lephone. Thus, these two studies were somewhat d iverse in their

approaches yet they are the only studies to show a rejection effect. A second quest ion

in this research concerns negat ive mood induction. Th is var iable was examined by all

ten studies. Aga in , the same two studies that found a rejection effect, found signif icant

di f ferences in the mood of part icipants who interacted with depressed targets (Coyne

1976b, Strack & Coyne , 1983).

A third i ssue has to do with others' soc ia l percept ions of dep ressed individuals

after a brief interaction. S e v e n of the ten studies located used a measure of soc ia l

percept ion of target individuals. The Impact M e s s a g e Inventory (IMI; Kiesler, et a l . ,

1976) was used in three studies (Borden & Baum, 1987; Dobson , 1989; McN ie l et a l . ,

1987) and sets of bipolar adject ives a s s e s s i n g a positivity / negativity d imens ion were

used in four s tudies (Coyne, 1976b; K ing & Heller, 1984; Rosenblat t & Greenberg ,

1991; St rack & C o y n e , 1983). O n e study us ing the Impact M e s s a g e Inventory found

that d e p r e s s e d individuals were seen as less soc iab le /warm than their non-depressed

counterparts; however these depressed participants were a lso highly anx ious (Dobson,

1989). Leve l of anxiety may have accounted for the dif ferences in percept ion, a

possibi l i ty wh ich becomes more likely when one cons iders that the other two studies

using the IMI had negat ive f indings. Of the fou rs tud ies using bipolar rating sca les ,

24

three found that dep ressed individuals were perce ived more negatively (Coyne, 1976b;

King & Heller, 1984; Strack & Coyne , 1983).

The final i ssue examined in these studies concerns the behaviour of dep ressed

targets and the behaviour of participants with whom they interact. Th is issue was

examined by four of these ten studies. Two studies examined behaviour in both

participants and targets (Borden & Baum, 1987; Got l ib & Rob inson , 1982), one study

examined the behav iour of part icipants only (McNie l , et a l . , 1987), and one study

examined the behaviour of targets only (Paddock & Nowick i , 1986). One of the three

studies which examined behavioural var iab les in part icipants found dif ferences

between individuals who interacted with dep ressed versus non-depressed targets

(Gotlib & Rob inson , 1982). Part ic ipants who interacted with depressed individuals

demonstrated less support, fewer posit ive conversat ion-maintenance statements,

smi led less, and were less pleasant. Two of the three studies that examined behaviour

of dep ressed targets found dif ferences between these individuals and their non-

depressed counterparts (Gotlib & Rob inson , 1982; P a d d o c k & Nowicki , 1986).

Dep ressed targets showed less direct support to others, less conversat ion-

maintenance, and engaged in more negat ive non-verbal behaviour and more negative

paral inguist ic behaviours.

Overa l l , f indings in each of these categor ies of var iab les (rejection, socia l

perception, and behaviour) are not particularly robust. T h e postulate that depressed

persons are rejected is least supported by the results of these studies. There is

somewhat more, but by no means overwhelming, ev idence that depressed individuals

25

are perce ived more negat ively and that there are behavioural d i f ferences in dyadic

interactions in wh ich a dep ressed member is present. Overa l l , the studies reviewed

showed signif icant d i f ferences on less than one third of the dependent var iables

examined. Wha t factor(s) might account for these equivocal f ind ings?

Severa l methodologica l possibi l i t ies have been sugges ted to explain these

inconsistencies. First, reviewers have pointed out that "depress ion" is def ined

inconsistently from one study to the next (Marcus & Nardone, 1992). Th is results in

studies using different, somet imes quest ionable, measures of depress ion . Future

studies require wel l -def ined criteria for depress ion and psychometr ical ly sound

measures of depress ive symptomatology. Another possibi l i ty concerns the types of

interactions general ly studied. It has been suggested that the typical "getting

acquain ted" tasks require only wel l -pract iced and rudimentary soc ia l ski l ls that could be

reasonably wel l -handled by almost anyone (Marcus & Nardone, 1992). Future studies

would be better se rved by us ing a task that requires greater depth and less pract iced

modes of interaction.

Final ly, it has been sugges ted that heterogeneity in depress ion may account for

the inconsistent results in this research literature (Marcus & Nardone, 1992). In other

words, individual d i f ferences in depress ives may obscure the hypothes ized rejection

effect sugges ted by C o y n e ' s model . Discover ing the d imens ions important to the

process of rejection shou ld clarify d iscrepant empir ical f indings and offer a further

elaborat ion of Coyne ' s model . Marcus and Nardone suggest that the sociotropy /

autonomy distinction may play an important role in interpersonal behaviour and "elicit

26

different types and degrees or rejection from others" (p. 443). Failure to consider this

distinction, and thus homogenizing depression, could result in large within-group

variances that might obscure any distinction between depressed and non-depressed

individuals. The following section explores both theoretical and empirical interpersonal

concomitants of sociotropy and autonomy, and then examines their implications for

Coyne's interpersonal model.

Interpersonal Concomitants of Sociotropy and Autonomy

Beck (1983) describes many attributes of the sociotropic or autonomous

depressive that fall within the interpersonal domain. The sociotropic person is

described as fearful of rejection, needy, and eager to give up control to others. Such

persons are thought to require continuous reassurance and to find such reassurance to

be gratifying. On the other hand, the autonomous individual's concerns with

preserving his or her independence, and defining his or her own boundaries also

influences the person's interactions with others. In particular, Beck noted that

autonomous individuals are not susceptible to social feedback, are decisive, and

reluctant to give up control to others. They are also described as oblivious to the effect

of their actions on other people and as non-empathic (Beck, 1983).

Beck theorized that these attributes were pervasive and highly salient to others.

He believed them to be exhibited both in the general social world and in the

patient/therapist relationship (Beck, 1983). In fact, these characteristics were thought

to have such a significant impact on the behaviour of the individual and subsequently in

27

the patient/therapist relat ionship, that c l in ic ians were adv ised to tailor therapy for e a c h

subtype (Beck, 1983). Autonomous indiv iduals were thought to require a col laborat ive

relat ionship in which they are a l lowed to set goals . Beck suggested that the patient's

underly ing rigidity shou ld only be examined in the latter s tage of therapy, with a focus

on creat ing an internal sense of f reedom. For sociotropic individuals, it was

recommended that therapy be more structured and that helping and gu idance be

emphas ized . Explanat ions and clari f icat ions by the therapist were thought to elicit a

posit ive response (Beck, 1983). T h e dominant themes in therapy would appear to be

"col laborat ion" for autonomous individuals and "gu idance" for sociotropic individuals.

T h e writ ings of Blatt offer similar interpersonal descr ipt ions. In anacl i t ic

depress ion , the individual was descr ibed as having an intense need to be cared for,

protected, and soothed (Blatt, 1974). There are "cr ies for love" (p. 116) and fears of

abandonment by others. Soc ia l support can result in a temporary sense of comfort, but

all sat isfact ion is s e e n as coming from external sources. Th is dependence on others

results in great apprehens ion about being abandoned , leading to an inability to express

anger or d issat isfact ion for fear of losing this source of sat isfact ion. Introjective

depress ion was character ized by strivings toward self-definit ion and autonomy. T h e

relat ionships of these individuals were character ized by ambiva lence or hosti le feel ings

toward others. T h e s e interpersonal character izat ions of Blatt and Beck, despi te their

evolution from two distinct theoretical approaches , share many attributes in common.

R e s e a r c h sugges ts that the interpersonal concomitants of the two personali ty features

are indeed simi lar (Alden & Biel ing, 1996). The fol lowing sect ion provides a summary

28

of this research and other studies examining interpersonal correlates of sociotropy and

autonomy.

R e s e a r c h Ev idence

A growing number of studies have examined the interpersonal correlates of

dependency and self-cr i t ic ism (Zuroff & deLorimier, 1989; Zuroff & Fitzpatrick, 1995;

Zuroff & Stot land, 1995). In one study, dating preferences of undergraduate women

who var ied in dependency and self-crit icism were examined (Zuroff & deLorimier,

1989). D e p e n d e n c y was assoc ia ted with desir ing a partner with a high need for

intimacy and a low level of masculinity. Self-cr i t ic ism was assoc ia ted with desir ing a

partner with high need for achievement and high masculinity. In terms of relat ionship

harmony, self-cri t ical women were less satisf ied with their current relat ionship than

were dependent w o m e n (Zuroff & deLorimier, 1989). In another study, daily

interactions of indiv iduals who differed in dependency and self-cri t icism were

examined. D e p e n d e n c y w a s assoc ia ted with more frequent and intimate interactions,

whereas self-cr i t ic ism was assoc ia ted with the lack of p leasure in socia l interactions

(Zuroff & Stot land, 1995). Final ly, adult attachment sty les of individuals who differed in

dependency and self-cr i t ic ism have been examined (Zuroff & Fitzpatrick, 1995).

Dependency w a s assoc ia ted with an anxious attachment style, whereas self-crit icism

was assoc ia ted with avoidant attachment concerns. Together, these results lend good

support to the notion that sociotropic and autonomous individuals may have different

2 9

interpersonal concerns and needs in both intimate relationships and other social

contacts.

In order to further elucidate the interpersonal correlates of sociotropy and

autonomy, Alden and Bieling (1996) sought to examine sociotropy and autonomy in an

established interpersonal framework. Undergraduate volunteers completed measures

of sociotropy and autonomy and the IIP-C, a circumplex-based measure of self-

reported interpersonal problems (Alden, Pincus, & Wiggins, 1990). The results

demonstrated that sociotropy was positioned in the octant of the interpersonal problem

circle which reflects a combination of non-assertive and overly-nurturant interpersonal

behaviour, generally labeled exploitable (Alden et al., 1990). Thus, high scores on

sociotropy were associated with self-reported problems expressing anger towards

others, asserting one's needs, and being taken advantage of by others. Autonomy

was positioned in the Cold octant of the interpersonal problem circle. Thus, autonomy

was associated with self-reported problems making long-term commitments, problems

experiencing love and affection toward others, and distancing oneself from other

people. Overall, this study also supported the notion that distinct interpersonal patterns

are associated with sociotropy and autonomy. However, the findings are limited by

exclusive reliance on self-report measures.

One study has examined the reactions of sociotropic and autonomous

dysphorics to a controlled social interaction (Bieling & Alden, 1996). In this study,

undergraduate women were selected for a dysphoric (BDI > 12) sociotropic, dysphoric

autonomous, non-dysphoric (BDI < 7) sociotropic group, and a non-dysphoric

30

autonomous group. Subjects part ic ipated in a therapy ana logue interaction with a

confederate who acted either in a control l ing or non-directive fashion throughout the

interaction. Resu l ts showed c lear interpersonal di f ferences between the two

personal i ty style groups. The sociot ropic and autonomous part icipants had different

cognit ive emot ional concerns dur ing the interpersonal task, perce ived the s a m e partner

behaviour differently, and differed in l iking for their partner. Speci f ical ly , sociotropic

part icipants were more motivated toward maintaining interpersonal re la tedness during

the interaction, including getting a long with and p leas ing their partner. They a lso

perce ived their partner as more approv ing and support ive. On the other hand,

autonomous individuals were more l ikely to be motivated by self-definit ional concerns,

such as their own objectives in the task, control l ing the interaction, and separat ing

themselves from their partner.

Most relevant to Coyne ' s model , results from this study showed di f ferences in

others' responses to each of the groups. The confederates, who were bl ind to the

part icipant's condit ion, exp ressed less l iking for the autonomous dysphor ic group.

Moreover , addit ional ana lyses sugges ted that a combinat ion of dysphor ia and

personal i ty resulted in less liking. Th is f inding corroborates a similar study (Zuroff,

Moskowi tz , W ie lgus , & Powers , 1983) examin ing dif ferences between dependent and

self-crit ical individuals participating in an interpersonal task. In this study, male and

female co l lege students part icipated in a variety of tasks with male and female peers

(Zuroff et a l . , 1983). There was a signif icant negative relat ionship between task

31

leadership and dependency in men. Moreover , there w a s a trend toward a negat ive

assoc ia t ion between self-cri t icism and likability in women.

Resu l t s from these studies do suggest that there are two distinct interpersonal

patterns assoc ia ted with sociotropy and autonomy. Sociot ropic and autonomous

individuals differ in their self-reported interpersonal problems, romantic preferences,

attachment sty les, react ions to an interpersonal task, and in the responses they

engender in others. Sociot ropic persons, wish ing to p lease others and concerned

about approva l , strive to get a long well with others. They bel ieve that they s u c c e e d in

this regard, perce iv ing their partner as p leased with them and as evaluat ing them

posit ively. Indeed, they are judged to be more l ikable than their autonomous

counterparts. A different p rocess seems to occur in autonomous individuals.

Conce rned about their own goa ls and about maintaining control of their environment,

they s e e others a s potentially avers ive intruders who may seek excess ive inf luence

over them. Thei r behaviour culminates in expected rejection, confirming their or iginal

distrust of others and tendencies to d is tance themselves from others. One would

expect that this p rocess might lead to inc reased dysphor ic affect.

The Present Study

T h e s tud ies rev iewed above suggest that not all depress ives suffer similar levels

of rejection from others. T h e s e data, though limited, imply that autonomous

depress ives wou ld be particularly prone to elicit a negat ive react ion in others, whereas

their soc iot ropic counterparts may be less suscept ib le to such negat ive responses .

This difference, if found to exist in dep ressed patients, may indeed constitute an

explanat ion for prev ious discrepant f indings. If individuals do respond differently to

sociotropic and au tonomous depressed individuals, the interpersonal model of

depress ion p roposed by C o y n e (1976a) may require refinement.

To my knowledge, no studies have examined the responses of others to

sociotropic and au tonomous depressed patients. P rev ious research examining

responses to sociotropy and autonomy has studied either non-depressed col lege

students (Zuroff et a l . , 1983) or dysphor ic students (Biel ing & A lden , 1996). There is.

an ongoing debate over the adequacy of student samp les as ana logues for

depress ives. W h e r e a s some have argued that most f indings in ana logue samples do

appear to genera l ize to cl in ical samp les (Vredenburg, Flett, & Krames, 1993), there are

fundamental d i f ferences between a d iagnos is of depress ion and self-reported

symptoms of depress ion in students (Coyne, 1994). Popula t ions of students are clearly

distinct in any number of demograph ic ways from a community sample of cl inical ly

depressed indiv iduals (Coyne, 1994). W h e r e a s the former populat ion may be prone to

transient ep i sodes of d ist ress that are al leviated by the diminution of a stressor, the

latter suffers from ongo ing depress ive disorders that deep ly impair their psychosoc ia l

functioning. There is no debate that cl inical ly dep ressed individuals are a preferable

sample for research in depress ion . Clear ly , the di f ferences between dysphor ic

students and indiv iduals with a depress ive disorder necess i ta te studying the latter

sample.

33

The current study ut i l ized a cl in ical populat ion and a wel l es tab l ished measure of

symptom severity. In addit ion, fol lowing the suggest ions of M a r c u s and Nardone

(1992), a more involving, novel conversat ional task was used to better engage all

part icipants in a col laborat ive task. The task used for the study w a s a therapy

ana logue in which a confederate he lped participants toward a def ined goal , and was

similar to one used in prev ious research (Biel ing & A lden , 1996). Moreover , such a

task is consistent with the theoret ical descr ipt ions provided by B e c k (1983) that outline

sociotropic and autonomous di f ferences in response to therapy situations.

In order to provide a level of compar ison with past research , measures similar to

those employed in prev ious studies were uti l ized. Thus, three c l a s s e s of var iables

were a s s e s s e d . First, level of rejection of targets was a s s e s s e d after the interaction

task. S e c o n d , soc ia l percept ions of others that were relevant to this type of

interpersonal task and sociotropy and autonomy were examined. Third, both verbal

and non-verbal behav iours of targets relevant to likability were coded by independent

raters. T h e s e behaviours, which were der ived from testing in a pilot sample (descr ibed

in the Dependent M e a s u r e s sect ion, p. 45), included attention to the task, difficulty in

schedul ing p leasant events, appear ing re laxed and calm, s p e e c h intonation, smil ing,

cheerful appearance , suggest ing opt ions for schedul ing, res is tance to suggest ions of

assistant, and sel f -d isc losure.

34

Hypotheses

Regarding rejection of target individuals, two specific hypotheses were made.

First, I predicted that autonomy would be associated with rejection. This hypothesis is

consistent with theoretical depictions, studies that have found an association between

autonomy and rejection, and self report findings (Alden & Bieling, 1996; Beck, 1983;

Bieling & Alden, 1996; Zuroff et al., 1983). Second, I predicted that depressed

individuals would be rejected to a greater extent than non-depressed individuals. Such

a prediction is consistent with Coyne's model and at least some past research (Coyne,

1976a, 1976b; Strack & Coyne, 1983). With regard to sociotropy, previous findings

suggest that it is associated with greater self-reported warmth, but not necessarily

rejection (Alden & Bieling, 1996; Bieling & Alden, In press, Zuroff et al., 1983).

However, some theorists have suggested that sociotropy may be associated with

higher levels of support seeking, and excessive reliance on others that does, indeed,

result in rejection (Blatt & Zuroff, 1992). Given these contradictory views, no specific

prediction was made for sociotropy and rejection of target individuals.

In terms of others' social perceptions, two hypotheses were made based on

theoretical descriptions and previous research (Alden & Bieling, 1996; Beck, 1983;

Bieling & Alden, In press). First, autonomy was hypothesized to be associated with

insensitivity to others, negating other's suggestions, being action-oriented, and having

more rigid goals in the task. Second, I predicted that sociotropy would be associated

with striving to garner others' approval, trying to please others, deferring to other's

35

dec is ions, and desi r ing reassurance. No speci f ic predict ions were made about how

depressed targets, as a group, would be rated on the soc ia l percept ion measures .

Wi th regard to behaviours in the task, two speci f ic predict ions were made. First,

autonomy was predicted to be related to fewer posit ive soc ia l behaviours in the

interaction task. S e c o n d , I hypothes ized that dep ressed individuals would engage in

fewer posi t ive soc ia l behaviours than non-depressed individuals in the interaction.

No speci f ic predict ion w a s made about assoc ia t ions of sociotropy with posit ive or

negative interaction behaviours.

M e t h o d

Overview

D e p r e s s e d outpatients and non-depressed controls were a s s e s s e d on

sociotropy and autonomy and then part icipated in a brief task in which a research

assistant he lped them plan adapt ive life changes . Upon complet ion of the interaction,

participants and the confederates completed measures in which they rated one another

on a variety of d imens ions. Al l part icipants were then administered a structured

diagnost ic interview, and debriefed.

Participants

Depressed Samp le .

A total of 41 (29 women, 12 men) part icipants were recruited from two sources :

an outpatient wai t ing list at the Heal th Psycho logy Cl in ic , Vancouve r Hospital , U B C ,

36

and a posting at the Mood Disorders Clinic in the Department of Psychiatry, Vancouver

Hospital, UBC (see Table 1 for details). Potential wait-list participants were contacted

by telephone and were told about the details of the study. A laboratory appointment

was booked if they were interested in taking part in the study. Potential volunteers who

telephoned based on the posting were asked about their current mood problems, and

were then invited to participate if they seemed initially appropriate.

Table 1 - Recruitment Sources

Source Total Potential Not willing or Screened out Total in Study

Volunteers able to

Participate

(at telephone

contact)

by interview or

BDI cutoff.

Depressives

Waiting List 42 4 4 34

Clinic Posting 14 5 2 7

Total 41

Control

Community 51 2 8 41

Postings Total 41

37

Inclusion Cri ter ia for Depress ives .

Part ic ipants were inc luded in the final sample if they: (1) met DSM-III-R criteria

for either a current major depress ive ep isode or dysthymia accord ing to the S C I D

interview, and (2) obtained a minimum score of 14 on the Beck Depress ion Inventory.

U s e of such a BDI cut-off is common in research on interpersonal factors in depress ion

and the minimum score used here is similar to one used in prev ious research (e.g.,

K ing & Heller, 1984, McN ie l , et a l . , 1987). In order for the samp le to be representative,

part icipants who met criteria for a concurrent anxiety d isorder were a lso included in the

sample, as were individuals currently using ant i -depressant medicat ion. Individuals

with a d iagnos is of bipolar d isorder were not included. Deta i led character ist ics of the

sample are d isp layed in Tab le 2.

Control Sample .

A total of 41 (29 women, 12 men) participants constituted the control sample.

T h e s e individuals were recruited from posters p laced in community centers and

l ibraries in the Vancouve r a rea . Potential volunteers were asked to contact the

laboratory about a study on conversat ion. Individuals were sc reened over the

te lephone for potential difficulties with mood, or history of such problems. Individuals

without such a history were invited to participate and were schedu led for an

appointment.

Tab le 2 - S a m p l e Character is t ics

38

Var iab le Depress i ve Control x ! t

A g e 37 .34 (10 .06 ) 35 .10(6 .81) 1.18

Marital Status

S ing le

Marr ied

Divorced

4 4 %

3 7 %

1 9 %

6 3 %

2 5 %

1 2 %

3.81

Educat ion (Years) 14.37 (2.44) 15.07 (1.93) -1.46

Employment Status 1.71

Work ing 7 1 % 8 2 %

Not Work ing 2 9 % 18%

S C I D d iagnos is (frequency) M D E 6 1 %

Dysthymia 4 6 %

G A D 1 2 %

Soc ia l Phob ia 5 %

P a n i c D/O 5 %

O C D 2 %

History of any previous

psychiatr ic contact

8 3 % 1 0 % 44 .14**

Ant idepressant U s a g e 4 4 % 0% 23.06**

BDI Sco re 23.44 (9.42) 3.95 (2.46) 12.82**

0 <.001

39

Inclusion Criteria for Control Group.

Volunteers were included in the final sample if they: (1) did not meet a diagnosis

for either a mood or anxiety disorder, and (2) reported few depression symptoms (BDI <

07).

Experimenters

The primary experimenter was a male graduate student, with prior clinical

training in assessment and diagnosis. The experimenter was responsible for carrying

out the experimental procedures, including giving instructions for the task,

administration of measures, screening interviews, and debriefing procedures. The

experimenter was blind with regard to participants' scores on the BDI, sociotropy, or

autonomy. A second graduate student with previous training in structured clinical

interviews served as a reliability rater on the diagnostic interviews.

Confederates

Two graduate student research assistants served as confederates. A male

confederate interacted with male participants and a female confederate interacted with

female participants. Both confederates were trained to enact the same scripted

behaviours in the interaction (see Appendix A for script), and the duration of the

interaction was monitored. Also, deviations from the script were recorded by the

experiementer to ensure that the task was carried out consistently by the confederates.

40

Procedure

Upon arrival at the laboratory, participants completed consent forms, a measure

of depress ive symptomatology, and a measure of sociotropy and autonomy.

Part ic ipants were then g iven the fol lowing instructions:

Th is study concerns helping styles and the way in which people work

together. W h a t we would like you to do today is to talk with a research

assistant, and the two of you will be working on p leasant events

schedul ing. Now, many people are very busy in their l ives, and don't

a lways do as many fun or interesting things as they might like to. The

idea for the task today is to take a look at your schedu le , and the kinds of

things you like to do. T h e n you'l l work on adding some things in that

might be fun or p leasant for you. The overal l idea is to reduce your level

of s t ress and increase your genera l sense of sat isfact ion.

Part ic ipants were then informed that the interaction would be v ideotaped and that the

experimenter would observe the conversat ion from behind a one-way mirror. After

complet ion of the conversat ion, the confederate and participant were separated to

complete the dependent measures .

After complet ing their sca les , participants were interviewed using the Structured

Cl in ical Interview for DSM-III-R (SCID; Spitzer, Wi l l iams, G ibbon , & First, 1990). Upon

complet ion of the S C I D , all part icipants were told the nature of the experiment and

41

received information about the purpose of the study. A n opportunity for quest ions and

feedback was g iven, and part icipants were then thanked and d ismissed .

Experimental Task

The task was based on the p leasant events schedul ing techniques deve loped by

Lewinsohn for behavioural treatment of depress ion (Appendix B, MacPh i l l amy &

Lewinsohn, 1971). Confedera tes worked with participants for the purpose of creating

a schedu le with inc reased f requency of p leasant events. Th is necess i ta ted

col laborat ion between part icipants and confederates toward a meaningful goal ;

part icipants were required to engage in act ive conversat ion, examine options, suggest

potential solut ions, and make personal ly relevant dec is ions. Th is type of task w a s

se lec ted because it s e e m e d likely to elicit a variety of interpersonal behaviours from

participants (Marcus & Nardone, 1992). A list of p leasant events and a weekly

schedu le were provided. The confederate then ass is ted part icipants in the select ion of

p leasant events and he lped them in fitting these to their schedu les .

Diagnostic and Classificatorv Measures

Structured Cl in ica l Interview for DSM-III-R (SCID; Spi tzer et a l . . 1990).

T h e S C I D is a d iagnost ic instrument that is consistent with criteria from the

Diagnost ic and Statist ical Manua l (DSM-III-R). The current study uti l ized the S C I D - N P

(Nonpatient edition) which a l lowed for the d iagnos is of all Ax is I d isorders including: (1)

mood syndromes (2) psychot ic sc reen ing (3) mood disorders (4) psychoact ive

42

subs tance use d isorders (5) anxiety d isorders (6) somatoform disorders (7) eat ing

d isorders and (8) adjustment disorder.

T h e S C I D w a s administered by an experimenter with prior cl inical exper ience

and training in administrat ion of structured interviews. In order to confirm d iagnoses

made by the experimenter, each interview was tape recorded and verif ied

independent ly for d iagnost ic validity by a second rater. The overal l inter-rater

agreement (kappa) for d iagnoses was .85.

Beck Depress ion Inventory (BDI: Beck. W a r d . Mende lson . Mock. & Erbauqh. 1961).

T h e Beck Depress ion Inventory (Appendix C) is an efficient and widely used

measure of depress ive symptomatology, appear ing in over 500 studies (Steer, Beck, &

Gar r i son , 1986). It is a 21-item sca le der ived from cl inical observat ions. T h e BDI has

been shown to have high reliability and stability and has a lso been demonstrated to

correlate highly with other self-report measures of depress ion and with c l in ic ians'

ratings of depress ion (Beck, Steer, & Garb in , 1988).

Pe rsona l Sty le Inventory. (PSI: Rob ins . Ladd, Welkowi tz . Blanev. Diaz. & Kutcher.

1994).

T h e PSI cons is ts of two 24-i tem sets, one measur ing autonomy and the other

sociotropy (Appendix D). Coeff icient a lpha for the autonomy sca le is .86, and for

sociotropy .90. Test-retest reliabil i t ies for a period of between 5 and 13 weeks were

.80 for the sociotropy sca le and .70 for the autonomy sca le (Robins et a l . , 1994). E a c h

43

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).

Dependent Measures

Desire for Future Interactions (Coyne. 1976b; Dobson, 1989: Segrin. 1993)

This scale consists of 9 items which reflect a desire to engage in future

interactions with the target person. Items were answered on a seven point Likert type

scale with affirmative (very much) and negative (not at all) anchors. Alpha reliabilities

for this scale have ranged between .87 and .96 in other studies (Dobson, 1989; Segrin,

1993). The current study used nine items related to future interactions. Items were: (1)

To what extent would you like to meet this person again, (2) To what extent would you

like to spend more time with them in the future, (3) To what extent would you like to

work with this person in the future, (4) To what extent would you like to sit next to this

person on a three hour bus trip, (5) To what extent would you like to invite this person

to visit you, (6) To what extent would you be likely to ask this person for advice, (7) To

what extent would you like sharing an apartment with this person, (8) To what extent

would you like eating lunch with him or her often, (9) To what extent would you like

having this person as a friend. For the present study, alpha reliabilities for the DFI

scale were .97 across the two confederates.

44

Confederate Percept ions of Part ic ipants.

Likert-type i tems were deve loped for rating part icipants intentions and perceived

behaviour. Content for these 21 items included the types of behaviours bel ieved to be

assoc ia ted with sociot ropy and autonomy that would be relevant in an interpersonal

cooperat ive help ing task (Appendix E). D imens ions relevant to sociotropy and

autonomy were der ived from empir ical and theoretical material (e.g., Beck, 1983; Blatt

& Zuroff, 1992; Rob ins et a l . , 1994), and the interpersonal concerns of sociotropic and

autonomous indiv iduals identif ied in a previous study (Biel ing and A lden , In p r e s s ) .

T h e m e s for behav iours thought to be assoc ia ted with sociotropy were: needing

others' help, requir ing reassurance , lack of assert ion, deferr ing to others to maintain a

posit ive relat ionship, fear of rejection, and being p leased with others' input. Together,

these items were des igned to a s s e s s the degree of the part icipants' interpersonal

deference and the a lpha reliability for this item set was .79. T h e m e s for behaviours

thought to be assoc ia ted with autonomy were: having speci f ic goa ls or standards, being

less suscept ib le to others ' feedback, being less sensi t ive to others' needs, being action

oriented, and be ing dogmat ic and decis ive. Together, these item a s s e s s e d the degree

of self-orientation the participant d isp layed and the a lpha reliabil ity for this sca le was

.88. Overa l l , the interpersonal deference scale, and the self-orientation sca les

d isp layed a moderate negat ive correlat ion ( r = -.39) suggest ing that these sca les were

not assess i ng redundant interpersonal percept ions.

45

Participant Behaviour.

The conversations were videotaped and the behaviour of participants rated

along a series of dimensions. The video camera was placed behind a one-way mirror,

and the camera was focused on the participant throughout the pleasant events task.

Behaviours related to liking of participants were identified rationally by three research

assistants in a sample of undergraduates (N=30). Videotapes of pilot participants were

subsequently coded by three trained raters along these rationally derived dimensions.

Nine types of behaviours were found to be significantly related to liking (r's > .40) and

could be coded with a high degree of inter-rater reliability (r's > .80) in the pilot sample.

These dimensions were used in the present study and included: (1) attentiveness (2)

ease of scheduling events, (3) smiling, (4) suggesting schedule options, (5) speech

inflection, (6) resisting the confederate's suggestions, (7) appearing cheerful, (8) being

disengaged from the task, and (9) self-disclosing.

Two undergraduate raters (who were not involved in the initial derivation of the

coding system) were trained using the videotapes of the undergraduate pilot

participants and found to have a high level of agreement on the nine items (all r's

greater than .80). The two raters then viewed each videotape and independently coded

each participant along these nine dimensions. The average inter-rater reliability

(pearson r) was .84 for the nine behaviours with a range of .73 to .92. Across the nine

behaviours, the two coders were within one point on the seven-point scale in 90% of

46

the c a s e s . Thus , the behaviour ratings made by each coder were averaged for each of

nine behaviours.

T h e s e nine ratings were then submitted to a principal components analys is .

A l though the ratio of the number of part icipants to the number of items was less than

ideal, this ana lys is suggested the p resence of a s ingle factor, with an e igenva lue of

4 .01, that exp la ined 40.1 % of the var iance. Items with posit ive loadings inc luded

attent iveness, e a s e of schedul ing, smi l ing, suggest ing schedu le options, s p e e c h

inflection, and appear ing cheerful. Items with negat ive loadings included resist ing the

confederate 's suggest ions, being d i sengaged from the task, and sel f -d isc los ing. T h e s e

nine items were used to create a s ingle index of posit ive interactive behaviour, with a

higher sco re indicating more posit ive behaviours. The a lpha reliability for this nine item

overal l behaviour index was .81.

Results

Preliminary Analyses

Part icipant Character is t ics.

T h e two groups, dep ressed and controls, were compared on a number of

demograph ic character ist ics uti l izing independent samples t-tests and ch i -square

ana lyses where appropriate. The two groups did not differ on age, level of educat ion,

employment status, or marital status (see Tab le 2 for statistical values). A s expected,

part icipants differed markedly on psychiatr ic history, medicat ion usage, and BDI score.

47

The overal l BDI sco re in the depressed group w a s indicative of a moderate level of

depress ion, whe reas the BDI score in the controls indicated no depress ion.

Confedera te C h e c k s .

In order to insure that the two confederates carr ied out the task in a simi lar

manner, a between groups (confederate identity) multivariate ana lys is of var iance

( M A N O V A ) w a s carr ied out on script deviat ions and duration of the interaction . There

were no signif icant di f ferences between the two confederates overal l , F (2, 79) = 1.69,

p_ > .10. Inspection of the mean number of script deviat ions showed that the first

confederate inc luded an average of 96 .10% of the script content, and the second

confederate inc luded 95 .89% of the content. M e a n duration for the first confederate

was 975s (sd=320s) compared to a mean of 830s (sd=340s) for the second

confederate (this di f ference approached but did not reach statistical s igni f icance, p_ >

.07). T h e s e a n a l y s e s and inspect ions of the means suggest that confederates were

largely equivalent in their presentat ion of the script to participants.

Part icipant Gender .

Corre la t ions between participant gender and all of the major dependent and

independent var iab les were calculated. Wi th one except ion, the correlat ions were non­

significant and extremely low, ranging from .13 to .02, with a mean of .06. G e n d e r did

correlate with interpersonal deference (r=.55, p_ < .001), which suggested that women

48

were rated more highly on this measure. Thus the effect of gender on this variable is

reported in the Perceptions of Participants section in the analysis.

Description of Independent Variables.

These analyses were conducted in order to examine levels of sociotropy and

autonomy in the depressed and control groups and to insure that these independent

variables were not redundant. A between groups multivariate analysis of variance

(MANOVA) was carried out with sociotropy and autonomy as the dependent variables

and depression status as the grouping variable. A main effect emerged for group, F (2,

79) = 7.44, rj < .01. Examination of the univariate values indicated a significant

difference for both sociotropy and autonomy, F (1, 80) = 7.60, p_< .01 and F (1, 80) =

12.65, p_< .01, respectively. The means for these measures in each group are

displayed in Table 3. Further descriptive statistics are available in Appendix F.

Table 3. Means and standard deviations of sociotropy and autonomy scores in

depressed and control groups.

Depressed Controls

Sociotropy 99.50 (19.36) 88.55 (16.5)

Autonomy 94.05(18.69) 81.76 (11.85)

Inspection of the means demonstrated that the depressed group scored higher on both

sociotropy and autonomy than controls. Such a result is not surprising given the

49

correlat ion between depress ive symptoms and the PSI reported in val idat ion work

(Robins, e t a l . , 1994).

In order to ensure that the personal i ty var iables were not redundant, correlat ions

were computed between sociotropy and autonomy in the overal l sample , and then in

each of the two groups. Overa l l , sociotropy and autonomy were correlated at .44, p_ <

.001. Th is correlat ion was .47, p. < .001 in the control group, and .32, p_ < .05, in the

dep ressed group. Though these personal i ty d imensions shou ld ideal ly be independent

(Robins et a l . 1994), these levels of correlat ions do not suggest a substant ial

redundancy.

Main Analyses

Overview.

For each of the major dependent var iables descr ibed below, two types of

ana lyses were used . First, correlat ions between the speci f ic dependent measure and

depress ion c lassi f icat ion (depressed vs. control), sociotropy, and autonomy were

calculated. S e c o n d , in order to s imul taneously examine the role of mood and

personal i ty and to understand the unique contribution of each of these var iables to

var iance exp la ined in the dependent measures , regression equat ions were calculated.

Order of entry for the predictors w a s b a s e d on theoretical considerat ions with the most

powerful potential predictor entered initially to control for its effect, fo l lowed by the

personal i ty var iab les of interest. Thus , in all equat ions, depress ion was entered first,

50

followed by sociotropy, autonomy, the interaction of sociotropy and depression, and the

interaction of autonomy and depression.

Desire for Future Interactions.

The correlations between the rejection measure (DFI), and the independent

variables are displayed in Table 4. There was a significant negative relationship

between DFI and depression, as well as autonomy. Thus, both autonomy and

depression were associated with rejection.

Table 4. Correlations between desire for future interactions (DFI). sociotropy.

autonomy, and depression.

Sociotropy Autonomy Depression

DFI -.13 -.35 -.51

ns p_< .01 p_< .001

The results of the regression analysis for the DFI are displayed in Table 5.

Depression status significantly, and negatively (beta = -.51) predicted desire for future

interactions in this analysis, accounting for 26% of the variance. Autonomy added

significantly (beta = -.23) to the prediction of DFI, accounting for a further 4% of the

variance. Finally the interaction of autonomy and depression significantly predicted a

further 4% of the variance in DFI scores (beta = - .24).

51

Tab le 5. Hierarchical regress ion ana lyses predicting rejection from depress ion,

sociotropy, autonomy, and interact ions.

Step, and Entered Var iab le R 2 beta df F c h ang

Desi re for Future Interactions

1. Depress ion .26 -.51 (1 ,80 ) 28.79, p_ < .001

2. Sociotropy .27 .03 (2, 79) .07

3. Autonomy .31 -.23 (3 ,78) 4.47, p_ < .05

4. Soc io , by Depress ion .31 -.02 (4, 77) .05

5. Auto, by Depress ion .35 -.24 (5, 76) 4.54, p_ < .05

The predict ive utility of the interaction between depress ion and autonomy

suggested that autonomy had different consequences for des i re for future interactions

(DFI) depending on depress ion status. Thus, in order to clarify the nature of the

interaction, correlat ions between the DFI and autonomy were ca lcu la ted in each of the

two groups (depressed and controls). There was a signif icant correlat ion between

autonomy and des i re for future interactions in the depressed group, but not in the

control group (r = - .39, p_ < .05 and r = 1 3 , ns, respectively). Thus autonomy was

signif icantly assoc ia ted with rejection, but this was true only in the dep ressed group.

Confedera te Percept ions of Part ic ipant

52

Interpersonal Deference.

A s descr ibed earl ier, there w a s a posit ive assoc ia t ion between gender and

percept ions of interpersonal deference (r = .55, p_ < .01). B e c a u s e of this, partial

correlat ions between the independent and dependent var iab les were computed,

control l ing for the effect of gender (see Tab le 6). There was a posit ive assoc ia t ion

between sociotropy and perce ived interpersonal deference in the task.

Tab le 6. Part ial correlat ions between interpersonal deference, sociotropy. autonomy,

and depress ion .

Sociotropy Autonomy Depress ion

Interpersonal .31 -.01 -.02

Deference E < .01 ns ns

The results of the regress ion equat ion for interpersonal deference are d isp layed in

Tab le 7. On ly sociotropy signif icantly predicted interpersonal deference (beta=.35) 1.

T h e addit ion of sociotropy to the equat ion resulted in a multiple R of .33, indicating that

sociotropy accounted for 11 % of the var iance in interpersonal deference. The results

of these ana l yses suggested that inc reased levels of sociotropy were assoc ia ted with

greater perce ived interpersonal deference.

53

Tab le 7. Hierarch ica l regress ion ana lyses predict ing interpersonal deference from

depress ion , sociotropy, autonomy, and interactions.

Step, and Entered Var iab le R 2 beta df F C han g

Interpersonal Deference

1. Depress ion .00 -.02 (1 ,80) .05

2. Sociot ropy .11 .35 (2 ,79) 9.78, p. < .001

3. Autonomy .13 - .15 (3, 78) 1.51

4. Soc io , by Depress ion .13 .08 (4, 77) .62

5. Auto, by Depress ion .15 .13 (5, 76) 1.02

Sel f -Or ientat ion.

T h e correlat ions between the independent var iab les and self-orientation are

reported in Tab le 8. There was a posit ive, signif icant assoc ia t ion between depress ion

and self-orientat ion. No such relat ionship w a s found for autonomy or sociotropy.

Tab le 8. Corre la t ions between self-orientat ion, sociotropy, autonomy, and depress ion .

Sociotropy Autonomy Depress ion

Self-or ientat ion -.06 .16 .41

ns ns p_< .001

In the regress ion ana lys is , only depress ion predicted self-orientation (beta=.41),

account ing for 1 7 % of the var iance (see Tab le 9). There was a trend for sociotropy to

54

predict self-or ientat ion in a negat ive direction (beta=-20, p. < .07). Overal l , the

regression and correlat ional ana lys is suggested that depress ion was assoc ia ted with a

greater sel f - focus in the task.

Tab le 9. H ierarch ica l regress ion ana lyses predicting self-orientation from depress ion,

sociotropy. autonomy, and interactions.

Step, and Entered Var iab le R 2 beta df F c h a n g

Self-orientation

1. Depress ion .17 .41 (1 ,80) 16.43, p_ <.001

2. Sociot ropy .21 -.19 (2, 79) 3.59

3. Autonomy .21 .09 (3, 78) .70

4. Soc io , by Dep ress ion .23 -.12 (4, 77) 1.54

5. Auto, by Depress ion .24 .15 (5, 76) 1.56

Participant Behav iours .

Corre la t ions between the dependent measures (the nine coded behaviours, and

the overal l index of behaviours) and independent measures (sociotropy, autonomy, and

depress ive c lassi f icat ion) are d isp layed in Tab le 10. G iven the exploratory nature of

the coding sys tem, examinat ion of the pattern of correlat ions ac ross behaviours was of

cons iderab le interest. However because the number of correlat ions to be examined

was large, only those signif icant at the .01 level or better are descr ibed here.

Depress ion w a s corre lated with being less attentive, less smi l ing, suggest ing fewer

55

Tab le 10. Corre lat ions between rated behaviours and sociotropy, autonomy, and

depress ion .

Sociot ropy Autonomy Depress ion

Attentive -.12 - .30** - .34**

Difficult to schedu le -.09 -.01 .27

Smi l ing -.07 - .30** - .53* * *

Sugges ted

schedu le opt ions

-.12 - .30** - .30**

Inflected speech -.09 -.21 - .46***

Res is ted assis tant 's

suggest ions

-.04 .03 .02

Cheerfu l - .10 - .33** - .55***

D isengaged -.00 .09 .27

Sel f -d isc los ive .07 .11 .33**

Posi t ive Behav iour

Index

-.07 - .29** - .50***

Note, **p_<.01, ***p_<.001

56

schedule options, less voice inflection, appearing less cheerful, and being more

disclosive. Autonomy was correlated with being less attentive, smiling less, suggesting

fewer schedule options, and appearing less cheerful. Sociotropy displayed no

significant correlations with any of the nine behaviours.

The regression analysis for behaviours was performed with the overall behaviour

index as the criterion. This was done for reasons of parsimony and to control for the

Type I error associated with separate regression for each of the nine behaviours. The

results of this analysis are displayed in Table 11. Depression significantly and

negatively predicted the behaviour index, with a beta weight of -.50, accounting for 34%

of the variance. The only other significant predictor was the interaction between

depression and autonomy, with a beta weight of -.28, which accounted for an additional

unique 5% of the variance in the positive behaviour index.

Table 11. Hierarchical regression analyses predicting positive behaviors from

depression, sociotropy. autonomy, and interactions.

Step, and Entered Variable R 2 beta df F c h a n g

Positive Behavior Index

1. Depression .25 -.50 (1,80) 26.82, p_ < .001

2. Sociotropy .26 .08 (2,79) .66

3. Autonomy .28 -.18 (3, 78) 2.66

4. Socio, by Depression .29 .06 (4, 77) .47

5. Auto, by Depression .34 -.28 (5, 76) 6.16, p.< .05

57

T h e signif icant interaction sugges ted that the effect of autonomy on the posi t ive

behaviour index var ied depend ing on whether or not an individual was dep ressed . In

order to clarify the nature of the interaction, correlat ions between the posit ive behav iour

index and autonomy were calculated in e a c h of the two groups (depressed and

controls). The re was a signif icant negat ive correlat ion between autonomy and the

posit ive behav ior index in the depressed group, but not in the control group (r = - .37, p_

< .05 and r = 1 6 , ns, respect ively). Thus , autonomy w a s assoc ia ted with fewer posit ive

interactive behav iors , but this assoc ia t ion was signif icant only in the dep ressed group.

Discussion

Overview

T h e s e f indings indicate that both depress ion and personal i ty impact on rejection

and interpersonal behav iours in socia l interactions. Consis tent with Coyne ' s

interpersonal theory, depress ion was assoc ia ted with rejection and fewer posit ive soc ia l

behaviours. However , autonomy modif ied these effects, inf luencing both rejection and

soc ia l behav iours in dep ressed individuals. Personal i ty and mood a lso inf luenced how

part icipants were perce ived. Sociotropy was assoc ia ted with higher levels of

interpersonal deference, whereas depress ion was assoc ia ted with higher levels of self-

orientation. Overa l l , data from both subjective (rejection, percept ions) and objective

(coded behav iours) sources suggested that an interplay of personal i ty and mood are

58

relevant to the interpersonal model of depress ion. E a c h speci f ic hypothesis will now be

d iscussed in turn.

Rejection, depression, and personality

Both depress ion and autonomy had been predicted to be assoc ia ted with

rejection, and this proved to be true. The correlat ion and regress ion ana lyses indicated

that depress ion w a s strongly assoc ia ted with rejection. In addit ion, levels of autonomy

were assoc ia ted with inc reased rejection, particularly when individuals were a lso

depressed . Thus , a combinat ion of low mood and a high level of autonomy seem

particularly l ikely to lead to this type of negative interpersonal outcome. Th is result

paral lels f indings made previously in a student sample (Biel ing & A lden , 1996), and

suggests that the c o n s e q u e n c e s of autonomy may partial ly depend on other factors in

the individual.

The robust rejection effect for depressed indiv iduals found in the present study

is highly consistent with C o y n e ' s (1976a) theoretical model . E v e n after a relatively brief

interaction (mean of approximately 14.5 minutes) there was c lear ev idence of rejection

of dep ressed individuals. T h e s e results confirm the notion that di f ferences between

depressed indiv iduals and normal controls are detectable upon an initial meeting and

that the rejection effect postulated by C o y n e does not require longer or repeated

interactions. However , this f inding is in the minority compared with the body of

empir ical f indings descr ibed earlier. Th is leads to the quest ion of why a rejection effect

was found here and not in other studies (see Marcus & Nardone, 1992, for a review).

59

Most likely, methodological factors, both in the task and participant select ion, account

for this dif ference. C o m p a r e d to the standard "getting acquainted" conversat ion, the

helping task used in the present study was perhaps more involving and personal .

Part ic ipants were required to col laborate with another person in a dynamic fashion that

required self-ref lect ion, express ion of one 's w ishes, and compromise. Moreover , unl ike

the "getting acquain ted" situation it is unl ikely that part icipants had a set "script" that ,

they cou ld fol low (Marcus & Nardone, 1992). Another strength of this study was the

use of a cl in ical sample. Al l dep ressed participants had a current d iagnos is of major

depress ion or dysthymia, suppor ted by a sufficient level of depress ive symptoms. A

number of past s tudies have used individuals who did not meet d iagnost ic criteria for a

mood disorder or undergraduate students with mild levels of dysphor ia (e.g. Borden &

Baum, 1987; McN ie l et a l . , 1987; Paddock & Nowicki , 1986). S u c h potentially

heterogeneous groups may have obscured the depress ion effect.

Depression. Personality, and Others' Perceptions

The two hypotheses regarding the percept ions of part icipants in the task were

partially supported. A s expected, sociotropy was assoc ia ted with greater perce ived

interpersonal deference. Thus , sociotropy was assoc ia ted with appear ing to be

concerned about getting a long in the interaction, and accomodat ing to the other

person 's w ishes . Th is f inding is highly congruent with descr ipt ions provided by Beck

(1983) regarding sociot ropic indiv iduals in a therapy context. However, despi te this

higher level of perce ived interpersonal deference, sociotropy did not impact level of

60

rejection or behaviours. Th is is particularly interesting in light of some theoret ical

descr ipt ions which suggest that sociotropy does, eventual ly, result in rejection (Blatt &

Zuroff, 1992). It may be that excess i ve deference and dependency c a u s e resentment

which ultimately lead to socia l rejection. However, such a p rocess may take a greater

length of time, or a greater level of intimacy to emerge. Th is may expla in why no

relat ionship w a s found here between sociotropy and rejection, and offers an interesting

avenue for research in long term interactions.

Contrary to the second percept ion hypothesis, autonomy w a s not assoc ia ted

with self-or ientat ion. However, there w a s a robust assoc ia t ion between depress ion and

self-orientat ion. Thus , dep ressed indiv iduals were seen as being less sensi t ive to

others and more se l f - focused. It is unc lear why autonomy was not corre lated with self-

orientation. It is poss ib le that autonomy is not assoc ia ted with sal ient d i f ferences in

perce ived behav iours and motivations. However, this conc lus ion is inconsistent with

both the rejection and behavioural f indings involving autonomy. Thus , the lack of

f indings regarding percept ions and autonomy are likely due to insensitivity in the

part icular measure used for this study.

Depression. Personality, and Behaviour

T h e two hypotheses regarding behaviour in the interaction were supported.

Cons is tent with predict ions, depress ion was assoc ia ted with fewer posit ive

interpersonal behaviours. Correlat ional ev idence suggested that depress ive status

w a s assoc ia ted with many of the coded behaviours including less attention to the task,

61

smil ing less , suggest ing fewer schedu le opt ions, hav ing less inflected speech ,

appear ing less cheerful , and engaging in se l f -d isc losure. A s predicted, autonomy w a s

a lso assoc ia ted with fewer posit ive behav iours including being less attentive, smi l ing

less, suggest ing fewer schedu le options, and being less cheerful . W h e n a s ingle,

composi te index of behaviour was examined, autonomy was found to interact with

depress ion , so that autonomy led to fewer posit ive behav iours in the dep ressed group,

but not in the control group. Interestingly this interaction mirrors the f indings on desi re

for future interact ions in which autonomy w a s assoc ia ted with rejection, but only in the

depressed group. T h e s e behavioural assoc ia t ions may wel l explain why autonomy was

assoc ia ted with rejection in the depressed group. The interaction may a lso expla in why

rejection was not assoc ia ted with autonomy in the control group; in non-depressed

part icipants, autonomy had no behavioural consequences . Moreover , the behavioural

correlates of autonomy may expla in previous f indings in which autonomous dysphor ics,

and self-crit ical students were rejected more than other groups (Biel ing, & A lden , 1996;

Zuroff et a l , 1983).

It is notable that the behavioural ratings were made independent of the

interaction task. T h e behavioural coding w a s carr ied out by two independent raters

whereas the des i re for future interactions measure was completed by the confederates.

Moreover, the behav iour coding was done us ing v ideo tape, whereas the rejection data

was taken from the interaction. That these two types of measures , from two distinct

modali t ies, shou ld conform to the same pattern of results is suggest ive of a rel iable and

robust effect.

62

The present f indings l inking fewer posit ive interpersonal behaviours to

depress ion are a l so consistent with at least some past s tudies of the interpersonal

model of dep ress ion (e.g. Got l ib & Rob inson , 1982; P a d d o c k & Nowicki , 1986). In

those studies, d e p r e s s e d part icipants who engaged in brief interactions with others

showed less direct support to others, less conversat ion maintenance, and engaged in

more negat ive non-verbal behav iour and more negat ive paral inguist ic behaviours.

These behav iours appear to be similar to those coded in the present study, al though

they are somewhat less speci f ic . It seems that depress ion is assoc ia ted with a variety

of negat ive interpersonal behaviours, and that these assoc ia t ions may be quite broad.

One potentially interesting avenue for further explorat ion may be to examine which of

these types of behav iours is most related to rejection.

Implications for the Interpersonal Model

T h e s e results have substant ive implications for the interpersonal model of

depress ion. First, the results from the rejection, percept ions, and behaviour var iables

offer good support for C o y n e ' s (1976) interpersonal model . Cons is tent with predict ions

made by this mode l , d e p r e s s e d individuals were more likely to be rejected, were

perce ived to be act ing in a sel f -centered manner, and engaged in fewer posit ive verbal

and non-verbal behav iours than non-depressed individuals. Interestingly, this pattern

of behaviour is highly consistent with more recent theoret ical formulat ions which

highlight def ic ienc ies in respons iveness and pol i teness that may occur in depressed

individuals (Segr in & Ab ramson , 1994).

63

T h e present f indings for the rejection and behavioural var iab les offer some

degree of support for the inclusion of personal i ty factors in the interpersonal model of

depress ion . Autonomy, in particular, is likely to be a va luable addit ion. T o the extent

that a dep ressed individual is autonomous, they will engage in fewer posit ive

interpersonal behaviours and exper ience a greater level of rejection. Th i s inc reased

ability to predict rejection within dep ressed individuals represents a signif icant step

toward greater specif icity in the model . However, this new specif ic i ty c a n best be

regarded as an addit ion to the model rather than a revision.

T h e s e results a lso highlight the dynamic, cycl ical nature of interpersonal

behaviour. S o m e writers have emphas i zed the role of early learning exper iences and

relat ionships with attachment f igures as foundat ions for excess i ve interpersonal

re la tedness and self-definit ion (Blatt, 1974; Blatt & Zuroff, 1992). Thus , over the

course of the individual 's development, the impact of these personal i ty styles on

depress ion "may be partially attributable to the interpersonal envi ronments they seek,

establ ish, and maintain" (Blatt & Zuroff, 1992, p. 538). In other words, ear ly learning

and subsequent express ion of these personal i ty patterns can result in a develop ing,

evolv ing interaction cyc le in which the individual 's fears and concerns are reinforced by

their own behav iours and result ing negat ive exper iences. The results for sociotropy,

autonomy, and depress ion in this interpersonal situation support the notion that the

individual p lays a role in the ongo ing formation of their soc ia l context. Au tonomous and

dep ressed individuals are likely to create soc ia l situations in wh ich their behaviour

leads to rejection, result ing in further soc ia l isolation and mistrust. Sociotropy, and the

interpersonal deference assoc ia ted with it, may result in an individual not hav ing their

own needs met, or p lacing an excess i ve burden on others.

T o il lustrate the potential impact of these personal i ty styles in ongo ing soc ia l

relat ionships, comments made to the exper imenter during the cl inical interview and de ­

briefing may offer some degree of insight. The first set of remarks were made by a 54 -

year old woman with an autonomy score of 115 (depressed mean 94) and a BDI of 23.

During the S C I D she was asked a s tandard quest ion which was intended to sc reen for

Gene ra l i zed Anxiety Disorder:

I: In the last six months, wou ld you say you've been particularly nervous

or anx ious?

P: I'd say that I've been eas i ly irritable, not so much anxious but.. .people

don't understand me. I don't suffer fools, and its like if somebody a s k s me

someth ing I'm quite wil l ing to do it. But they start sort of, you know, what

to me is a bunch of nonsense , and I..lose my pat ience. Th is has been a

retreat from people; there are only certain people I want to be around. M y

to lerance level is not very high. I'd just as soon be alone.. . . S o it has had

an impact on my fr iendships, and I guess that if people are comfortable

enough to want to do stuff aga in , they can contact me. I mean I have

ca l led people.. . there's no response .

Th is person 's responses indicate cons iderab le confl icts about relat ionships and limited

motivation to be with others. The re are concerns that others may take advantage, and

lack of to lerance for others "nonsense. " A s a result, there is a protective withdrawal

65

and lack of contact. O n c e the person feels more motivated to be soc ia l , there is a lack

of response from others. Th is rejection makes it difficult for this woman to have posit ive

socia l contacts, conf i rming the belief that people "don't understand." A second

illustration is drawn from a de-brief ing with a 43-year old woman who scored 99 on

autonomy and had a BDI of 27. In the informal d i scuss ion fol lowing the de-brief ing, the

interviewer and participant are d iscuss ing the c a u s e s of depress ion :

I: S o what do you think caused your depress ion .

P: W e l l , I know what the cause is. I feel l ike a fai lure because I'm not

ach iev ing things. I get into great despai r over that. A n d I am angry and

frustrated with society.

I: W o u l d you say that you feel lonely?

P: Somet imes I feel lonely but a lot of the time I prefer to be by myself.

I'm furiously... I mean I wont even let my father come and visit me, or wont

let anybody stay with me because I'm afraid they will take advantage of

me. I just can't cope with that basical ly..I 've not iced considerably,

espec ia l l y in the last six, seven years, that I don't want to do anything for

anybody. I feel very protective about my energy and my space . S o I'd

say that I'm probably into extreme protect ionism. W h i c h has its dire

consequences . . .o f lonel iness and . . .a more shut down life.

Aga in , there is s t rong ev idence of negative feel ings about relat ionships, and a des i re to

protect l imited energy for the self. However, there appear to be feel ings of lonel iness,

66

as wel l . T h e s e excerpts are brief, and non-systematic; however they offer some degree

of il lustration of these p rocesses and their impact on the person 's exper ience.

Caveats

Although the present study represents an advance over previous research, there

are severa l methodologica l i ssues that may limit the study's general izabi l i ty. The first

concerns the measurement of sociotropy and autonomy. A s descr ibed in the

introduction, at least three other measure of these constructs exist ( D E Q , S A S , and the

D A S ) . Though the measure used in the present study is thought to be psychometr ical ly

superior to these other measures , the majority of research completed in this a rea to

date has used the D E Q . Whe the r or not f indings for sociotropy and autonomy would

paral lel f indings for dependency and self-crit icism remains an empir ical quest ion.

Inclusion of the D E Q would have provided such an answer and thus increase the

general izabi l i ty of these f indings. A second concern for the present study is the

measurement of percept ions and behaviours. A l though these measures were

empir ical ly der ived and pi loted, other commonly used measu res of interpersonal

percept ion (e.g. the IMI) would have resulted in addit ional information and

general izabi l i ty. T h e final i ssue concerns gender effects. A l though general ly no

significant gender effects were found, there was one except ion which was control led for

in the subsequent ana lyses . However, because of the smal l percentage of men in the

study, any between-groups compar isons would have been difficult to make. Future

studies may wish to examine equal numbers of men and women.

67

Future Directions

T h e present results suggest a number of future directions for researchers

examining the interpersonal model of depression. First, the present results suggest

that the type of interaction in which participants are engaged may strongly influence the

strength of findings on behaviour and rejection. Determining which aspects of an

interaction (e.g. collaboration, personalized content) are germane to the rejection

phenomenon may tell us more about the various everyday social contexts in which this

effect is likely to arise. Overall, the results of this study serve to encourage greater

creativity in designing interpersonal situations for laboratory study.

Outside of the interpersonal model, these findings may be of interest to

clinicians and researchers examining process and outcome in psychotherapy. S o m e

research examining sociotropy and autonomy and outcome/has suggested that

autonomy is associated with response to antidepressant medication (Peselow et al.,

1992). Beck (1983) also suggested that different forms of intervention might be

required in sociotropic and autonomous depressives, and that the therapist's stance

should be varied to suit the individual's personality style. T h e present findings

reinforce these suggestions. In the structured helping task used in this study,

sociotropy was associated with a greater perceived interpersonal deference. This

might increase the level of rapport or therapeutic alliance, and ultimately impact the

outcome of psychotherapy. Similarly autonomy, especially when combined with

depression, may be associated with less positive interpersonal behaviours,

68

undermining the formation of a strong therapeut ic a l l iance, and perhaps lessen ing the

posit ive impact of therapy. Ultimately, psychotherapy with autonomous depress i ves

may n e e d to take into account the interpersonal style of the patient, and its consequent

impact on the therapist. Thus , examining these personal i ty var iables in the context of

p rocess or outcome studies is of cons iderab le interest and importance.

Overall Summary

T a k e n together, the results from this study offer c lear support for the

interpersonal mode l of depress ion . However, this study a lso supports the inc lus ion of

sociotropy a n d autonomy as a step toward bui lding a more comprehens ive explanatory

model of interpersonal behaviour in depress ion . T h e s e personal i ty factors are

part icularly attractive to c l in ic ians and researchers al ike because they go beyond

symptomatology to focus on the motivational aspec ts of depress ive interpersonal

behaviour. T h e personal i ty var iable of sociotropy, an investment in interpersonal

re lat ionships, is detectable by others in a theoretical ly consistent manner. Autonomy,

investment in self-str iving and achievement, is not only detectable by others, but leads

to rejection and fewer posit ive soc ia l behaviours. Ultimately, these f indings enhance

our unders tand ing of these personal i ty constructs and their consequences .

6 9

Footnotes

1 Because of the correlation between interpersonal deference and gender the

regression analysis was repeated with participant's gender entered as the first

predictor. All other predictor variables were entered in the same order as in the

main analysis. For the regression involving sociotropy, gender was a predictor of

interpersonal deference (beta = .51, F(1, 80) = 27.81, rj < .001) and accounted

for 26% of the variance. However sociotropy continued to predict interpersonal

deference (beta = .30, F(1, 78) = 9.31, p_ < 01), and accounted for a further 8% of

the variance. Thus, women were rated as higher in interpersonal deference, but

this did not negate or modify the effects of sociotropy on this variable.

References

Abramsbn, L. Y. , Seligman, M . E . P., & Teasdale , j . D. (1978). Learned

helplessness in humans: Critique and reformulation. Journal of Abnormal Psychology.

87, 49-74.

Alden, L. E . , & Bieling, P. J . (1996). Interpersonal convergence of personality

constructs in dynamic and cognitive models of depression. Journal of Research in

Personality. 30,60-75.

Alden, L. E . , Wiggins, J . S., & Pincus, A . (1990). Construction of circumplex scales

for the Inventory of Interpersonal Problems. Journal of Personality Assessment . 55,

521-536.

Allen, N. B., de L. H o m e , D. J. , &Trinder , J . (1996). Sociotropy, autonomy, and

dysphoric emotional responses to specific classes of stress: A psychophysiological

evaluation. Journal of Abnormal Psychology, 105.25-33.

Arieti, S., & Bemporad, J . R. (1980). T h e psychological organization of depression.

American Journal of Psychiatry. 137. 1360-1365.

Beck, A . T . (1983). Cognitive therapy of depression: New perspectives. In P. J .

C l a y t o n & 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. P., & Emery, G . (1983). Development of the

Sociotropy-Autonomy Scale: A measure of personality factors in psychopathology.

Unpublished Manuscript. University of Pennsylvania, Philadelphia.

•71

Beck, A . T . , Rush, A . J. , Shaw, B. F. & Emery. G . (1979V Cognitive Therapy of

Depression. New York: Guildford Press.

Beck, A . T , Steer, R. A . , & Garbin, M . (1988). Psychometric properties of the Beck

Depression Inventory: Twenty-five years of evaluation. Clinical Psychology Review, 8,

77-100.

Beck, A . T , Ward , C . H . , Mendelson, M . , Mock, J. , & Erbaugh, J . (1961). A n

inventory for measuring depression. Archives of General Psychiatry, 4, 561-571.

Bieling, P. J , , & Alden, L. E . (In press). Interpersonal patterns in dysphoria: T h e role

of sociotropy and autonomy. Cognitive Therapy and Research.

Billings, A . , & Moos , R. (1982). Psychosocial theory and research on depression: A n

integrative framework and review. Clinical Psychology Review. 2. 213-237.

Blaney, P ; H . , & Kutcher, G . S; (1991). Measures of depressive dimensions: Are they

interchangeable? Journal of Personality Assessment . 56, 502-512.

Blatt, S. J . (1974). Levels of object representation in anaclitic and introjective

depression. Psychoanalytic Study of the Child. 29 , 107-157.

Blatt, S. J . (1995). T h e destructiveness of perfectionism: Implications for the

treatment of depression. American Psychologist, 50, 1003-1020.

Blatt, S. J „ D'Affliti, J . P., & Quinlan, D. M (1976). Experiences of depression in

young adults. Journal of Abnormal Psychology, 85, 383-389

Blatt, S. J. , Quinlan, D. M . , Chevron, E . S., McDonald , C . , & Zuroff, D. (1982).

Dependency and self-criticism: Psychological dimensions of depression. Journal of

Consulting and Clinical Psychology. . 50(1). 113-124.

Blatt, S. J . , W e i n , S. J . , Chevron, E . , & Quinlan, D. M . (1979). Parental

representations and depression in normal young adults. Journal of Abnormal

Psychology. 88, 388- 397.

Blatt, S. J . , Schaffer, C . E . , Bers, S. A . , & Quinlan, D. M . (1992). Psychometric

properties of the Depressive Experiences Questionnaire for adolescents. Journal of

Personality Assessment . 59, 82-98.

Blatt, S. J. , & Zuroff, D. C . (1992). Interpersonal relatedness and self-definition: Two

prototypes for depression. Clinical Psychology Review 12, 527-562.

Blatt, S. J . , Zuroff, D. C , Quinlan, D. M . , & Pilkonis, P. A . (1996). Interpersonal

factors in brief treatment of depression: Further analyses of the National Institute of

Mental Health Treatment of Depression Collaborative Treatment Program. Journal of

Consulting and Clinical Psychology., 64, 162-171.

Borden, J . W . & Baum, C . G . (1987) Investigation of a social-interactional model of

depression with mildly depressed males and females. Sex Roles, 17, 449-465.

Bowlby, J . (1977). T h e making and breaking of affectional bonds: I. Aetiology and

psychopathology in light of attachment theory. British Journal of Psychiatry. 130, 201 -

210.

73

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.

Chevron, E . S., Quinlan, D. M . , & Blatt, S. J . (1978). S e x roles and gender

differences in the experience of depression. Journal of Abnormal Psychology, 88, 680-

683.

Cappeliez, P. (1993). T h e relationship between Beck's concepts of sociotropy and

autonomy and the NEO-Personali ty Inventory. British Journal of Clinical Psychology.

32,78-80.

Costa, P. T , & M c C r a e , R. R. (1985). T h e NEO-Personali ty Inventory manual.

O d e s s a : F L : Psychological Assessment Resources.

Coyne, J . C . (1976a). Toward an interactional description of depression. Psychiatry,

39, 28- 40.

Coyne, J . C . (1976b). Depression and the response of others. Journal of Abnormal

Psychology, 85, 186-193.

Coyne, J . C . (1985). Studying depressed persons' interactions with strangers and

spouses. Journal of Abnormal Psychology, 94, 231-232.

Coyne, J . C . (1994). Self-reported distress: Analogue or ersatz depression?

Psychological Bulletin, 116. 29-45.

74

Dobson, K. S. (1989). Real and perceived interpersonal response to subcl inical^

anxious and depressed targets. Cognitive Therapy and Research. 13(1), 37-47.

Doerfler, L. A . , & Chaplin, W . F. (1985). T y p e III error in research on interpersonal

models of depression. Journal of Abnormal Psychology, 94. 227-230.

Gotlib, I. H . , & Robinson, L. A . (1982). Responses to depressed individuals:

Discrepancies between self-report and observer rated behaviour. Journal of Abnormal

Psychology. 9_1, 231-240.

Goldberg, J . 0. , Segal , Z . V . , Vella , D. D. , & Shaw, B. F. (1989). Depressive

personality: Millon Clinical Multiaxial Inventory profiles of sociotropic and autonomous

subtypes. Journal of Personality Disorders, 3, 193-198.

Hammen, C , Ellicott, A . , & Gitlin, M . (1989). Vulnerability to specific life events and

prediction of course of disorder in unipolar depressed patients. Canadian Journal of

Behavioural Science. 21_, 377-388.

Hammen, C , Ellicott, A . , & Gitlin, M . (1992). Stressors and sociotropy/autonomy: A

longitudinal study of their relationship to the course of bipolar disorder. Cognitive

Therapy and Research. 16. 409-418.

Hammen, C , Ellicott, A . , Gitlin ; M . , & Jamison, K. R. (1989). Sociotropy/autonomy

and vulnerability to specific life events in patients with unipolar and bipolar disorders.

Journal of Abnormal Psychology, 98, 154-160.

75

Hammen, C , Marks, T . , Mayol , A . , & deMayo, R. (1985). Depressive self-schemas,

life stress and vulnerability to depression. Journal of Abnormal Psychology, 94, 308-

319.

Horowitz, L. M . , Rosenberg, S. E . , Baer, B. A . , Ureno, G . , & Villasenor, V . S., (1988).

Inventory of Interpersonal Problems: Psychometric properties and clinical applications.

Journal of Consulting and Clinical Psychology, 56, 885-892.

Joiner, T . E . , Alfano, M . S., & Metalsky, G . I. (1992). W h e n depression breeds

contempt: Reassurance seeking, self-esteem, and rejection of depressed college

students by their roommates. Journal of Abnormal Psychology. 101, 165-173.

Joiner, T . E . , & Metalsky, G . I. (1995). A prospective study of an integrative

interpersonal theory of depression: A naturalistic study of college roommates. Journal

of Personality and Social Psychology, 69, 778-788.

Keller, M . B., Lavori, P . W . , Rice, J . , Coyell , W. , & Hirschfeld, R. M . A . (1986). T h e

persistent risk of chronicity in recurrent episodes of non-bipolar major depressive

disorder: A prospective follow-up. American Journal of Psychiatry, 143, 24-28.

Kiesler, D. J . , Anchin, J . C , Perkins, M . J. , Chirico, B. M . , Kyle, E . M . , & Federman,

E . J . (1976). T h e Impact M e s s a g e Inventory. Richmond: Virginia Commonwealth

University.

King, D. A . , & Heller, K. (1984). Depression and the response of others: A re-

evaluation. J p j j m a l p f j A b ^ ^

7 6

Libet, J. , & Lewinsohn (1973). T h e concept of social skill with special reference to

the behaviour of depressed persons. Journal of Consulting and Clinical Psychology,

40 ,304-312 .

MacPhillamy, D. J. , & Lewinsohn, P. M . (1971). Pleasant Events Schedule .

Unpublished manuscript, University of Oregon.

Marcus, D. K. & Nardone, M . E . (1992). Depression and interpersonal rejection.

Clinical Psychology Review. 12, 433- 449.

McCranie , E . C , & Bass, J . D. (1984). Childhood family antecedents of dependency

and self-criticism: Implications for depression. Journal of Abnormal Psychology, 93, 3-

8.

McNiel , D. E . , Arkowitz, H . S., & Pritchard, B. E . (1987). T h e response of others to

face- to-face interaction with depressed patients. Journal of Abnormal Psychology, 96,

341-344. j •

Millon, T . (1981). Millon Clinical Multiaxial Inventory Manual (3rd ed.). Minneapolis,

M N : National Computer Systems.

Mongrain, M . (1993). Dependency and self-criticism located within the five factor

model of personality. Personality and Individual Differences, 15; 455-462.

Mongrain, M . , & Zuroff, D. C . (1989). Cognitive vulnerability to depressed affect in

dependent and self-critical college women. Journal of Personality Disorders, 3, 240-

251.

77

Nietzel, M . T . , & Harris, M . J . (1990). Relationship of dependency and

achievement/autonomy to depression. Clinical Psychology Review, 10, 279-297.

Ouimette, P. C , Klein, D. N. , Anderson, R., Riso, L. P., & Lizardi, H . (1994).

Relationship of sociotropy/autonomy and dependency/self-criticism to DSM-III-R

personality disorders. Journal of Abnormal Psychology, 103, 743-749.

Paddock, J . R., & Nowicki, S. (1986). Paralanguage and the interpersonal impact of

dysphoria: It's not what you say but how you say it. Social Behavior and Personality,

14, 29-44.

Peselow, E . D. , Robins, C . J . , Sanfilipo, M . P., Block, P., & Fieve, R. R. (1992).

Sociotropy and autonomy: Relationship to antidepressant drug treatment response to

and endogenous- nonendogenous dichotomy. Journal of Abnormal Psychology, 101,

479-486.

Pilkonis, P. (1988). Personality prototypes among depressives: T h e m e s of

dependency and autonomy. Journal of Personality Disorders, 2, 144-152.

Riley, W . T . , & McCranie , E . W . (1990). T h e Depressive Experiences Questionnaire:

Validity and psychological correlates in a clinical sample. Journal of Personality

Assessment, 54, 523-533.

Robins, C . J . (1990). Congruence of personality and life events in depression.

Journal of Abnormal Psychology, 99, 393-397.

' ' 7 8'

Robins, C J. , & Block, P. (1988). Personal vulnerability, life events, and depressive

symptoms: A test of the specific interactional model. Journal of Personality and Social

Psychology. 54, 847-852.

Robins, C . J . , & Block, P. (1988). Cognitive theories of depression viewed from a

diathesis- stress perspective: Evaluations of the models of Beck and of Abramson,

Seligman, and Teasdale . Cognitive Therapy and Research, 13, 297-313.

Robins, C . J . , Block, P., & Peselow, E . D. (1989). Relations of sociotropic and

autonomous personality characteristics to specific symptoms in depressed patients.

Journal of Abnormal Psychology, 98, 86-88.

Robins, C . J . , Ladd, J . , Welkowitz, J . , Blaney, P. H . , Diaz, R , & Kutcher, G . (1994).

T h e Personal Style Inventory: Preliminary validation studies of new measures of

sociotropy and autonomy. Journal of Psychopathology and Behavioural Assessment ,

16, 277-300.

Robins, C . J . , & Luten, A . G . (1991). Sociotropy and autonomy: Differential patterns

of clinical presentation in unipolar depression. Journal of Abnormal Psychology, 100,

74- 77.

Rosenblatt, A . , & Greenberg, J . (1991). Examining the world of the depressed: Do

depressed people prefer others who are depressed. Journal of Personality and Social

Psychology, 60, 620-629.

Sadeh, A . , Rubin, S. S., & Berman, E . (1993). Parental relationship representations

and experiences of depression in college students. Journal of Personality Assessment,

60, 192-204.

Segal , Z . V , Shaw, B. F., & Veila , D. D. (1989). Life stress and depression: A test of

the congruency hypothesis of life event content and depressive subtype. Canadian

Journal of Behavioural Science, 21, 389-400.

Segrin, C . (1993). Interpersonal reactions to dysphoria: T h e role of relationship with

partner and perceptions of rejection. Journal of Social and Personal Relationships, 10,

83-97.

Segrin, C , & Abramson, L. Y. (1994): Negative reactions to depressive behaviors; A

communications theories analysis. Journal of Abnormal Psychology, 103, 655-668.

Segrin, C , & Dillard, J . D. (1992). T h e interactional theory of depression: A meta­

analysis of the research literature. Journal of Social and Clinical Psychology, 11_, 43-70,

Spitzer, R. L , Williams, J . B. W . , Gibbon, M . & First, M . B. (1990). User's Guide for

the Structured Clinical Interview for DSM-III-R. Washington, D C : American Psychiatric

Press. '

. Steer, R. A . , Beck, A . T . , & Garrison, B. (1986). Applications of the Beck Depression

Inventory. In N. Sartorius & T . A . Ban eds.), Assessment of Depression. Heidelberg:

Springer-Verlag.

. . . . 8 0

Strack, S. & Coyne , J . C . (1983). Social confirmation of dysphoria: Shared and

private reactions to depression. Journal of Personality and Social Psychology, 44, 798-

806.

Viglione, D. J . , Clemmey, P. A . , & Camenzuli , L. (1990). T h e Depressive

Experiences Questionnaire: A critical review/Journal of Personality Assessment , 55,

52-64.

Vredenburg, K., Flett G . L , & Krames, L. (1993). Analogue versus clinical

depression: A clinical reappraisal. Psychological Bulletin, 113, 327-344.

Welkowitz, J . , Lish, J . D. , & Bond, R N. (1985). T h e Depressive Experiences

Questionnaire: Revision and validation. Journal.of Personality Assessment , 49, 89-94.

Whiffen, V . E . , & Sasseville, T . M . (1991). Dependency, self-criticism, and

recollections of parenting: Sex differences and the role of depressive affect. Journal of

Social and Clinical Psychology, 10. 121-133.

Zuroff, D. C . (1994). Depressive personality styles and the five factor model of

personality. Journal of Personality Assessment , 63, 453-472.

Zuroff, D. C , & deLorimier, S. (1989). Ideal and actual romantic partners of women

varying in dependency and self-criticism. Journal of Personality, 57, 825-846.

Zuroff, D. C . & Fitzpatrick, D. K. (1995)! Depressive personality styles: Implications

for adult attachment. Personality and Individual Differences, 18,253-365.

81

A p p e n d i c e s

Append ix A

P L E A S A N T E V E N T S T A S K - S C R I P T

S o the idea here is for us to try to come up with some p leasant things that you could

add in to your dai ly or week ly routine. T h e s e fun activit ies shou ld be things that you

think you cou ld fit in your schedu le without increasing your s t ress level, and you

shouldn't feel that these are things you have to do. Y o u shou ld do these things just for

fun and enjoyment, O K ?

[go get schedu les ]

Somet imes it's real ly hard for people to come up with new ideas for interesting kinds of

activit ies that they cou ld do. S o , I have a list of p leasant activit ies that we can use to

help us a long. A n y ques t ions?

I think that the best way to approach this task is to do it in a coup le of steps. I guess

the first thing we n e e d to know is how your dai ly schedu le looks, and what kinds of

things you are a l ready doing, including things like work or schoo l , or any other activit ies

that take up time in your day. O n c e we s e e what's a l ready there, we can try to add

some things that you might like to do. Does that s e e m o k ?

Al l right, I guess first of all I'll g ive you this schedule , the t imes are a long here and the

days a long here. M a y b e you can just take a minute or two to c ross off those t imes

when you're a l ready busy throughout the week.

82

[subject completes schedule ]

Ok. S o that he lps us to get a feel for what t imes you have avai lab le . Now, the next

thing I'd like for us to do is to go over this list of activit ies. T h e list is made up of a

whole bunch of things that s o m e people f ind to be enjoyable activi t ies or past imes. Not

all of the things on here will appea l to you, but there are probably some things which

you might f ind p leasant or enjoyable. I'd like to go over the list, I'll just give you each

item, and maybe you can tell me whether that's something you might like to do or not.

O n c e we've gone over the list we' l l try to fit those things into your schedu le .

How about [first through last item]?

[Check off items which are agreed to.]

[If they say " yes " to ten items stop.]

[If they say " yes " to less than f ive items on first list, go to s e c o n d list and cont inue until

you get five.]

S o let's start with [first item]. Do you think that's something you could do or are there

things that might interfere with [item one]?

W h e n do you think would be a good time to do that?

Wha t about [second item]. Do you think there's anything that might interfere with that

or keep you from doing [second item]?

W h e r e might that one fit on the schedu le?

83

[Cont inue cyc le through to last item, e a c h time ask ing "What about...what cou ld

interfere...and when?" ]

Ok, so that's all the things you p icked out. Do you think that this is a reasonab le plan,

or are there some activit ies that you might not like to do or others that you would like to

a d d ?

Ok, great. S o , now that we're done with that I'll go and get [experimenter]. It w a s n ice

meeting you.

Append ix B

1. G o i n g to a movie

2. L is tening to mus ic

3. Go ing to a bookstore

4. S leep ing late

5. P lann ing a trip

6. Buy ing someth ing for myself

7. Be ing at the b e a c h

8. Redecora t ing my living s p a c e

9. Do ing art work

10. G o i n g to a sports event

11. R e a d i n g a novel or story

12. G o i n g to a bar

13. Boat ing

14. C a m p i n g

15. P lay ing ca rds

16. Do ing puzz les , c rosswords

17. P lay ing tennis

18. Woodwork ing or carpentry

19. Wri t ing stor ies, poems

20. Hik ing or explor ing

21 . Learn ing a foreign language

22. P lay ing a musical instrument

23 . Sk i ing

24. Tak ing a bath

25. P lay ing bi l l iards or pool

26. P lay ing c h e s s or checkers

27. Bowl ing

28. G o i n g to a park

29. Buy ing something for my family

30. Photography

31 . Be ing in the mountains

32. See ing beautiful scenery

33. Eat ing good meals

34. G o i n g to a museum or exhibit

35. Hav ing peace and quiet

36. Vis i t ing fr iends

37. Meditat ing or doing yoga

38. Runn ing , jogging, or aerob ics

39. Hav ing coffee or tea with f r iends

40. Doing volunteer work

Append ix C

85

On this questionnaire are groups of statements. Please read each group of statements carefully. Then pick out the one statement in each group which best describes the way you have been feeling the PAST WEEK, INCLUDING TODAY! Circle the number beside the statement you picked. If several statements in the group seem to apply equally well, circle each one. Be sure to read all the statements In each group before making your choice.

1. 0 I do not feel sad. 12. 0 I have not lost interest in other people. 1 1 feel sad. 1 1 am less interested in other people than 1 used to be. 2 1 am sad all the lime and 1 can't snap out of it. 2 1 have lost most Of my interest in other people. 3 1 am so sad or unhappy that 1 can't stand it. 3 1 have lost all of my interest in other people.

2. 0 I am not particularly discouraged about the future. 13. 0 1 make decisions about as well as 1 ever could. 1 I feel discouraged about the future. 1 1 put off making decisions more than 1 used to. 2 I feel I have nothing to look forward to. 2 1 have greater difficulty in making decisions than before. 3 I feel that the future is hopeless and that things cannot 3 1 can't make decisions at all any more.

improve. 14. 14. 0 1 don't feel 1 look any worse than 1 used to.

3. 0 I do not feel like a failure. 1 1 am worried that 1 am looking old or unattractive. 1 I feel I have failed more than the average person. 2 1 feel that there are permanent changes in my appearance 2 As I look back on my life, all I can see is a lot of failures. that make me look unattractive. 3 I feel I am a complete failure as a person. 3 1 believe that I look ugly.

4. 0 I get as much satisfaction out of things as I used to. 15. 0 1 can work about as well as before. 1 I don't enjoy things the way I used to. 1 It takes an extra effort to get started at doing something. 2 I don't get real satisfaction out of anything anymore. 2 1 have to push myself very hard to do anything. 3 I am dissatisfied or bored with everything. 3 1 can't do any work at all.

5. 0 I don't feel particularly guilty. 16. 0 1 can sleep as well as usual. 1 1 feel guilty a good part of the time. 1 1 don't sleep as well as 1 used to. 2 1 feel quite guilty most of the time. 2 1 wake up 1 -2 hours earlier than usual and find it hard to get 3 1 feel guilty all of the time. back to sleep. 3 1 feel guilty all of the time.

3 1 wake up several hours earlier than 1 used to and cannot get 6. 0 1 don't feel I am being punished. back to sleep.

1 1 feel 1 may be punished. 17. 2 1 expect to be punished. 17. 0 1 don't get more tired than usual.'

3 1 feel 1 am being punished. 1 1 get tired more easily than I used to. 3 1 feel 1 am being punished. 2 1 get tired from doing almost anything.

7. 0 1 don't feel disappointed in myself. 3 1 am too tired to do anything. 1 1 am disappointed in myself.

18-2 1 am disgusted with myself. 18- 0 My appetite is no worse than usual. 3 1 hate myself. 1 My appetite is not as good as it used to be. 3 1 hate myself.

2 My appetite is much worse now. 8. 0 1 don't feel 1 am any worse than anybody else. 3 1 have no appetite at all any more.

1 1 am critical of myself for my weaknesses or mistakes. 19. 2 1 blame myself all the time for my faults. 19. 0 1 haven't lost much weight, if any lately.

3 1 blame myself for everything bad that happens. 1 1 have lost more than S pounds. 3 1 blame myself for everything bad that happens. 2 1 have lost more than 10 pounds.

9. 0 1 don't have any thoughts of killing myself. 3 1 have lost more than 15 pounds. 1 1 have thoughts of killing myself, but 1 would not carry them

out. 1 am purposely trying to lose weight by eating less. Yes • No L.

2 1 would like to kill myself. 20. 3 1 would kin myself if 1 had the chance. 20. 0 I am no more worried about my health than usual. 3 1 would kin myself if 1 had the chance.

1 I am worried about physical problems such as aches and 10. 0 I don't cry any more than usual. pains, or upset stomach, or constipation.

1 I cry more now than 1 used to. 2 I am very worried about physical problems and it's hard.to 2 1 cry all the time now. think of much else. 3 1 used to be able to cry, but now 1 can't cry even though 1 3 I am so worried about my physical problems, that 1 cannot

want to. think about anything else.

11. 0 I am no more irritated now than 1 ever am. 21. 0 1 have not noticed any recent change in my interest in sex. 1 1 get annoyed or irritated more easily than 1 used to. 1 1 am less interested in sex than 1 used to be. 2 1 feel irritated all the time now. 2 1 am much less interested in sex now. 3 1 don't get irritated at all by the things that used to irritate

me. 3 1 have lost interest in sex completely.

(BDI-52E)

Appendix D

86

Here a re a number o f s ta tements about p e r s o n a l c h a r a c t e r i s t i c s . P l e a s e read each one c a r e f u l l y , and I n d i c a t e whether you agree a number.

or d i s a g r e e , and to v h a t e x t e n t , by c i r c l i n g

S t r o n g l y Di sagree D i s a g r e e

S l i g h t l y D i s a g r e e

S l i g h t l y Agree Agree

S t r o n g l Agree

1. I o f t e n put o t h e r p e o p l e ' s 1 needs b e f o r e my own.

2 3 4 5 6

2 . I t end to keep o t h e r 1 p e o p l e a t a d i s t a n c e .

2 3 4 S 6

3. I f i n d I t d i f f i c u l t to be 1 s e p a r a t e d from people I l o v e .

2 3 4 5 6

4 . I am e a s i l y b o t h e r e d by o t h e r 1 p e o p l e making demands o f me.

2 3 4 5 6

5 . I am v e r y s e n s i t i v e to the 1 e f f e c t s I have on the f e e l i n g s o f o t h e r p e o p l e .

2 3 4 5 6

6 . I d o n ' t l i k e r e l y i n g on 1 o t h e r s f o r h e l p .

2 3 4 5 6

7 . I an v e r y s e n s i t i v e to 1 c r i t i c i s m by o t h e r s .

2 3 4 5 6

8. I t b o t h e r s me when I f e e l 1 t h a t I am o n l y average and o r d i n a r y .

2 3 4 S 6

9 . I worry a l o t about h u r t i n g 1 o r o f f e n d i n g o t h e r p e o p l e .

2 3 4 5 6

10. When I ' m f e e l i n g b l u e , I d o n ' t 1 l i k e to be o f f e r e d sympathy .

2 3 4 5 6

11 . I t i s h a r d f o r me to b r e a k 1 o f f a r e l a t i o n s h i p even i f i t I s making me unhappy.

2 3 4 5 6

12. I n r e l a t i o n s h i p s , p e o p l e 1 a r e o f t e n too demanding o f one a n o t h e r .

2 3 4 5 6

13 . I am e a s i l y persuaded by 1 o t h e r s .

2 3 4 5 6

14. I u s u a l l y v iew my per fo rmance 1 2 3 4 5 6 as e i t h e r a complete s u c c e s s o r a complete f a i l u r e .

87

S t r o n g l y S l i g h t l y S l i g h t l y Strong Di sagree D i s a g r e e Di sagree Agree A g r e e Agree

15. I t r y to p l e a s e o ther people too much.

16. I d o n ' t l i k e people to invade my p r i v a c y .

17. I f i n d i t d i f f i c u l t i f I have to be a l o n e a l l day .

18. I t i s h a r d f o r me to take I n s t r u c t i o n s from people vho have a u t h o r i t y over me.

19. I o f t e n f e e l r e s p o n s i b l e f o r s o l v i n g o t h e r p e o p l e ' s p r o b l e m s .

20 . I o f t e n h a n d l e b i g d e c i s i o n s w i t h o u t t e l l i n g anyone e l s e about them.

2 1 . I t i s v e r y h a r d f o r me to get o v e r t h e f e e l i n g o f l o s s when a r e l a t i o n s h i p has ended.

22 . I t i s h a r d f o r me to have someone dependent on me.

2 3 . I t i s v e r y Important t o me t o be l i k e d o r admired by o t h e r s .

24 . I f e e l b a d l y about m y s e l f when I am n o t a c t i v e l y a c c o m p l i s h i n g t h i n g s .

25 . I f e e l I h a v e to be n i c e t o o t h e r p e o p l e .

26 . I t i s h a r d f o r me to expres s a d m i r a t i o n o r a f f e c t i o n .

2 7 . I l i k e to b e c e r t a i n t h a t t h e r e i s somebody c l o s e I can c o n t a c t i n case something u n p l e a s a n t happens to me.

28 . I t i s d i f f i c u l t f o r me to make a l o n g - t e r m committment to a r e l a t i o n s h i p .

88

- 3 -

S t r o n g l y S l i g h t l y S l i g h t l y S t r o n g ! Di sagree D i s a g r e e D i s a g r e e Agree Agree Agree

29 . I am too a p o l o g e t i c t o o t h e r p e o p l e .

30 . I t i s h a r d f o r me t o open up and t a l k about my f e e l i n g s and o t h e r p e r s o n a l t h i n g s .

31 . I am v e r y c o n c e r n e d w i t h how peop le r e a c t t o me.

32. I have a h a r d t ime f o r g i v i n g m y s e l f when I f e e l I h a v e n ' t worked up to .my p o t e n t i a l .

33. I get v e r y u n c o m f o r t a b l e when I 'm n o t s u r e w h e t h e r . o r not someone l i k e s me.

34. When making a b i g d e c i s i o n , I u s u a l l y f e e l t h a t a d v i c e from o t h e r s i s i n t r u s i v e .

35. I t i s h a r d f o r me t o say " n o " to o t h e r p e o p l e ' s r e q u e s t s .

36 . I r e s e n t i t when p e o p l e t r y to d i r e c t my b e h a v i o r o r a c t i v i t i e s .

37 . I become u p s e t when something happens to me and t h e r e ' s nobody around to t a l k t o .

38 . P e r s o n a l q u e s t i o n s f rom o t h e r s u s u a l l y f e e l l i k e a n i n v a s i o n o f my p r i v a c y .

39 . I am most c o m f o r t a b l e when I know my b e h a v i o r i s what o t h e r s e x p e c t o f me.

40 . I am v e r y upse t when o t h e r peop le o r c i r c u m s t a n c e s i n t e r f e r e w i t h my p l a n s .

4 1 . I o f t e n l e t p e o p l e t a k e advantage o f me.

89

S t r o n g l y S l i g h t l y S l i g h t l y Strong D i s a g r e e D i sagree D i s a g r e e Agree Agree Agree

42. I r a r e l y t r u s t the a d v i c e 1 2 3 4 5 6 o f o t h e r s when making a b i g d e c i s i o n .

43 . I become v e r y upset when a 1 2 3 4 5 6 f r i e n d breaks a date o r f o r g e t s to c a l l me as p l a n n e d .

44. I become upset more than most 1 2 3 4 5 6 p e o p l e I know when l i m i t s a r e p l a c e d on my p e r s o n a l independence and freedom.

45. I j u d g e m y s e l f based on how 1 2 3 4 5 6 I t h i n k o t h e r s f e e l about me.

46 . I become upse t when o t h e r s 1 2 3 4 5 6 t r y t o i n f l u e n c e my t h i n k i n g on a p r o b l e m .

47 . I t i s h a r d f o r me to l e t 1 2 3 4 5 6 p e o p l e know when I am angry w i t h them.

48. I f e e l c o n t r o l l e d when o t h e r s 1 2 3 4 5 6 h a v e a say i n my p l a n s .

90

Append ix E

Percept ions of Part ic ipants

Self-or ientat ion items

(All i tems rated on 1 [not at all] - 7 [very much] Likert sca le)

My partner s e e m e d to have their own objectives in this task.

My partner was more concerned about their own goa ls rather than mine in this task.

My partner didn't s e e m to respond to my suggest ions.

I felt that my partner didn't want to take any direction from me.

M y partner wasn' t very sensi t ive to my efforts.

I found my partner irritating and hard to get a long with.

My partner wanted to get through this task quickly.

My partner s e e m e d somewhat brusque or abrupt.

My partner negated my suggest ions.

My partner wanted to make their own dec is ions.

Interpersonal deference descr iptors

(All i tems rated on 1 [not at all] - 7 [very much] Likert scale)

M y partner needed help and encouragement .

I felt that my partner was trying hard to get a long with me.

My partner wanted reassurance from me.

My partner looked to me for accep tance of their suggest ions.

My partner did not assert their pre ferences but deferred to mine.

My partner didn't s e e m to want to take the initiative.

M y partner wanted me to like them.

My partner w a s concerned about how I wou ld evaluate them.

My partner took pa ins to be agreeab le .

My partner greeted my suggest ions with p leasure and approval .

My partner really appreciated my help.

Appendix F

Descriptive Statistics of Dependent Variables

Means and Standard Deviations of Dependent Variables

Entire Sample Depressed Controls

(N=82) (n=41) (n=41)

Variables

DFI 3.58(1.19) 2.97(1.11) 4.19 (.94)

Interpersonal

Deference 41.94(9.60) ' 41.71 (11.00) 42.17(8.11)

Self-Orientation 32.01 (9.19) 35.78(9.12) 28.24(7.65)

Positive Behaviour

Index 16.42(5.08) 13.89(3.98) 18.95(4.83)

Note. DFI=Desire for future interactions. Means are followed by standard deviations in brackets.

9 3

Intercorrelations of Dependent Measures

Variables DFI Interpersonal • Self- Positive

Deference Orientation Behaviour Index

DFI

Interpersonal .08 —

Deference

Self-Orientation -.72** -.39**

Positive

Behaviour-Index .71** .04 -.63'

Note. ** p. <.001. DFI=Desire for future interactions