Personality and Depression: Explanatory Models and Review ...faculty.csusm.edu/sbufferd/Klein, Kotov, Bufferd 2011 Personality and... · Personality and Depression: Explanatory Models

CP07CH11-Klein ARI 24 February 2011 7:50

Personality and Depression:Explanatory Models andReview of the EvidenceDaniel N. Klein, Roman Kotov, and Sara J. BufferdDepartments of Psychology and Psychiatry, State University of New York at Stony Brook,Stony Brook, New York 11794-2500; email: [email protected]

Annu. Rev. Clin. Psychol. 2011. 7:269–95

First published online as a Review in Advance onDecember 6, 2010

The Annual Review of Clinical Psychology is onlineat clinpsy.annualreviews.org

This article’s doi:10.1146/annurev-clinpsy-032210-104540

Copyright c! 2011 by Annual Reviews.All rights reserved

1548-5943/11/0427-0269$20.00

Keywordstraits, temperament, mood disorders, neuroticism, extraversion

AbstractUnderstanding the association between personality and depression hasimplications for elucidating etiology and comorbidity, identifying at-risk individuals, and tailoring treatment. We discuss seven major modelsthat have been proposed to explain the relation between personality anddepression, and we review key methodological issues, including studydesign, the heterogeneity of mood disorders, and the assessment of per-sonality. We then selectively review the extensive empirical literatureon the role of personality traits in depression in adults and children.Current evidence suggests that depression is linked to traits such asneuroticism/negative emotionality, extraversion/positive emotionality,and conscientiousness. Moreover, personality characteristics appear tocontribute to the onset and course of depression through a variety ofpathways. Implications for prevention and prediction of treatment re-sponse are discussed, as well as specific considerations to guide futureresearch on the relation between personality and depression.

269

Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


Temperament:generally used todescribe personality inchildhood; largelyinterchangeable withthe term “personality”

Neuroticism:a tendency to copepoorly with stress andto experience feelingsof sadness, anxiety, andanger

ContentsINTRODUCTION . . . . . . . . . . . . . . . . . . 270THE CONSTRUCT OF

PERSONALITY . . . . . . . . . . . . . . . . . . 270MODELS OF PERSONALITY

AND DEPRESSION . . . . . . . . . . . . . . 271Classical Models of

Personality-DepressionRelations . . . . . . . . . . . . . . . . . . . . . . . 271

Dynamic Models ofPersonality-DepressionRelations . . . . . . . . . . . . . . . . . . . . . . . 274

METHODOLOGICAL ISSUES . . . . . 275Study Design . . . . . . . . . . . . . . . . . . . . . . 275Heterogeneity of Depressive

Disorders . . . . . . . . . . . . . . . . . . . . . . . 276Assessment . . . . . . . . . . . . . . . . . . . . . . . . 276

AFFECTIVE TEMPERAMENTS . . . . 277PERSONALITY TRAIT

DIMENSIONS . . . . . . . . . . . . . . . . . . . 279The Five-Factor Model . . . . . . . . . . . . 279Psychobiological Models . . . . . . . . . . . 283Clinical Traits . . . . . . . . . . . . . . . . . . . . . 284

CHILD TEMPERAMENT . . . . . . . . . . 285CLINICAL IMPLICATIONS . . . . . . . . 286

Prevention . . . . . . . . . . . . . . . . . . . . . . . . 286Treatment Response . . . . . . . . . . . . . . . 286

CONCLUSIONS AND FUTUREDIRECTIONS . . . . . . . . . . . . . . . . . . . . 287

INTRODUCTIONThe hypothesis that depression is linked to per-sonality can be traced to antiquity, when Hip-pocrates, and later Galen, argued that particular“humors” were responsible for specific per-sonality types and forms of psychopathology.In this article, we discuss the major conceptualmodels that have been proposed to explain theassociation between personality and depression,comment on some important methodologicalissues, and selectively review the empiricalliterature. Due to space limitations, we limitour review to nonbipolar forms of depression.

This literature has developed along severaldistinct lines: (a) early clinical psychiatrists’

descriptions of affective temperaments;(b) research on the structure and neurobiologyof personality; (c) psychoanalytic and cognitive-behavioral theory and observations; and(d ) developmental psychologists’ work ontemperament. In recent years, there has beensubstantial convergence between these linesof work, and it is increasingly possible to viewthem within a single integrative framework.Understanding the associations between per-sonality and depression has a number of poten-tially important implications for research andpractice. First, personality traits associated withemotional experience, expression, and regula-tion may be intermediate phenotypes that pro-vide more tractable targets for genetic and neu-robiological research than depressive diagnoses(Canli 2008). Second, personality may be usefulin identifying more homogeneous subgroupsof depressive disorders that differ in develop-mental trajectories and etiological influences(e.g., Beck 1983). Third, tracing the pathwaysbetween personality and depressive disorderscan help elucidate more proximal processesinvolved in the development of mood disorders(Compas et al. 2004, Klein et al. 2008a, Lahey2009). Fourth, personality may be useful in tai-loring treatment (Zinbarg et al. 2008) and pre-dicting treatment response (Quilty et al. 2008a).Fifth, temperament/personality may providea means to identify at-risk individuals whocould benefit from prevention and earlyintervention efforts (Kovacs & Lopez-Duran2010). Finally, there is substantial comorbiditybetween depressive disorders and other formsof psychopathology. Some personality traits,such as neuroticism, are associated with mul-tiple psychiatric conditions. Thus, personalitycould help explain patterns of comorbidityand point toward more etiologically relevantclassification systems (Brown & Barlow 2009,Kotov et al. 2007, Watson 2009).

THE CONSTRUCT OFPERSONALITYBefore addressing the relation between per-sonality and depression, several conceptual

270 Klein · Kotov · Bufferd

Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


issues regarding the construct of personalityshould be considered. First, personality hastraditionally been conceptualized as havingtwo components: temperament, which refersto biologically based, early-emerging, stableindividual differences in emotion and its regu-lation, and character, which refers to individualdifferences due to socialization. However,the distinctions between these constructs arequestionable, as a large body of evidence hasaccumulated indicating that personality traitshave all the characteristics of temperament, in-cluding strong genetic and biological bases andsubstantial stability over the lifespan (Krueger& Johnson 2008, Watson et al. 2006). Hence,the terms “personality” and “temperament”are now often used interchangeably (Caspi& Shiner 2006, Clark & Watson 1999). Asmost research on personality in childhoodhas been conducted under the temperamentrubric, in this review we refer to this workusing the term “temperament” and reserve theterm “personality” for discussing the literatureon adolescents and adults. However, this isintended to reflect traditional usage ratherthan a conceptually meaningful distinction.

Second, a variety of personality classifica-tions have been proposed over the past century,but in the 1980s they were integrated in aconsensus taxonomy, the Five-Factor Model(FFM). The FFM recognized that personalityis ordered hierarchically from a large numberof specific traits to five general characteristics(Digman 1994, Goldberg 1993, Markon et al.2005). These “Big Five” traits are neuroticism,extraversion, conscientiousness, agreeableness,and openness to experience. Importantly,the FFM can be further reduced to threedimensions of negative emotionality, positiveemotionality, and disinhibition versus con-straint that form the next level of the personalityhierarchy (Clark & Watson 1999, Markonet al. 2005). This “Big Three” model is used instudies of temperament as well as personality,although disinhibition is often labeled aseffortful control in the child literature (Caspi &Shiner 2006, Rothbart & Bates 2006). The BigFive and Big Three schemes are closely related,

Five-Factor Model(FFM): hierarchicalpersonality taxonomywith five general traitsat its apex:neuroticism,extraversion,conscientiousness,agreeableness, andopenness to experience

Extraversion: atendency to engage theenvironment and otherpeople with vigor andenthusiasm

Conscientiousness: atendency to approachtasks in a planful anddeliberate manner andto be reliable andself-disciplined

N/NE: neuroticism/negative emotionality

E/PE: extraversion/positive emotionality

with neuroticism being essentially identical tonegative emotionality and extraversion cor-responding to positive emotionality (Clark &Watson 1999, Markon et al. 2005); we refer tothese two dimensions as neuroticism/negativeemotionality (N/NE) and extraversion/positiveemotionality (E/PE), respectively. Disinhi-bition does not have an exact counterpart inthe FFM but instead reflects a combination oflow conscientiousness and low agreeableness.Finally, openness to experience is outside theterritory covered by the Big Three.

Third, there is increasing recognitionthat temperament and personality are not afixed, static set of characteristics, but ratherare dynamic constructs that develop over thelifespan and change in response to maturationand life circumstances (Fraley & Roberts 2005,Rothbart & Bates 2006). For example, althoughthe rank-order stability of most personalitytraits is in the moderate range, it increasesover the course of development (Roberts &DelVecchio 2000). In addition, mean levelsof conscientiousness and some facets of E/PEincrease, and levels of N/NE decrease, overtime, particularly in young adulthood (Robertset al. 2006). A number of processes contributeto stability and change of personality. For ex-ample, genes are a major influence on stability(Krueger & Johnson 2008, Kandler et al. 2010).In addition, people often select, create, and con-strue environments in ways that reinforce andmaintain their initial trait dispositions (Caspi &Shiner 2006). However, life stressors and majorshifts in social roles and relationships can con-tribute to personality change (Fraley & Roberts2005, Kandler et al. 2010). We consider theimplications of these processes for the relationbetween personality and depression below.

MODELS OF PERSONALITYAND DEPRESSION

Classical Models ofPersonality-Depression Relations

A variety of models of the relation betweenpersonality and mood disorders have been

www.annualreviews.org • Personality and Depression 271

Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


Table 1 Summary of key predictions of the classic models

Model Predictions about a target trait and its relation to depressionCommon cause Shared etiology accounts for the observed associationContinuum/spectrum Similar etiology; association is fairly specific and nonlinearPrecursor Similar etiology; predicts depression onsetPredisposition Predicts depression onset; other variables mediate or moderate this linkPathoplasticity Predicts variation in presentation or outcome of depression above and beyond other baseline

characteristicsConcomitants Is altered during a depressive episode but returns to premorbid level afterConsequences/scars Is altered during and after a depressive episode

proposed (e.g., Akiskal et al. 1983, M.H. Kleinet al. 1993, Krueger & Tackett 2003). Theseproposed relations include: (a) personality anddepressive disorders have common causes;(b) personality and depressive disorders forma continuous spectrum; (c) personality is a pre-cursor of depressive disorders; (d ) personalitypredisposes to developing depressive disor-ders; (e) personality has pathoplastic effectson depression; ( f ) personality features arestate-dependent concomitants of depressiveepisodes; and ( g) personality features areconsequences (or scars) of depressive episodes.The distinctions between some of these ac-counts are subtle (cf. Kendler & Neale 2010),and other models, as well as combinationsof these scenarios, are plausible. However,these seven models provide a useful conceptualframework for approaching the issue.

These models can be divided into threegroups. The first three models (common cause,continuum/spectrum, and precursor) view per-sonality and depression as having similar causalinfluences but do not see one as having acausal influence on the other. The fourth andfifth models (predisposition and pathoplastic-ity) hold that personality has causal effects onthe onset or maintenance of depression. Finally,the sixth and seventh models (concomitantsand consequences) view depression as having acausal influence on personality. These models,and their unique predictions, are summarizedin Table 1.

The common cause model views personalityand depressive disorders as distinct entities

that arise from the same, or at least an over-lapping, set of etiological processes. From thisperspective, personality and depression are notdirectly related; rather, the association is dueto a shared third variable. The common causemodel would be supported by evidence thatpersonality traits and depression have sharedetiological influences.

The continuum/spectrum model em-phasizes the conceptual overlap betweendepressive disorders and certain personalitytraits and argues for a fundamental continuitybetween them. A depressive diagnosis isthought to simply identify individuals whohave the most extreme scores on a relevanttrait. Like the common cause model, thecontinuum/spectrum model assumes thatpersonality and depression arise from a similar,if not identical, set of causal factors. However,the continuum/spectrum model goes further inpositing that the association between the traitand disorder should be fairly specific becausethey are on the same continuum.1 Moreover,this association is expected to be nonlinear,so that almost nobody below the definitionalthreshold on the trait has the diagnosis butnearly everyone above the threshold meets thecriteria. Thus, the continuum/spectrum model

1There may not be complete specificity owing to diagnosticheterogeneity. As discussed below, depression is a hetero-geneous disorder with multiple etiological pathways (equi-finality). A personality trait may be part of only one of thepathways. In contrast, multifinality, in which the trait is as-sociated with multiple disorders, is less consistent with thecontinuum/spectrum view.


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


would be supported by evidence that the traitand depression are associated with the same eti-ological influences and that the trait-disorderrelationship is fairly specific and nonlinear.

The precursor model views personality asan early manifestation or “forme fruste” of de-pressive disorder. Like the common cause andcontinuum/spectrum accounts, the precursormodel posits that personality and depressivedisorders are caused by similar etiologic factors.Also like the continuum/spectrum account, itimplies considerable phenomenologic similar-ity between the relevant trait and depression.However, the precursor model differs fromboth of these other models in that it assumesa particular developmental sequence, with thepersonality traits being evident prior to the on-set of depressive disorder. In other words, boththe common cause and continuum/spectrummodels assume a fixed clinical expression astraits or disorder, whereas the precursor modelimplies escalation from traits to disorder withinindividuals over time. Support for the precursormodel would come from evidence that the traitand depression are associated with the sameetiological influences and that individuals withhigh levels of the trait are at increased risk fordeveloping the disorder over time.2

The common cause, continuum/spectrum,and precursor models do not posit causalrelations between personality and depression.In contrast, the predisposition model holds thatpersonality plays a causal role in the onset ofdepression. However, the predisposition modeloverlaps with the precursor model in that bothpropose that the relevant traits are evident priorto the onset of depressive disorder. The majordifference between these two accounts is that

2Application of the continuum/spectrum and precursor mod-els to depressive disorders is not straightforward. Person-ality traits are relatively stable, whereas depression is oftenepisodic. Existing formulations of the continuum/spectrummodel have not explained how stable trait characteristicsmanifest as an episodic illness. Similarly, the precursor modeldoes not account for why a stable trait would subsequentlydevelop into a nonstable depressive state. Thus, the contin-uum/spectrum and precursor models may provide a betterexplanation for chronic than episodic forms of depression.

Precursor model:considers personalityan early manifestationof the disorder

Predispositionmodel: posits thatpersonality is distinctfrom psychopathologyand plays a causal rolein its development

Pathoplasticitymodel: posits thatpersonality influencesthe expression of thedisorder after onset

the precursor model assumes that personalityand depression derive from the same set of etio-logical processes, but the predisposition modelposits that the processes that underlie person-ality differ from those that lead to depression.Thus, the predisposition account implies acomplex interplay among risk factors involvingmoderation and/or mediation, and this is whatdistinguishes it from the precursor model.3

The most common example—the diathesis-stress model—conceptualizes personality asthe diathesis and stress as a moderator thatprecipitates the onset of depressive disorder.Alternatively, stress may be a mediator, sothat personality vulnerability leads to negativeexperiences (e.g., interpersonal rejection, jobloss), which in turn increase the probabilityof a depressive episode. A second differencebetween these models is that the predispositionmodel does not assume any phenomenologicallinks between personality traits and depressivesymptoms. Consequently, the predisposingtrait may not have any phenotypic similarity todepression. Thus, the two most critical sourcesof support for the predisposition model wouldinvolve demonstrating that individuals withthe trait are at increased risk for subsequentlydeveloping depression, and that other variablesplay a role in mediating or moderating thistransition.

The pathoplasticity model is similar tothe predisposition model in that it also viewspersonality as having a causal influence ondepressive disorder. However, rather than con-tributing to the onset of depression, the patho-plasticity model posits that personality influ-ences the expression of the disorder after onset.This influence can include the severity or pat-tern of symptomatology, course, and responseto treatment. The pathoplasticity model wouldbe supported by evidence that personality

3Although moderating and mediating variables play an ex-plicit and central role in the predisposition model, it shouldbe acknowledged that they are not incompatible with theprecursor account. That is, the escalation from personalitytraits to depressive disorders in the precursor model impliesthat additional variables (e.g., maturational or environmentalfactors) must be involved to precipitate the change.


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


explains variation among depressed individualsin their clinical presentation or outcome.

The final two models also assume that thereis a causal relation between personality anddepression. However, these models reversethe direction of causality. In the concomitants(or state-dependent) model, assessments ofpersonality are colored, or distorted, by theindividual’s mood state. This model impliesthat personality returns to its baseline formafter recovery from the episode. In contrast,the consequences (or scar) model holds thatdepressive episodes have an enduring effect onpersonality, such that changes in personalitypersist after recovery. These models would besupported by evidence that depression alterslevels of personality traits, either concurrently(concomitants model) or over the longer-term(consequences model).

Dynamic Models ofPersonality-Depression RelationsThe models above consider traits to be perfectlystable. As noted earlier, there is now extensiveevidence indicating that personality shows plas-ticity in childhood, with long-term test-retestcorrelations of r ! 0.35, and continues tochange across the lifespan, although personalityconsistency gradually increases up to r ! 0.75after the age of 50 (Roberts & DelVecchio2000). Models of personality-psychopathologyrelations can be expanded to recognize themalleability of traits (e.g., Ormel et al. 2001).For example, one can posit a dynamic precursormodel4 in which early temperament defines thebaseline level of risk but subsequent experiencesmodify personality liability to depression. Thismodel explains variability in disorder onset as afunction of the initial level of risk and steepnessof the trait trajectory over time. Given the evi-dence on patterns of personality continuity andchange (Roberts & DelVecchio 2000), it ap-

4This could also be called the dynamic continuum modelbecause once the dynamic element is introduced, it becomesvirtually impossible to distinguish the continuum/spectrumand precursor models.

pears likely that trait vulnerability is more mal-leable early in life, but significant life events canalter its trajectory even in old age. A depressivedisorder is thought to emerge when personalityliability crosses the threshold. Thus, individu-als who are born with an elevated personalityliability or those with a rapidly increasing traittrajectory would have a childhood onset of thedisorder, whereas those with a more slowlyincreasing trait trajectory would not cross thethreshold until much later, if ever. Moreover,a pathological trait trajectory may be checkedor reversed by positive experiences (Ormel &de Jong 1999). In fact, personality generallytends to change in a more adaptive directionwith age (Roberts et al. 2006), although thispattern is not universal ( Johnson et al. 2007).This may help to explain why the probability offirst-episode depression peaks in adolescence,as trait deviance is more common at that age.

Similarly, the predisposition model can beexpanded to recognize personality change. Thisdynamic predisposition model (Ormel & deJong, 1999, Ormel et al. 2001) acknowledgestransactions between personality and the envi-ronment and integrates them with the environ-mental moderation and mediation mechanismsof the classic predisposition model. In the envi-ronmental moderation version of this account,negative life experiences influence not only de-pression onset but also levels of trait vulnera-bility (Middledorp et al. 2008). This increase inpersonality liability may then lead to additionallife stress. If this vicious cycle is perpetuatedunchecked, personality liability would continueto increase, and at some point, a negative lifeevent could overwhelm coping capabilities andelicit a depressive disorder. Importantly, and incontrast to the dynamic precursor model, in thisaccount maladaptive traits alone are not suffi-cient to cause depression, and an environmentaltrigger is necessary.

The vicious cycle of increasing trait vulner-ability and stress exposure does not necessarilyindicate that personality per se influencesdepression onset. Indeed, certain traits mayincrease stress exposure but have no effect ondepression otherwise (e.g., it is possible that low


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


conscientiousness does not cause depressiondirectly but leads to depressogenic experiences,such as academic difficulties, job loss, andrelationship problems; Roberts et al. 2007),consistent with the environmental-mediationpathway.

Dynamic models offer richer and more com-plete accounts of the role of personality in theonset of depression. Moreover, it is importantto recognize that depressive disorders have beenlinked to multiple traits (as reviewed below),and it is likely that different personality charac-teristics contribute through different pathways.

METHODOLOGICAL ISSUESA number of methodological issues must beconsidered in evaluating the relation betweenpersonality and mood disorders, including(a) study design, (b) heterogeneity of depressivedisorders, and (c) assessment of personality.

Study DesignA number of research designs can be usefulin studying the relation between personalityand depressive disorders. The common cause,continuum/spectrum, precursor, and predispo-sition models would all be supported by familystudies demonstrating personality differencesbetween nonaffected relatives of probandswith and without a history of depression. Thecommon cause, continuum/spectrum, and pre-cursor models would be supported by twin andgenetic association studies demonstrating thatthe same genes predispose to both personalityand depressive disorders. The precursor andpredisposition models posit that personality ab-normalities are trait markers and hence shouldbe present prior to the onset, and after recoveryfrom, depressive episodes. Hence, these modelscan be tested by comparing individuals with ahistory of depression that is currently in remis-sion to persons with no history of depressionon relevant personality traits. An even strongerapproach to testing the precursor and predispo-sition models is to use prospective longitudinalstudies of persons with no prior history of

mood disorder to determine whether particularpersonality traits predict the later onset ofdepressive disorder. Although no single designcan distinguish among these four models, thecombination of designs can bolster the casefor particular accounts. For example, findingsubstantial common genetic variance in twinstudies, but no evidence of developmental se-quencing in longitudinal studies, would supportthe common cause and continuum/spectrummodels. In turn, these two models could becompared by examining the specificity of theassociation between trait and disorder andwhether there is a nonlinear relation betweentrait level and probability of disorder. On theother hand, if there were evidence of develop-mental sequencing in longitudinal studies aswell as substantial common genetic variancein twin studies (or overlap of other etiologicalfactors in other designs), it would support theprecursor model (particularly if the trait wasalso phenomenologically similar to depression).In contrast, developmental sequencing but lessshared genetic (or other etiological) variancewould support the predisposition model. Alsocrucial for the predisposition model is evidencefrom longitudinal studies demonstrating thatother variables (e.g., life stress) moderate ormediate the association between personalityand subsequent depression.

The pathoplasticity model can be evaluatedin longitudinal studies of persons with depres-sive disorders by examining the associationsamong personality traits and clinical features,course, and treatment response. Specifically,the pathoplasticity model posits that the traitwould predict these outcomes even aftercontrolling for initial illness severity and otherprognostic factors. Of note, an alternative ex-planation of such results is that the personalitytrait is a marker for a more severe, chronic,or etiologically distinct subgroup, rather thanhaving a causal influence on the expressionof the disorder. A multiwave follow-up ofindividuals with a depressive disorder could behelpful in ruling out this possibility. If the traitinfluences the disorder course directly, ratherthan because it is an indicator of a latent


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


Dysthymic disorder:a condition defined bychronic but relativelymild feelings ofdepression lasting atleast two years

disorder class, changes in personalityscores should predict subsequent changes inoutcomes.

The concomitants model can be testedthrough cross-sectional studies comparing per-sons who are currently depressed, persons whohave recovered from depressive episodes, andhealthy controls. An even better approach isto conduct longitudinal studies assessing indi-viduals when they are in a depressive episodeand again after they have recovered. If person-ality measures are abnormal during depressiveepisodes but not after recovery, it would sug-gest that they are concomitants of the depressedstate. Multilevel analyses can also be used toseparate personality variance into trait and statecomponents and to test whether state varianceis associated with concurrent measures of de-pression (e.g., Duncan-Jones et al. 1990).

The consequences (or scar) hypothesis canbe evaluated by assessing persons before andafter a first depressive episode. If personalitydeviance is much greater after the episode hasremitted, it would suggest that scarring hasoccurred.

Testing dynamic theories requires longitu-dinal data with at least three assessment points.These assessments should measure relevantcontextual factors (e.g., life stress) in additionto depression and personality to allow the ex-amination of dynamic and transactional effects.Multilevel modeling and structural equationmodeling offer powerful approaches to evalu-ating such effects with longitudinal data.

Heterogeneity of DepressiveDisordersThe depressive disorders are almost certainlyetiologically heterogeneous, reflecting the con-vergence of multiple developmental pathways.Hence, it is likely that the role of personalityfactors and, as suggested above, the applicabil-ity of different models of the relation betweenpersonality and depression differ for differentforms of depression. The current classificationsystem for depressive disorders is based onclinical features and is a poor approximation of

etiological distinctions. Nonetheless, it is im-portant to consider whether the role of person-ality varies as a function of the specific depres-sive diagnosis (e.g., major depressive disorder,dysthymic disorder), subtype (e.g., psychotic,melancholic, atypical), and clinical character-istics such as age of onset, recurrence, andchronicity. Failure to take heterogeneity intoaccount may obscure important personality-depression associations. Conversely, person-ality may provide a basis for identifying morehomogeneous subgroups within the depressivedisorders. Unfortunately, few studies of person-ality and depressive disorders have attemptedto take this heterogeneity into account.

It is important to note, however, thatassociations between personality and specificsubtypes and clinical characteristics do notnecessarily indicate etiological heterogeneity.Instead, they could reflect pathoplasticity,in which personality influences symptompresentation and/or course, but the primaryetiological process is the same, or they could re-flect differential severity of subtypes that resultsin quantitative differences in their trait profiles.

Finally, a significant source of heterogeneityin depression is comorbidity with other formsof psychopathology. Given the high rates ofcomorbidity, particularly with the anxietydisorders, associations between personalityand depressive disorders may actually reflectthe relation of personality with a co-occurringnonmood disorder. Indeed, personality may bea third variable that explains broad patterns ofcomorbidity among many disorders. For exam-ple, recent hierarchical models of classificationposit that trait dispositions such as N/NEaccount for much of the comorbidity betweendepression and other disorders (Griffith et al.2010, Kotov et al. 2007). Thus, it is importantfor researchers to consider whether traitshave specific relations with depression overand above more general associations with thebroader group of internalizing disorders.

AssessmentTemperament/personality can be assessed us-ing a variety of methods, including self-report


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


inventories, semistructured interviews, infor-mants’ reports, and observations in naturalis-tic settings and the laboratory. Unfortunately,most of the literature examining the associa-tion between personality and depressive dis-orders has assessed personality via self-report.This is potentially problematic because self-reports of personality can be complicated bycurrent mood state, limited insight, responsestyles, and the difficulty of distinguishing traitsfrom the effects of stable environmental con-texts (Chmielewski & Watson 2009). In addi-tion, when the same individual provides infor-mation on both personality and depression, ashas been the case in almost all studies in thisarea, common method variance can inflate as-sociations. Hence, there is a need for greateruse of informant report and observationalmeasures.

A second issue concerns the overlapbetween some personality constructs andpsychopathology (Lahey 2004). For example,many items on N/NE scales are similar to de-pressive symptoms (Ormel et al. 2004b). Thiscan inflate associations between measures ofpersonality and depression. On the other hand,personality and symptom assessments usuallyhave different time frames, with trait scalesreflecting long-standing patterns and depres-sion measures tapping more recent experiences(e.g., past week, past month). This trait versusstate distinction parallels that between person-ality and other related constructs. For example,measures of N/NE and negative affect havenearly identical content but are distinguishedby their time frames (Watson 2000). Thus, thedegree to which this content overlap threatensthe validity of personality-psychopathologyresearch depends, at least in part, on theduration/chronicity of the disorders of interest.The extent to which this is a concern also de-pends on one’s model of personality-depressionrelations. From the continuum/spectrum per-spective, personality and depression are variantsof the same phenomenon, so the two constructsshould overlap. In contrast, the predispositionmodel views personality and depression asdistinct domains, so from this perspective it

DSM-IV: Diagnosticand Statistical Manualof Mental Disorders,Fourth Edition

is important to define and assess these sets ofconstructs as independently as possible and tojudiciously delineate their time frames.

AFFECTIVE TEMPERAMENTSThe classical European descriptive psy-chopathologists in the late-nineteenth andearly-twentieth centuries observed that manypatients with mood disorders, as well astheir relatives, exhibited particular patternsof premorbid personalities that appeared tobe attenuated versions of their illnesses. Forexample, Kraepelin (1921) described fourpatterns of personality that he consideredthe “fundamental states” underlying manic-depressive illness: depressive, manic, irritable,and cyclothymic temperament. He believedthat these were precursors or “rudimentaryforms” of the major mood disorders. Schneider(1958) described similar types; however, heviewed them as personality disorders that werenot necessarily related to the mood disorders.Two variants of these types, cyclothymicdisorder and dysthymic disorder, are includedas mood disorder diagnoses in the Diagnosticand Statistical Manual of Mental Disorders,Fourth Edition (DSM-IV; Am. Psychiatr. Assoc.1994). However, these disorders are definedas fairly severe conditions, with the criteriaemphasizing symptomatology rather thanpersonality traits. As a result, these categoriesappear to be limited to the more severe,symptomatic manifestations of the affectivetemperaments described by Kraepelin andSchneider (Akiskal 1989).

On the basis of Kraepelin’s and Schneider’sdescriptions, Akiskal (1989) proposed formalcriteria for the affective temperament types,and he and his colleagues developed interviewand self-report measures of these constructsthat have been applied in a number of settingsand cultures (e.g., Akiskal et al. 2005). Akiskal’swork also provided the basis for including de-pressive temperament as a personality disorderin the DSM-IV appendix. Of the four affectivetemperament types, depressive temperamenthas been the most systematically studied in


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


Major depressivedisorder (MDD):a conditioncharacterized byepisodes of depressedmood or loss ofinterest or pleasurelasting at least twoweeks

relation to nonbipolar depressive disorders.The terms “depressive temperament,” “depres-sive personality,” and “depressive personalitydisorder” have been used interchangeablyin the literature to refer to the followingconstellation of traits: introversion, passivity,and nonassertiveness; gloominess, cheer-lessness, and joylessness; self-reproach andself-criticism; pessimism, guilt, and remorse;being critical and judgmental of others; consci-entiousness and self-discipline; brooding andgiven to worry; and feelings of inadequacy andlow self-esteem.

Data on the nature of the relation betweendepressive personality and depressive disordersare consistent with most of the causal modelsdescribed above, illustrating the complexityof the associations between personality anddepression. The strongest support for the com-mon cause and continuum/spectrum modelsderives from twin and family studies. In a largetwin study, Ørstavik et al. (2007) found that de-pressive personality and major depressive disor-der (MDD) shared substantial genetic variance,although there was evidence for unique geneticfactors as well. Family studies indicate thatindividuals with depressive personality have anincreased rate of mood disorders in their first-degree relatives (e.g., Klein & Miller 1993).In addition, patients with MDD, particularlythose with chronic forms of depression, haveelevated levels of depressive personality traitsin their first-degree relatives (Klein 1999).

Consistent with Kraepelin’s (1921) retro-spective observations, prospective longitudinaldata indicate that depressive personality traitsprecede the onset of depressive disorders. Kwonet al. (2000) found that young women with de-pressive personality and no comorbid Axis Iand II disorders had a significantly increasedrisk of developing dysthymic disorder (but notMDD) over the course of a three-year follow-up. Taken together with the twin and familystudies, these findings provide compelling sup-port for the precursor model. In addition, inlight of the conceptual issue regarding traits andstates raised above for the continuum/spectrumand precursor models, it is noteworthy that

depressive personality is most closely associatedwith chronic forms of depression at both thefamily and individual levels.

Evidence also supports the predisposition,pathoplasticity, and consequences models.Rudolph & Klein (2009) reported that youthwith elevated levels of depressive personalitytraits experienced a significant increase indepressive symptoms 12 months later. Whileconsistent with the precursor model, it is note-worthy that this association was moderatedby pubertal status and timing. Thus, youthwith elevated levels of depressive personalitytraits and more advanced pubertal statusand earlier pubertal timing experienced thegreatest increase in depressive symptoms. Thissupports the predisposition model, suggestingthat depressive personality traits confer vulner-ability to depression in the presence of othermaturational and psychosocial processes.

Depressive personality also appears to havea pathoplastic effect on the course of depres-sive disorders, predicting poorer outcomes andresponse to treatment (Laptook et al. 2006,Ryder et al. 2010). Moreover, Rudolph & Klein(2009) recently reported preliminary supportfor the consequences model, at least in youth.They found that in a sample of early ado-lescents, higher levels of depressive symptomspredicted an increase in depressive personalitytraits 12 months later. Finally, the limited ev-idence available suggests that semi-structuredinterview assessments of depressive personalitytraits are not influenced by a depressive episode(Klein 1990), arguing against the concomitantsmodel.

Although the work on affective temper-aments is important in understanding thedevelopment of depressive disorders, it isunlikely that these types actually reflect basictemperamental processes that originate inearly childhood, as their defining featuresinclude a number of developmentally complexcognitive and interpersonal characteristics.Instead, these temperament types are morelikely to be intermediate outcomes that reflectthe interaction of more basic temperamenttraits that are elaborated over development in


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


conjunction with early socialization and otherenvironment influences.

In recent years, considerable evidencehas accumulated indicating that depressivepersonality is associated with several of thebasic personality trait dimensions discussedbelow, particularly high N/NE and low E/PEand a number of their facets (e.g., Huprich2003, Vachon et al. 2009).

PERSONALITY TRAITDIMENSIONSThe affective temperaments are conceptualizedwithin a categorical framework. In contrast,most of the other work on personality anddepression views personality in dimensionalterms. In this section, we focus on the FFM,but we also briefly consider Gray’s (1994) andCloninger and colleagues’ (1993) psychobi-ological models and several additional traitsfrom the clinical literature (e.g., self-criticism,dependency, and rumination). This sectionfocuses primarily on studies of adults andadolescents using self-report measures ofpersonality. Studies of younger children usingobservational measures of temperament arereviewed in a later section.

The Five-Factor ModelCross-sectional associations. In their influ-ential theory of personality and depression,Clark & Watson (1999, Clark et al. 1994)posited that depressive disorders are character-ized by high levels of N/NE and low levels ofE/PE. A large number of cross-sectional stud-ies have evaluated these relations as well as thelinks between depression and the other FFMdimensions. Kotov et al. (2010) recently con-ducted a meta-analysis of this literature, whichrevealed that MDD is associated with very highN/NE (Cohen’s d = 1.33) and low conscien-tiousness (d = "0.90). The link to low E/PEwas more modest (d = "0.62) and inconsis-tent, with some studies finding positive effects.The associations with the other two traits wereweak and unremarkable. The N/NE finding is

consistent with expectations, but the effect forE/PE was smaller and that for conscientious-ness was larger than anticipated. Dysthymic dis-order exhibited a more extreme profile with re-markably strong and consistent links to E/PE(d = "1.47), N/NE (d = 1.93), and conscien-tiousness (d = "1.24). This is not surprisingas dysthymic disorder is thought to be moretrait-like than MDD, and a greater contribu-tion from personality might be expected.

To determine whether the observed per-sonality links are specific to depression, Kotovet al. (2010) also examined personality profilesof anxiety disorders. They found that with theexception of specific phobia, which had rela-tively weak associations with all five traits, allanxiety disorders showed stronger effects onN/NE, E/PE, and conscientiousness (averaged = 1.91, "1.05, "1.02, respectively) than didMDD. Several also scored above dysthymic dis-order on N/NE. Dysthymia had stronger asso-ciations than anxiety disorders on the other twotraits, but the differences were slight.

It is conceivable that more specific associ-ations were not evident because these analysesfocused on broad personality dimensions.Narrow traits that comprise the general di-mensions may have stronger associations withdepressive disorders. Indeed, self-harm—acomponent of N/NE that reflects propensity toself-deprecation and self-injury—was found tocontribute to depression even after controllingfor the broad traits, and this effect was specificrelative to other common mental disorders(Watson et al. 2006). With regard to E/PE,evidence is emerging that the positive affec-tivity facet, but not the sociability/extraversionfacet, is related to depression (Durbin et al.2005, Naragon-Gainey et al. 2009). This mayexplain the surprisingly modest associationbetween MDD and E/PE, if this generaltrait includes much variance not relevant todepression. Thus, facet-level research promisesto yield stronger and more specific evidence ofpersonality-depression links.

Evidence bearing on causal models. Be-cause most attempts to tease apart explanatory


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


models of the association between depressionand personality have focused on N/NE andE/PE, we consider only these two traits in thissection. The section is organized by the type ofresearch design used to address the models.

Personality during and after a depres-sive episode. Studies of personality and psy-chopathology may be complicated by the in-fluence of participants’ mood states on reportsof their personalities (the concomitants model).For example, many studies have found that in-dividuals with MDD report higher levels ofN/NE when they are depressed than whenthey are not depressed (Hirschfeld et al. 1983b,Kendler et al. 1993, Ormel et al. 2004a). Incontrast, the evidence for mood state effects onE/PE is weaker and less consistent (de Fruytet al. 2006, Kendler et al. 1993, Morey et al.2010). However, the influence of mood stateon personality should not be overstated. Eventhough levels of N/NE decline significantly af-ter remission from a depressive episode (i.e.,absolute stability), individuals’ relative posi-tions with respect to levels of N/NE (i.e., rank-order stability) tend to be moderately well pre-served (de Fruyt et al. 2006, Morey et al. 2010).Moreover, clinical trials suggest that changesin depressive symptoms are not necessarily ac-companied by changes in personality (Quiltyet al. 2008b, Tang et al. 2009).

Cross-sectional comparisons of remitted pa-tients and controls. A number of early stud-ies used remission designs, comparing patientswho had recovered from a depressive episode tonever-depressed controls or population normson self-rated personality traits. These studiesfound that E/PE is significantly lower in for-merly depressed patients than in healthy con-trols (Hirschfeld et al. 1983a, Reich et al. 1987),arguing against the concomitants model and infavor of the precursor, predisposition, and/orconsequences models. However, the results forN/NE were less consistent (Hirschfeld et al.1983a, Reich et al. 1987). This inconsistencymay be due to a number of factors, includ-ing insufficiently stringent criteria for recovery,

thereby possibly confounding personality andresidual symptoms; using normative data col-lected by other investigators, which may intro-duce demographic and sociocultural differencesbetween the formerly depressed and compari-son samples; and selection effects, as N/NE isassociated with a poorer course (discussed be-low) and thus samples of remitted depressivesmay include a disproportionate number withlow levels of this trait.

Personality before and after a depressiveepisode. Several studies have tested the conse-quences (or scar) hypothesis by comparing per-sonality measures in depressed individuals be-fore and after a MDD episode. The results ofthese studies have been inconsistent. Kendlerand colleagues reported increases in N/NE (butnot E/PE) after a depressive episode in twoseparate samples (Fanous et al. 2007, Kendleret al. 1993); however, other studies have foundthat N/NE and E/PE do not change from be-fore to after a MDD episode (e.g., Ormel et al.2004a, Shea et al. 1996). Importantly, the stud-ies reporting scarring used less stringent criteriafor recovery and shorter follow-ups, suggestingthat the findings may be due to residual symp-toms and/or that the scars dissipate over time.

Personality in relatives of depressed indi-viduals. A number of studies have tested thecommon cause, continuum/spectrum, precur-sor, and predisposition models by comparingpersonality traits in the never-depressed rela-tives of patients with mood disorders and never-depressed controls (e.g., Farmer et al. 2002,Hecht et al. 1998, Ouimette et al. 1996). Theresults have been mixed, with some studies re-porting higher N/NE and/or lower E/PE inthe never-depressed relatives of probands withmood disorders, and other studies reporting nodifferences. However, interpretation of thesestudies is complicated by two factors. First, per-sonality traits may not play the same role inrisk for depression among familial as nonfa-milial forms of depression. Second, there maybe selection biases in samples using well rela-tives who are already partly through the risk


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


period for mood disorder. Thus, those rela-tives with the strongest personality vulnerabil-ities may have already developed the disorderand be excluded from the study.

Twin studies. As discussed above, a valu-able approach to testing the common cause,continuum/spectrum, and precursor models isthrough twin studies. These studies indicatethat there are substantial associations betweenthe liabilities for N/NE and MDD, but onlyweak associations between the genetic liabili-ties for E/PE and MDD (Fanous et al. 2007;Kendler et al. 1993, 2006).

Prospective longitudinal studies. The most di-rect approach to testing the precursor and pre-disposition models is to conduct prospectivestudies of personality in never-depressed partic-ipants to determine whether personality char-acteristics predict the subsequent onset of de-pressive disorders. Several studies using largecommunity samples have reported that higherlevels of N/NE predict the onset of first lifetimeMDD episodes (de Graaf et al. 2002; Fanouset al. 2007; Kendler et al. 1993, 2006; Ormelet al. 2004a). In addition, several studies us-ing measures of other traits that overlap withN/NE or its facets have reported similar find-ings (Hirschfeld et al. 1989, Rorsman et al.1993). Although there is some evidence thatE/PE predicts the first onset of MDD (Kendleret al. 2006, Rorsman et al. 1993), it is muchweaker, and several studies have failed to findan association (Fanous et al. 2007, Hirschfeldet al. 1989, Kendler et al. 1993).

Personality and the subsequent course of de-pression. Finally, there is evidence that bothN/NE and E/PE have pathoplastic influenceson the course of depression after the onset ofthe disorder. For example, many studies havereported that higher N/NE and lower E/PEpredict a poorer course and response to treat-ment, although the findings regarding E/PE areslightly less consistent (de Fruyt et al. 2006,Duggan et al. 1990, Morris et al. 2009, Quilty

et al. 2008a, Tang et al. 2009). As noted above,however, these findings are also consistent withdiagnostic heterogeneity, such that personal-ity dysfunction is a marker for a more severeor etiologically distinct group. Indeed, there isevidence that the nonmelancholic subtype ischaracterized by more vulnerable personalitystyles than is melancholia and that chronic de-pressions are associated with higher N/NE andlower E/PE than is nonchronic MDD (Klein2008, Kotov et al. 2010).

Evidence relevant to dynamic models.Transactions between N/NE and environmen-tal contexts have received the most attentionin the literature (Ormel & de Jong 1999, vanOs & Jones 1999). N/NE shows reciprocalrelations with a range of significant life expe-riences, such as initiation and break-up of acommitted relationship, relationship quality,occupational attainment, and financial security(Neyer & Lehnart 2007, Roberts et al. 2003,Scollon & Diener 2006). Furthermore, N/NEhas been repeatedly implicated in the gen-eration of stressful life events (Kercher et al.2009, Lahey 2009, Middeldorp et al. 2008),which suggests an environmentally mediatedrelationship between this trait and depression.The environmentally moderated mechanismhas also received support, as several studiesfound that N/NE interacts with stressful lifeevents to predict first onset of major depression(Kendler et al. 2004, Ormel et al. 2001, van Os& Jones 1999).

E/PE has demonstrated bidirectional effectswith many significant social and occupationalexperiences (Neyer & Lehnart 2007, Robertset al. 2003, Scollon & Diener 2006). In addi-tion, a decrease in E/PE over time was foundto predict future internalizing problems (Vanden Akker et al. 2010). However, little atten-tion has been given to mechanisms underlyingthe association between this trait and depres-sion. Support for an environmentally mediatedeffect is limited and mixed (Middeldorp et al.2008, Wetter & Hankin 2009), and the environ-mental moderation model is largely untested,although there is some evidence that positive


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


affect moderates the effects of daily stressors ondepressive symptoms (Wichers et al. 2007).

Finally, conscientiousness may play an im-portant role in dynamic models of personality-depression relations. It has reciprocal associa-tions with family support, divorce, occupationalattainment, and job involvement (Roberts et al.2003, Roberts & Bogg 2004). Conscientious-ness is hypothesized to influence depressionby increasing exposure to negative life events(Anderson & McLean 1997, Compas et al.2004), but mediation and moderation effectshave not been tested.

Interactions between temperament dimen-sions. Finally, personality-depression rela-tions may be multivariate, rather than bivariate,with multiple traits interacting to influ-ence depression. Indeed, in their influentialmodel of personality and depression, Clark &Watson (1999, Clark et al. 1994) hypothesizedthat depression is characterized by high N/NEand low E/PE, raising the possibility that it isthe combination of the two traits that is particu-larly important in depressive disorders. A grow-ing number of studies have reported that the in-teraction of high N/NE and low E/PE predictssubsequent depressive symptoms or disordersin adults and youth (Gershuny & Sher 1998,Joiner & Lonigan 2000, Wetter & Hankin2009), although several studies have not foundsuch an interaction ( Jorm et al. 2000, Kendleret al. 2006, Verstraeten et al. 2009). The inter-action between N/NE and conscientiousness isalso of interest, as the latter construct includesaspects of self-regulation and effortful control(Rothbart & Bates 2006) and may therefore re-flect the ability to modulate one’s affective reac-tivity. Indeed, there is cross-sectional evidencethat effortful control moderates the associationbetween N/NE and depressive symptoms inadolescents (Verstraeten et al. 2009).

Summary and discussion. Cross-sectionalstudies have documented strong links ofdepressive disorders to N/NE, conscien-tiousness, and E/PE, although the lattereffect is substantial in dysthymic disorder but

only moderate in MDD. In fact, personalitygenerally appears to play a greater role indysthymia. None of these relations are specific,however, as anxiety disorders have very similartrait profiles. This observation argues againstthe continuum/spectrum model at least withregard to these broad dimensions. It may bepossible to find traits that are specific to de-pression by targeting lower-order personalitydimensions. Narrower traits may also explainthe surprisingly modest link between E/PE andMDD, as some, but not other, facets of thisgeneral dimension are relevant to depression.

The nature of relations between thesepersonality traits and depression is complex,and our understanding is still limited. N/NE,the most widely studied personality trait indepression, raises challenging conceptual andmethodological issues due to the overlapbetween some of its features and depressivesymptoms (Ormel et al. 2004b). Nonetheless,this cannot completely explain the associationbetween these constructs (Tang et al. 2009).N/NE is moderately influenced by clinical state(the concomitants model), shares commonetiological influences with MDD (commoncause, continuum/spectrum, and precursormodels), predicts the subsequent onset ofMDD (precursor and predisposition models),and influences the course of depression (patho-plasticity model). In addition, N/NE appearsto contribute to subsequent stress and adversityand increases the risk of depression in the faceof negative life events (predisposition model).Finally, it may also be changed by experience ofMDD episodes (consequences model), but theevidence for this is weaker and less consistent.

The role played by E/PE in depression is lessclear. Its cross-sectional association with dys-thymia is substantial, but its relation to MDD ismore modest. E/PE is not influenced by clinicalstate or changed by the experience of depres-sive episodes. It appears to be abnormally loweven during remission, which is consistent withthe continuation of trait deviance from the pre-morbid stage (precursor or predisposition ac-counts). Moreover, low E/PE tends to predicta poorer course of depression. However, the


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


degree of shared etiological influences betweenE/PE and MDD is low, and the evidence thatE/PE predicts the onset of MDD in prospectivelongitudinal studies is weak. As noted above,three possible reasons for the weaker and lessconsistent findings regarding E/PE are (a) itplays a greater role in some forms of depres-sion than others (e.g., chronic depressions);(b) only some facets of the broader trait (e.g.,low positive affective and approach motivation)are related to depression; and (c) E/PE maymake a greater contribution to depression bymoderating N/NE than as a main effect (Olinoet al. 2010).

Finally, there appears to be a strong negativeassociation between conscientiousness and de-pression, at least in cross-sectional studies. Thismay appear surprising in light of the positiveassociations discussed below between depres-sion and a number of other constructs thatare thought to be related to conscientiousness,such as behavioral inhibition system sensitivity,harm avoidance, perfectionism, and temper-amental behavioral inhibition. It is importantto note, however, that these latter constructsare more strongly associated with high N/NE,and in some cases, low E/PE, than with consci-entiousness (de Fruyt et al. 2000, Muris et al.2009a, Smits & Boeck 2006). Hence, it is likelythat their positive correlations with depressionare driven by their shared variance with highN/NE. Unfortunately, few studies testingcausal models of personality and depressionhave considered conscientiousness. However,evidence indicating that this trait may mod-erate the effects of N/NE on depression andthat it increases the likelihood of subsequentadversity that could then, in turn, producedepression suggests that further research onthe role of conscientiousness is warranted.

Psychobiological ModelsGray’s model. Gray’s (e.g., Gray 1994) influ-ential theory proposes that there are two majorneurobehavioral systems that underlie behav-ior: the behavioral activation system (BAS),which responds to signals of reward, and the be-

BAS: behavioralactivation system

BIS: behavioralinhibition system

havioral inhibition system (BIS), which is sensi-tive to cues for punishment. Although BAS andBIS differ conceptually and empirically fromE/PE and N/NE, their relations with depres-sion are thought to be similar. Thus, it has beenhypothesized that depression is associated withreduced BAS and/or heightened BIS sensitivity(Depue & Iacono 1989, Gray 1994). Althoughmuch of this work has focused on bipolar disor-der (e.g., Alloy et al. 2008, Johnson et al. 2008),several recent studies have examined self-reportmeasures of BAS and BIS sensitivity in MDD.Consistent with Gray’s model, compared withhealthy controls, currently depressed patientsreport lower levels of BAS and higher levels ofBIS, and patients with a past history of MDDreport lower levels of BAS (Pinto-Meza et al.2006). In addition, lower BAS sensitivity, butnot higher BIS sensitivity, is associated with apoorer course of MDD (e.g., Kasch et al. 2002,McFarland et al. 2006), suggesting that BASmay have a pathoplastic effect on depression.

Cloninger’s model. Cloninger (e.g.,Cloninger et al. 1993) has proposed amodel of personality that includes four tem-perament and three character dimensions.The temperament dimensions include noveltyseeking (an appetitive/approach system), harmavoidance (an inhibition/avoidance system),reward dependency (a system that is responsiveto signals of social approval and attachment),and persistence. The character dimensions areself-directedness (responsible, goal-directed),cooperativeness (helpful, empathic versushostile and alienated), and self-transcendence(imaginative, unconventional). Harm avoid-ance is conceptually and empirically associatedwith BIS, and novelty seeking and persis-tence are associated with BAS. Similarly,harm avoidance is positively correlated withN/NE and negatively associated with E/PE,self-directedness is negatively correlated withN/NE, and novelty seeking and persistenceare associated with E/PE (e.g., de Fruyt et al.2000).

A number of studies have reported that pa-tients with MDD report higher levels of harm


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


avoidance and lower levels of self-directednessthan do healthy controls (e.g., Celikel et al.2009). Most of the traits in Cloninger’s systemare influenced by the respondent’s mood state(e.g., Farmer et al. 2003); however, abnormallevels of harm avoidance and self-directednessare present even after remission (e.g., Smithet al. 2005). Increased harm avoidance andlower self-directedness are also characteristicof most anxiety disorders, indicating that theseeffects are not specific to MDD (Ongur et al.2005).

Few studies have explicitly tested the com-mon cause, precursor, predisposition, and con-sequences hypotheses for Cloninger’s model.Farmer et al. (2003) found that the never-depressed siblings of patients with MDD re-ported significantly greater harm avoidanceand less self-directedness than did the never-depressed siblings of healthy controls. In addi-tion, Cloninger et al. (2006) reported that in alarge community sample, high harm avoidanceand persistence and low self-directedness pre-dicted an increase in self-reported depressivesymptoms 12 months later. A larger number ofstudies have addressed the pathoplasticity hy-pothesis, albeit with mixed results. Low harmavoidance, self-directedness, and reward de-pendency have predicted a poorer response totreatment in some, but not all, studies; the otherdimensions have generally not been associatedwith course and treatment outcome ( Joyce et al.2007, Kennedy et al. 2005, Morris et al. 2009).

Clinical TraitsIndependent of the traditional personality field,clinical researchers have developed a number oftrait-like constructs to describe dispositions todepression. These clinical traits are similar inscope to personality facets, and their stability iscomparable to that of a typical personality di-mension (e.g., Kasch et al. 2001, Zuroff et al.2004). Also, factor analytic studies have shownthat most of these clinical traits can be success-fully incorporated in the personality taxonomyas components of neuroticism (Watson et al.2006). Next, we briefly discuss three of the most

studied constructs: ruminative response style,self-criticism, and dependency.

Ruminative response style, a tendency todwell on sad mood and thoughts (Nolen-Hoeksema 1991), is correlated with concurrentdepressive symptoms and predicts future symp-toms as well as increases in symptoms over time(Rood et al. 2009). Also, one study reportedthat ruminative response style prospectivelypredicts onset of MDD (Nolen-Hoeksema2000). The trait has also been linked to anxietydisorders, but the association with depressionis appreciably stronger (Cox et al. 2001,Nolen-Hoeksema et al. 2008).

Blatt’s (1974, 1991) theory of depressionfocuses on two trait vulnerabilities: self-criticism (an inclination to feelings of guiltand failure stemming from unrealistically highexpectations for oneself ) and dependency (adisposition to feelings of helplessness and fearsof abandonment resulting from a preoccupa-tion with relationships). These constructs aresimilar, although not identical, to Beck’s (1983)constructs of autonomy and sociotropy. Studiesindicate that the link between dependencyand depressive disorders is relatively weak andnonspecific, whereas self-criticism has beenestablished as an important and specific factorin these conditions (Zuroff et al. 2004). Bothtraits have been conceptualized as dynamicpredispositions to depressive disorders, andthere is some support for this view, includingevidence of transactions with life stress as wellas environmental mediation and moderationof personality effects (Zuroff et al. 2004). Self-criticism, and to a lesser extent dependency,have also been found to predict future increasesin depressive symptoms. In addition, there isevidence that dependency predicts the subse-quent onset of major depression in older, butnot younger, individuals (Hirschfeld et al. 1989,Rohde et al. 1990). The concomitants andpathoplasty models have also received empiri-cal support (Zuroff et al. 2004). Finally, there issome research indicating that dependency mayincrease as a function of depressive episodes(consequences model) in youth but not adults(Rohde et al. 1990, 1994; Shea et al. 1996).


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


As noted above, all of these constructs arestrongly linked to N/NE (Cox et al. 2001, Kaschet al. 2001), and some (particularly ruminationand self-criticism) can be considered facets ofthis broader trait (Watson et al. 2006). Lower-order facets can account for variance over andabove that of higher-order traits (Paunonen& Ashton 2001), and several cross-sectionalstudies have supported the incremental validityof ruminative response style and self-criticism(Cox et al. 2004, Muris et al. 2009b) in associa-tions with depressive symptoms. However, thisissue requires more research, particularly usinglongitudinal designs.

CHILD TEMPERAMENTMost of the literature on personality and de-pression has focused on adolescents and adults.Research that is grounded in the child temper-ament literature in developmental psychologyhas the potential to extend existing work onpersonality in depression by (a) providing thestrongest test of the precursor and predisposi-tion models; (b) more precisely delineating thebehavioral manifestations of temperamentalvulnerabilities to mood disorders in youngchildren; (c) tracing the development and conti-nuity of trait vulnerabilities across the lifespan;and (d ) examining the neurobiological, cog-nitive, and interpersonal processes that maymediate the association between early tempera-ment traits and the subsequent development ofdepressive disorder (Compas et al. 2004, Kleinet al. 2008a, Kovacs & Lopez-Duran 2010).

The early childhood temperament dimen-sions that have received the greatest attentionwith respect to depression are N/NE, E/PE,and behavioral inhibition (BI). BI refers to wari-ness, fear, and low exploration in novel situa-tions (Kagan et al. 1987). It combines aspectsof N/NE (fear and anxiety), E/PE (low ap-proach), and conscientiousness (constraint/ ef-fortful control) that do not have a direct analogin most models of adult personality.

Cross-sectional and longitudinal studies ofolder children and adolescents using self-reportmeasures have generally reported associations

BI: behavioralinhibition

of low E/PE and high N/NE with depressionsimilar to those in the adult literature (e.g.,Lonigan et al. 2003).5 Observational studies ofyounger children of depressed mothers also in-dicate that these traits may be associated withrisk for depression (Kovacs & Lopez-Duran2010). For example, in a community sample of100 three-year-olds, Durbin et al. (2005) re-ported that children of mothers with a historyof mood disorder exhibited low PE in emotion-eliciting laboratory tasks. Importantly, this ef-fect was limited to the affective (positive af-fect) and motivational (approach/engagement),rather than the interpersonal (sociability), com-ponents of PE. Furthermore, low PE at age3 predicted depressotypic cognition and mem-ory biases at age 7 (Hayden et al. 2006) andparent-reported depressive symptoms at age 10(Dougherty et al. 2010).

Subsequently, using a larger communitysample (N = 543), Olino et al. (2010) foundthat preschool-aged children of parents with ahistory of depression had higher levels of NEand BI. However, both main effects were qual-ified by interactions with child PE. At highand moderate (but not low) levels of child PE,higher levels of NE and BI were each associ-ated with higher rates of parental depression.Conversely, at low (but not high and moder-ate) levels of child NE, low PE was associatedwith higher rates of parental depression. Takentogether, these results suggest that children ofdepressed parents may exhibit diminished PEor elevated NE and BI. In this latter sample,low PE was also associated with elevated levelsof cortisol shortly after awakening, an index ofhypothalamic-pituitary-adrenal axis dysregula-tion that has been shown to predict MDD inadolescents and adults (Dougherty et al. 2009).

In both the Durbin et al. (2005) and Olinoet al. (2010) studies, the child temperament–parental psychopathology associations were

5Few studies have examined the association of conscientious-ness/effortful control with depression in youth, but analo-gous to the adult literature, there is cross-sectional evidencethat effortful control is negatively correlated with depression(Verstraeten et al. 2009).


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


specific to depression. However, other worksuggests that children of parents with anxietydisorders may also exhibit elevated BI. For ex-ample, Rosenbaum et al. (2000) assessed BI us-ing laboratory measures in 2- to 6-year-old chil-dren of parents with a history of MDD and/orpanic disorder and parents with no history ofmood or anxiety disorders. Children of patientswith both MDD and panic disorder exhibitedsignificantly greater BI than did children of par-ents with no history of mood or anxiety dis-order. Children of parents with panic disor-der alone and children of parents with MDDalone had intermediate levels of BI that didnot differ significantly from children of par-ents in the comorbid and no-psychopathologygroups.

Finally, there is some direct evidence thatpersonality traits assessed in childhood predictthe development of depressive disorders inadults. Caspi et al. (1996) reported that childrenwho were rated as socially reticent, inhibited,and easily upset at age 3 had elevated rates ofdepressive (but not anxiety or substance use)disorders at age 21. Moreover, van Os et al.(1997) found that physicians’ ratings of behav-ioral apathy at ages 6, 7, and 11 were predictiveof both adolescent mood disorder and chronicdepression in middle adulthood. However, BIappears to predict the development of anxietydisorders at least as strongly as depression(Hirshfeld-Becker et al. 2008).

CLINICAL IMPLICATIONS

Prevention

Personality research has important implicationsfor the prevention of depression. Meta-analyticevidence indicates that existing preventiveinterventions can reduce the incidence of de-pressive disorders by 25% (Cuijpers et al. 2008).However, the available strategies are a mix ofuniversal (intervention is administered to theentire population), selective (to a well-definedat-risk group), and indicated (to those withsubthreshold disorder) approaches. Universalinterventions are costly, lack a personalized

focus, and require very large samples to yielddetectable effects, whereas indicated interven-tions may be better described as treatment thanprevention (Munoz et al. 2010). In contrast,selective interventions are true preventivemeasures that are cost effective and can betailored to a specific mechanism of risk. How-ever, implementation of selective strategiesrequires knowledge of risk factors and causalprocesses that lead from the vulnerability to thedisorder.

The majority of established risk factors fordepressive disorders are either immutable (e.g.,demographic characteristics, family history) orpredict onset only in the short term (e.g., stress-ful life events). In contrast, personality is at leastsomewhat malleable, especially in youth, butmay forecast the onset of depression years inadvance, which makes traits a potentially attrac-tive means of identifying individuals at risk andinforming selection of interventions. Differenttrait-disorder pathways would point to differentpreventive strategies; hence, further researchon the nature of personality-depression rela-tions can significantly facilitate developmentof preventive interventions. Another advantageof traits is that they can be assessed relativelyeasily and efficiently and thus are ideal forscreening.

Treatment ResponsePersonality also can inform treatment ofdepressive disorders post onset. In particular,traits can predict response to treatment.Substantial evidence has accumulated thatindividuals with lower N/NE have better treat-ment outcomes across modalities (Kennedyet al. 2005, Mulder 2002, Tang et al. 2009).Other Big Five traits have been studied less andtheir role is not yet certain. However, a recentlarge investigation of a combination interven-tion (medication plus psychotherapy) foundthat low N/NE and high conscientiousnesspredicted who would respond to treatment,and although high E/PE did not contributedirectly, it amplified the effect of high consci-entiousness (Quilty et al. 2008a). As discussed


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


above, investigations of Cloninger’s traits haveproduced inconsistent results ( Joyce et al.2007, Kennedy et al. 2005, Mulder 2002). Fewstudies have examined personality facets, butpreliminary evidence suggests that lower-ordertraits can add substantially to the prediction oftreatment response (Bagby et al. 2008). Amongclinical traits, self-criticism, but not depen-dency, was found to forecast poor treatmentoutcomes (Blatt et al. 1995). Furthermore, per-sonality may be useful in matching patients tointerventions. For instance, Bagby et al. (2008)reported that patients high on N/NE or low onsome agreeableness facets respond better to an-tidepressant medication than to psychotherapy.

The processes underlying these predictiveassociations are not entirely clear. One hypoth-esis is that personality change mediates the ef-fect of treatment on depression. Indeed, thereis a fair amount of evidence that depressiontreatment reduces N/NE and increases E/PE(Zinbarg et al. 2008) and that this effect isnot due to confounding by the depressive state(Tang et al. 2009). Quilty et al. (2008b) tested amediation model and found direct support forthis hypothesis. Other possibilities need to beruled out, however, particularly the hypothesesthat traits predict poorer response because theyindicate a more severe form of depression orthat they interfere with treatment complianceand the therapeutic relationship, thus reducingthe efficacy of the intervention.

CONCLUSIONS AND FUTUREDIRECTIONSThe literature on the relation between person-ality and depression is large, but it has manygaps and inconsistent findings. Nonetheless,it is possible to draw a number of conclu-sions. First, there are moderate-to-large cross-sectional associations between depression andthree general personality traits—N/NE, E/PE,and conscientiousness—as well as with a vari-ety of related traits (e.g., harm avoidance, ru-mination, and self-criticism) and personalitytypes (depressive personality). Second, most ofthe personality traits associated with depression

also are related to other forms of psychopathol-ogy, particularly anxiety disorders. This may re-flect the phenomenon of multifinality, in whichvariables early in the causal chain lead to multi-ple outcomes depending on subsequent eventsin the causal pathway. On the other hand, manyof the disorders that are currently classified asdistinct conditions are closely related; hence,research on personality-psychopathology asso-ciations can provide important information forrevising our nosological system. Third, reportsof some traits (e.g., N/NE and harm avoidance)are influenced by clinical state, whereas othertraits (e.g., E/PE) appear to be independent ofmood state. However, state effects cannot fullyaccount for the associations between person-ality and depression. Fourth, shared etiologi-cal factors (e.g., genes) account for a portion ofthe association between N/NE and depression.Fifth, depressive personality and some traits,particularly N/NE, predict the subsequent on-set of depressive disorders. However, it is un-clear at this point whether they are best con-ceptualized as precursors or predispositions, asit is difficult to tease these models apart, andthere is evidence supporting both accounts.In either case, there is growing evidence thattemperamental risk factors are evident at anearly age, suggesting a promising approach toidentifying young children at risk for depres-sion. Sixth, there is evidence suggesting thatother traits, such as low E/PE and low con-scientiousness/effortful control, may moderatethe relationship between N/NE and depres-sion. Seventh, it appears unlikely that depres-sive episodes produce enduring changes in mostpersonality traits. Finally, personality traits pre-dict, and may in fact influence, the course andtreatment response of depression.

To make further progress in elucidating therelation between personality and mood disor-ders, future studies should be guided by sixbroad considerations. First, most of the litera-ture on personality and depression has focusedon the broad traits of N/NE and E/PE. Thereis a need for further work on conscientiousnessand on lower levels in the trait hierarchy (i.e.,facets). It is important to determine whether a


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


more specific level of analysis will yield morepowerful effects and increase the specificity ofassociations between personality constructs andparticular forms of psychopathology. Clinicaltraits, such as ruminative response style andself-criticism, need to be included in these stud-ies and evaluated jointly with traditional per-sonality dimensions. Finally, it is importantto continue to explore interactions betweentraits.

Second, there is a critical need for prospec-tive, longitudinal studies. Most existing longi-tudinal studies have begun in late adolescenceor adulthood. However, a substantial propor-tion of mood disorders have already developedby mid-adolescence. Therefore, in order to fur-ther test the precursor and predisposition mod-els, and to trace the developmental pathwaysbetween personality and depression, it is neces-sary to conduct longitudinal studies that start asearly as possible in order to obtain a sufficientnumber of first-onset cases and avoid selectionbiases caused by excluding participants who al-ready have a history of mood disorder at initialassessment.

Third, depression researchers have treatedpersonality as static. However, personalitychanges over the course of development.Future work must begin to consider thecomplex personality-environment transactionsthat can strengthen or attenuate personalitytrajectories and predispositions for depressivedisorder. In addition, as understanding ofepigenetics increases, it will be important to

explore epigenetic influences on personalitychange and their relation to depression.

Fourth, if personality is a precursor of, orpredisposes to, the development of depressivedisorders, it is critical to identify the moderat-ing factors and mediating processes involved inthese pathways. There is some evidence sug-gesting that moderators may include gender,early adversity, and life stress, and mediatorsmay include interpersonal deficits, depresso-typic cognitions, maladaptive coping, and be-havioral and neurobiological stress reactivity(Klein et al. 2008a). There is a need for moresystematic research examining these modera-tors and mediators in a longitudinal framework.

Fifth, self-reports have borne the brunt ofmost research in this area and have made impor-tant contributions. However, like all methods,they have limitations and cannot be applied inall contexts (e.g., young children). Thus, thereis a need for further work using complementarymethods such as informant reports and obser-vations in naturalistic and laboratory settings.

Finally, the role of personality/temperamentmay differ for different forms of depressive dis-order. Personality appears to play an especiallyimportant role in early-onset, chronic, and re-current depressive conditions (e.g., Klein 2008,Kotov et al. 2010, van Os et al. 1997). Focusingon broad diagnostic categories such as MDDmay obscure important associations with partic-ular forms of depression; hence, future studiesneed to give greater consideration to the het-erogeneity of depressive disorders.

SUMMARY POINTS

1. There are moderate-to-large cross-sectional associations between depression and threegeneral personality traits—N/NE, E/PE, and conscientiousness—as well as with a varietyof related traits (e.g., harm avoidance, rumination, and self-criticism) and personalitytypes (e.g., depressive personality).

2. Most of the personality traits associated with depression also are related to other formsof psychopathology, particularly anxiety disorders. This may reflect the phenomenonof multifinality, in which variables early in the causal chain lead to multiple outcomesdepending on subsequent events in the causal pathway. On the other hand, many of the


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


disorders that are currently classified as distinct conditions are closely related; hence,research on personality-psychopathology associations can also provide important infor-mation for revising our nosological system.

3. Reports of some traits (e.g., N/NE and harm avoidance) are influenced by clinical state,whereas other traits (e.g., E/PE) appear to be independent of mood state. However, stateeffects cannot fully account for the associations between personality and depression.

4. Shared etiological factors (e.g., genes) account for a portion of the association betweenN/NE and depression.

5. Depressive personality and some traits, particularly N/NE, predict the subsequentonset of depressive disorders. However, it is unclear at this point whether they arebest conceptualized as precursors or predispositions, as it is difficult to tease thesemodels apart, and there is evidence supporting both accounts. In either case, there isgrowing evidence that temperamental risk factors are evident at an early age, suggestinga promising approach to identifying young children at risk for depression.

6. There is evidence suggesting that other traits, such as low E/PE and low conscientious-ness/effortful control, may moderate the relationship between N/NE and depression.

7. It appears unlikely that depressive episodes produce enduring changes in most personalitytraits.

8. Personality traits predict, and may in fact influence, the course and treatment responseof depression.

DISCLOSURE STATEMENTThe authors are not aware of any affiliations, memberships, funding, or financial holdings thatmight be perceived as affecting the objectivity of this review.

LITERATURE CITED

Akiskal HS. 1989. Validating affective personality types. In The Validity of Psychiatric Diagnosis, ed. L Robins,J Barrett, pp 217–27. New York: Raven

Akiskal HS, Akiskal KK, Haykal RF, Manning JS, Connor PD. 2005. TEMPS-A: progress towards validationof a self-rated clinical version of the Temperament Evaluation of the Memphis, Pisa, Paris, and San DiegoAutoquestionnaire. J. Affect. Disord. 85:3–16

Akiskal HS, Hirschfeld RMA, Yerevanian BI. 1983. The relationship of personality to affective disorders: acritical review. Arch. Gen. Psychiatry 40:801–10

Alloy LB, Abramson LY, Walshaw PD, Cogswell A, Grandin LD, et al. 2008. Behavioral approach systemand behavioral inhibition system sensitivities and bipolar spectrum disorders: prospective prediction ofbipolar mood episodes. Bipolar Disord. 10:310–22

Am. Psychiatr. Assoc. 1994. Diagnostic and Statistical Manual of Mental Disorders. Arlington, VA: Am. Psychiatr.Assoc. 4th ed.

Anderson KW, McLean PD. 1997. Conscientiousness in depression: tendencies, predictive utility, and longi-tudinal stability. Cogn. Ther. Res. 21:223–38

Bagby RM, Quilty LC, Segal ZV, McBride CC, Kennedy SH, Costa PT. 2008. Personality and differentialtreatment response in major depression: a randomized controlled trial comparing cognitive-behaviouraltherapy and pharmacotherapy. Can. J. Psychiatry 53:361–70


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


Beck AT. 1983. Cognitive therapy of depression: new approaches. In Treatment of Depression: Old and NewApproaches, ed. P Clayton, J Barrett, pp. 265–90. New York: Raven

Blatt SJ. 1974. Levels of object representation in anaclitic and introjective depression. Psychoanal. Study. Child.29:107–57

Blatt SJ. 1991. A cognitive morphology of psychopathology. J. Nerv. Ment. Dis. 179:449–58Blatt SJ, Quinlan DM, Pilkonis PA, Shea MT. 1995. Impact of perfectionism and need for approval on the brief

treatment of depression: the National Institute of Mental Health Treatment of Depression CollaborativeResearch Program revisited. J. Consult. Clin. Psychol. 63:125–32

Brown TA, Barlow DH. 2009. A proposal for a dimensional classification system based on the shared featuresof the DSM-IV anxiety and mood disorders: implications for assessment and treatment. Psychol. Assess.21:256–71

Canli T. 2008. Toward a neurogenetic theory of neuroticism. Ann. N. Y. Acad. Sci. 1129:153–74Prospective birth-cohort study of relationof temperament inpreschoolers andsubsequentpsychopathology inearly adulthood.

Caspi A, Moffitt TE, Newman DL, Silva PA. 1996. Behavioral observations at age 3 years predictadult psychiatric disorders. Arch. Gen. Psychiatry 53:1033–39

Caspi A, Shiner RL. 2006. Personality development. In Handbook of Child Psychology: Social, Emotional, andPersonality Development, ed. W Damon, R Lerner, N Eisenberg, pp. 300–65. New York: Wiley. 6th ed.

Celikel FC, Kose S, Cumurcu BE, Erkorkmaz U, Sayar K, et al. 2009. Cloninger’s temperament and characterdimensions of personality in patients with major depressive disorder. Compr. Psychiatry 50:556–61

Chmielewski M, Watson D. 2009. What is being assessed and why it matters: the impact of transient error ontrait research. J. Personal. Soc. Psychol. 97:186–202

Clark LA, Watson D. 1999. Temperament: a new paradigm for trait psychology. In Handbook of Personality,ed. LA Pervin, OP John, pp. 399–423. New York: Guilford. 2nd ed.

Reviews theories of, anddata on, relationsbetween personality andthe mood and anxietydisorders.

Clark LA, Watson D, Mineka S. 1994. Temperament, personality, and the mood and anxiety disorders.J. Abnorm. Psychol. 103:103–16

Cloninger CR, Svrakic DM, Przybeck TR. 1993. A psychobiological model of temperament and character.Arch. Gen. Psychiatry 50:975–90

Cloninger CR, Svrakic DM, Przybeck TR. 2006. Can personality predict future depression? A twelve-monthfollow-up of 631 subjects. J. Affect. Disord. 92:35–44

Compas BE, Connor-Smith J, Jaser SS. 2004. Temperament, stress reactivity, and coping: implications fordepression in childhood and adolescence. J. Clin. Child. Adolesc. Psychol. 33:21–31

Cox BJ, Enns MW, Walker JR, Kjernisted K, Pidlubny SR. 2001. Psychological vulnerabilities in patientswith major depression versus panic disorder. Behav. Res. Ther. 39:567–73

Cox BJ, McWilliams LA, Enns MW, Clara IP. 2004. Broad and specific personality dimensions associatedwith major depression in a nationally representative sample. Compr. Psychiatry 45:246–53

Cuijpers P, van Straten A, Smit F, Mihalopoulos C, Beekman A. 2008. Preventing the onset of depressivedisorders: a meta-analytic review of psychological interventions. Am. J. Psychiatry 165:1272–80

de Fruyt F, van de Wiele L, van Heeringen C. 2000. Cloninger’s psychobiological model of temperament andcharacter and the five-factor model of personality. Personal. Individ. Differ. 29:441–52

de Fruyt F, van Leeuwen K, Bagby RM, Rolland JP, Rouillon F. 2006. Assessing and interpreting personalitychange and continuity in patients treated for major depression. Psychol. Assess. 18:71–80

de Graaf R, Bijl RV, Ravelli A, Smit F, Vollenbergh WAM. 2002. Predictors of first incidence of DSM-III-Rpsychiatric disorders in the general population: findings from the Netherlands Mental Health Survey andIncidence Study. Acta Psychiatr. Scand. 106:303–13

Depue RA, Iacono WG. 1989. Neurobehavioral aspects of affective disorders. Annu. Rev. Psychol. 40:457–92Digman JM. 1994. Child personality and temperament: Does the five-factor model embrace both domains?

In The Developing Structure of Temperament and Personality from Infancy to Adulthood, ed. CF Halverson,GA Kohnstamm, RP Martin, pp. 323–38. Hillsdale, NJ: Erlbaum

Dougherty LR, Klein DN, Durbin CE, Hayden EP, Olino TM. 2010. Temperamental positive and negativeemotionality and children’s depressive symptoms: A longitudinal prospective study from age three to ageten. J. Soc. Clin. Psychol. 29:462–88

Dougherty LR, Klein DN, Olino TM, Dyson M, Rose S. 2009. Increased waking salivary cortisol and de-pression risk in preschoolers: the role of maternal history of melancholic depression and early childtemperament. J. Child Psychol. Psychiatry. 50:1495–503


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


Duggan CF, Lee AS, Murray RM. 1990. Does personality predict long-term outcome in depression? Br. J.Psychiatry 157:19–24

Duncan-Jones P, Fergusson DM, Ormel J, Horwood LJ. 1990. A model of stability and change in minorpsychiatric symptoms: results from three longitudinal studies. Psychol. Med. Monogr. Suppl. 18:1–28

Durbin CE, Klein DN, Hayden EP, Buckley ME, Moerk KC. 2005. Temperamental emotionality inpreschoolers and parental mood disorders. J. Abnorm. Psychol. 114:28–37

Fanous AH, Neale MC, Aggen SH, Kendler KS. 2007. A longitudinal study of personality and major depressionin a population-based sample of male twins. Psychol. Med. 37:1163–72

Farmer A, Redman K, Harris T, Mahmood A, Sadler S, et al. 2002. Neuroticism, extraversion, life events anddepression: the Cardiff Depression Study. Br. J. Psychiatry 181:118–22

Farmer A, Mahmood A, Redman K, Harris T, Sadler S, McGuffin P. 2003. A sib-pair study of the temperamentand character scales in major depression. Arch. Gen. Psychiatry 60:490–96

Fraley RC, Roberts BW. 2005. Patterns of continuity: a dynamic model for conceptualizing the stability ofindividual differences in psychological constructs across the life course. Psychol. Rev. 112:60–74

Gershuny BS, Sher KJ. 1998. The relation between personality and anxiety: findings from a 3-year prospectivestudy. J. Abnorm. Psychol. 107:252–62

Goldberg LR. 1993. The structure of phenotypic personality traits. Am. Psychol. 48:26–34Gray JA. 1994. Framework for a taxonomy of psychiatric disorder. In Emotions: Essays on Emotion Theory, ed.

SHM van Goozen, NE Van de Poll, pp. 29–59. Hillsdale, NJ: ErlbaumGriffith JW, Zinbarg RE, Craske MG, Mineka S, Rose RD, et al. 2010. Neuroticism as a common dimension

in the internalizing disorders. Psychol. Med. 40:1125–36Hayden EP, Klein DN, Durbin CE, Olino TM. 2006. Low positive emotionality at age three predicts depres-

sotypic cognitions in seven-year-old children. Dev. Psychopathol. 18:409–23Hecht H, van Calker D, Berger M, von Zerssen D. 1998. Personality in patients with affective disorders and

their relatives. J. Affect. Disord. 51:33–43Hirschfeld RMA, Klerman GL, Clayton PJ, Keller MB. 1983a. Personality and depression: empirical findings.

Arch. Gen. Psychiatry 40:993–98Hirschfeld RMA, Klerman GL, Clayton PJ, Keller MB, Mc Donald-Scott P, Larkin BH. 1983b. Assessing

personality: effects of the depressive state on trait measurement. Am. J. Psychiatry 140:695–99Hirschfeld RMA, Klerman GL, Lavori P, Keller MB, Griffith P, Coryell W. 1989. Premorbid personality

assessments of first onset of major depression. Arch. Gen. Psychiatry 46:345–50Hirshfeld-Becker DR, Micco J, Henin A, Bloomfield A, Biederman J, Rosenbaum J. 2008. Behavioral inhibi-

tion. Depress. Anxiety 25:357–67Huprich SK. 2003. Evaluating facet-level predictions and construct validity of depressive personality disorder.

J. Personal. Disord. 17:219–32Johnson SL, Cueller AK, Ruggero C, Winett-Perlman C, Goonick P, et al. 2008. Life events as predictors of

mania and depression in bipolar I disorder. J. Abnorm. Psychol. 117:268–77Johnson W, Hicks BM, McGue M, Iacono WG. 2007. Most of the girls are alright, but some aren’t: personality

trajectory groups from ages 14 to 24 and some associations with outcomes. J. Personal. Soc. Psychol. 93:266–84

Joiner TE, Lonigan CJ. 2000. Tripartite model of depression and anxiety in youth psychiatric inpatients:relations with diagnostic status and future symptoms. J. Clin. Child Psychol. 29:372–82

Jorm AF, Christensen H, Henderson AS, Jacomb PA, Korten AE, Rodgers B. 2000. Predicting anxiety anddepression from personality: Is there a synergistic effect of neuroticism and extraversion? J. Abnorm.Psychol. 109:145–49

Joyce PR, McKenzie JM, Carter JD, Rae AM, Luty SE, et al. 2007. Temperament, character, and personalitydisorders as predictors of response to interpersonal psychotherapy and cognitive-behavioural therapy fordepression. Br. J. Psychiatry 190:503–8

Kagan J, Reznick JS, Snidman N. 1987. The physiology and psychology of behavioral inhibition in children.Child Dev. 55:1459–73

Kandler C, Bleidorn W, Riemann R, Spinath FM, Thiel W, Angleitner A. 2010. Sources of cumulativecontinuity in personality: a longitudinal multiple-rater twin study. J. Personal. Soc. Psychol. 98:995–1008


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


Kasch KL, Klein DN, Lara ME. 2001. A construct validation study of the Response Styles QuestionnaireRumination Scale in participants with a recent-onset major depressive episode. Psychol. Assess. 13:375–83

Kasch KL, Rottenberg J, Arnow BA, Gotlib IH. 2002. Behavioral activation and inhibition systems and theseverity and course of depression. J. Abnorm. Psychol. 111:589–97

Kendler KS, Kuhn J, Prescott CA. 2004. The interrelationship of neuroticism, sex, and stressful life events inthe prediction of episodes of major depression. Am. J. Psychiatry 161:631–36

Kendler KS, Neale MC. 2010. Endophenotype: a conceptual analysis. Mol. Psychiatry 15:789–97Kendler KS, Neale MC, Kessler RC, Heath AC, Eaves LJ. 1993. A longitudinal twin study of personality and

major depression in women. Arch. Gen. Psychiatry 50:853–62Tests models ofpersonality and MDD ina prospective study of alarge twin sample.

Kendler KS, Gatz M, Gardner CO, Pederson NL. 2006. Personality and major depression. Arch. Gen.Psychiatry 63:1113–20

Kennedy SH, Farvolden P, Cohen NL, Bagby RM, Costa PT. 2005. The impact of personality on thepharmacological treatment of depression. In Depression and Personality, ed. M Rosenbluth, SH Kennedy,RM Bagby, pp. 97–119. Arlington, VA: Am. Psychiatr. Publ.

Kercher AJ, Rapee RM, Schniering CA. 2009. Neuroticism, life events and negative thoughts in the develop-ment of depression in adolescent girls. J. Abnorm. Child Psychol. 37:903–15

Klein DN. 1990. Depressive personality: reliability, validity, and relation to dysthymia. J. Abnorm. Psychol.99:412–21

Klein DN. 1999. Depressive personality in the relatives of outpatients with dysthymic disorder and episodicmajor depressive disorder and normal controls. J. Affect. Disord. 55:19–27

Klein DN. 2008. Dysthymia and chronic depression. In Psychopathology: History, Theory, and Diagnosis, ed. WECraighead, DJ Miklowitz, LW Craighead, pp. 329–65. Hoboken, NJ: Wiley

Klein DN, Dougherty LR, Laptook RS, Olino TM. 2008a. Temperament and risk for mood disorders inadolescents. In Adolescent Emotional Development and the Emergence of Depressive Disorders, ed. N Allen,L Sheeber, pp. 238–61. Cambridge, UK: Cambridge Univ. Press

Klein DN, Miller GA. 1993. Depressive personality in nonclinical subjects. Am. J. Psychiatry 150:1718–24Klein DN, Torpey DC, Bufferd SJ, Dyson MW. 2008b. Depressive disorders. In Child and Adolescent Psy-

chopathology, ed. TP Beauchaine, SP Hinshaw, pp. 477–509. Hoboken, NJ: WileyKlein MH, Wonderlich S, Shea MT. 1993. Models of relationships between personality and depression:

toward a framework for theory and research. In Personality and Depression: A Current View, ed. MH Klein,DJ Kupfer, TM Shea, pp. 1–54. New York: Guilford

Documents personalityprofiles of commonmental disordersthrough a systematicmeta-analysis.

Kotov R, Gamez W, Schmidt FL, Watson D. 2010. Linking “Big” personality traits to anxiety, de-pressive, and substance use disorders: a meta-analysis. Psychol. Bull. 136:768–821

Kotov R, Watson D, Robles J, Schmidt NB. 2007. Personality traits and anxiety symptoms: the MultilevelTrait Predictor Model. Behav. Res. Ther. 45:1485–503

Kovacs M, Lopez-Duran N. 2010. Prodromal symptoms and atypical affectivity as predictors of major depres-sion in juveniles: implications for prevention. J. Child. Psychol. Psychiatry 51:472–96

Kraepelin E. 1921. Fundamental states. In Manic-Depressive Insanity and Paranoia, ed. GM Robertson, pp. 117–32. Edinburgh, Scotland: Livingstone

Krueger RF, Johnson W. 2008. Behavioral genetics and personality: a new look at the integration of nature andnurture. In Handbook of Personality: Theory and Research, ed. OP John, RW Robins, LA Pervin, pp. 287–310.New York: Guilford. 3rd ed.

Krueger RF, Tackett JL. 2003. Personality and psychopathology: working toward the bigger picture. J. Per-sonal. Disord. 17:109–28

Shows that depressivepersonality predicts thelater first-lifetime onsetof dysthymic disorder.

Kwon JS, Young-Me K, Chang-Gok C, Byung-Joo P, Kim L. 2000. Three-year follow-up of womenwith the sole diagnosis of depressive personality disorder: subsequent development of dysthymiaand major depression. Am. J. Psychiatry 157:1966–72

Lahey BB. 2004. Commentary: role of temperament in developmental models of psychopathology. J. Clin.Child. Adolesc. Psychol. 33:88–93

Lahey BB. 2009. Public health significance of neuroticism. Am. Psychol. 64:241–56Laptook RS, Klein DN, Dougherty LR. 2006. Ten-year stability of depressive personality disorder in depressed

outpatients. Am. J. Psychiatry 163:865–71


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


Lonigan CJ, Phillips BM, Hooe ES. 2003. Relations of positive and negative affectivity to anxiety and depres-sion in children: evidence from a latent-variable longitudinal study. J. Consult. Clin. Psychol. 71:465–81

Markon KE, Krueger RF, Watson D. 2005. Delineating the structure of normal and abnormal personality:an integrative hierarchical approach. J. Personal. Soc. Psychol. 88:139–57

McFarland BR, Shankman SA, Tenke CE, Bruder GE, Klein DN. 2006. Behavioral activation system deficitspredict the six-month course of depression. J. Affect. Disord. 91:229–34

Middledorp CM, Cath DC, Beem AL, Willemsen G, Boomsma DI. 2008. Life events, anxious depression andpersonality: a prospective and genetic study. Psychol. Med. 38:1557–65

Morey LC, Shea MT, Markowitz JC, Stout RL, Hopwood CJ, et al. 2010. State effects of major depressionon the assessment of personality and personality disorder. Am. J. Psychiatry 167:528–35

Morris BH, Bylsma LM, Rottenberg J. 2009. Does emotion predict the course of major depressive disorder?A review of prospective studies. Br. J. Clin. Psychol. 48:255–73

Mulder RT. 2002. Personality pathology and treatment outcome in major depression: a review. Am. J. Psy-chiatry 159:359–71

Munoz RF, Cuijpers P, Smit F, Barrera AZ, Leykin Y. 2010. Prevention of major depression. Annu. Rev. Clin.Psychol. 6:181–212

Muris P, Bos AER, Mayer B, Verkade R, Thewissen V, Dell’Avvento V. 2009a. Relations among behavioralinhibition, Big Five personality factors, and anxiety disorder symptoms in nonclinical children. Personal.Individ. Differ. 46:525–29

Muris P, Fokke M, Kwik D. 2009b. The ruminative response style in adolescents: an examination of its specificlink to symptoms of depression. Cogn. Ther. Res. 33:21–32

Naragon-Gainey K, Watson D, Markon K. 2009. Differential relations of depression and social anxiety symp-toms to the facets of extraversion/positive emotionality. J. Abnorm. Psychol. 118:299–310

Neyer FJ, Lehnart J. 2007. Relationships matter in personality development: evidence from an 8-year longi-tudinal study across young adulthood. J. Personal. 75:535–68

Nolen-Hoeksema S. 1991. Responses to depression and their effects on the duration of depressive episodes.J. Abnorm. Psychol. 100:569–82

Nolen-Hoeksema S. 2000. The role of rumination in depressive disorders and mixed anxiety/depressive symp-toms. J. Abnorm. Psychol. 109:504–11

Nolen-Hoeksema S, Wisco BE, Lyubomirsky S. 2008. Rethinking rumination. Perspect. Psychol. Sci. 3:400–24 Examines relation ofparental depressionwith observations oftemperament in largesample of preschool-aged offspring.

Olino TM, Klein DN, Dyson MW, Rose SA, Durbin CE. 2010. Temperamental emotionality inpreschool-aged children and depressive disorders in parents: associations in a large communitysample. J. Abnorm. Psychol. 119:468–78

Ongur D, Farabaugh A, Iosifescu DV, Perils R, Fava M. 2005. Tridimensional Personality Questionnaire fac-tors in major depressive disorder: relationship to anxiety disorder comorbidity and age of onset. Psychother.Psychosom. 74:173–78

Ormel J, de Jong A. 1999. On vulnerability to common mental disorders. An evidence-based plea for adevelopmental perspective. In Common Mental Disorders in Primary Care: Essays in Honour of Professor SirDavid Goldberg, ed. M Tansella, G Thornicroft, pp. 34–51. London: Routledge

Prospectively tested adynamic predispositionmodel of neuroticism-depression relations in asample of older adults.

Ormel J, Oldehinkel AJ, Brilman EI. 2001. The interplay and ecological continuity of neuroticism, dif-ficulties, and life events in the etiology of major and subsyndromal, first and recurrent depressiveepisodes in later life. Am. J. Psychiatry 158:885–91

Prospectively testedcompeting models ofpersonality-depressionrelations in a largecommunity sample.

Ormel J, Oldehinkel AJ, Vollebergh W. 2004a. Vulnerability before, during, and after a major de-pressive episode. Arch. Gen. Psychiatry 61:990–96

Ormel J, Rosmalen J, Farmer A. 2004b. Neuroticism: a noninformative marker of vulnerability to psy-chopathology. Soc. Psychiatry Psychiatr. Epidemiol. 39:906–12

Ørstavik RE, Kendler KS, Czajkowski N, Tambs K, Reichborn-Kjennerud T. 2007. The relationship betweendepressive personality disorder and major depressive disorder: a population-based twin study. Am. J.Psychiatry 164:1866–72

Ouimette PC, Klein DN, Pepper CM. 1996. Personality traits in the first degree relatives of outpatients withdepressive disorders. J. Affect. Disord. 39:43–53

Paunonen SV, Ashton MD. 2001. Big Five factors and facets and the prediction of behavior. J. Personal. Soc.Psychol. 81:524–39


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


Pinto-Meza A, Caseras X, Soler J, Puigdemont D, Perez V, Torrubia R. 2006. Behavioral inhibition andbehavioral activation systems in current and recovered major depression participants. Personal. Individ.Differ. 40:215–26

Quilty LC, De Fruyt F, Rolland JP, Kennedy SH, Rouillon PF, Bagby RM. 2008a. Dimensional personalitytraits and treatment outcome in patients with major depressive disorder. J. Affect. Disord. 108:241–50

Quilty LC, Meusel LAC, Bagby RM. 2008b. Neuroticism as a mediator of treatment response to SSRIs inmajor depressive disorder. J. Affect. Disord. 111:67–73

Reich J, Noyes R, Hirschfeld R, Coryell W, O’Gorman T. 1987. State and personality in depressed and panicpatients. Am. J. Psychiatry 144:181–87

Roberts BW, Bogg T. 2004. A 30-year longitudinal study of the relationships between conscientiousness-related traits, and the family structure and health-behavior factors that affect health. J. Personal. 72:325–54

Roberts BW, Caspi A, Moffitt T. 2003. Work experiences and personality development in young adulthood.J. Personal. Soc. Psychol. 84:582–93

Roberts BW, DelVecchio WF. 2000. The rank-order consistency of personality from childhood to old age: aquantitative review of longitudinal studies. Psychol. Bull. 126:3–25

Roberts BW, Kuncel NR, Shiner R, Caspi A, Goldberg LR. 2007. The power of personality: the comparativevalidity of personality traits, socioeconomic status, and cognitive ability for predicting important lifeoutcomes. Perspect. Psychol. Sci. 2:313–45

Roberts BW, Walton KE, Viechtbauer W. 2006. Patterns of mean-level change in personality traits acrossthe life course: a meta-analysis of longitudinal studies. Psychol. Bull. 132:1–25

Rohde P, Lewinsohn PM, Seeley JR. 1990. Are people changed by the experience of having an episode ofdepression? A further test of the scar hypothesis. J. Abnorm. Psychol. 99:264–71

Rohde P, Lewinsohn PM, Seeley JR. 1994. Are adolescents changed by an episode of major depression?J. Am. Acad. Child Adolesc. Psychiatry 33:1289–98

Rood L, Roelofs J, Bogels SM, Nolen-Hoeksema S, Schouten E. 2009. The influence of emotion-focusedrumination and distraction on depressive symptoms in non-clinical youth: a meta-analytic review. Clin.Psychol. Rev. 29:607–16

Rorsman B, Grasbeck A, Hagnell O, Isberg PE, Otterbeck L. 1993. Premorbid personality traits and psycho-metric background factors in depression: the Lundby Study 1957–1972. Neuropsychobiology 27:72–79

Rosenbaum JF, Biederman J, Hirshfeld-Becker DR, Kagan J, Snidman N, et al. 2000. A controlled studyof behavioral inhibition in children of parents with panic disorder and depression. Am. J. Psychiatry157:2002–10

Rothbart MK, Bates JE. 2006. Temperament. In Handbook of Child Psychology: Vol. 3. Social, Emotional, andPersonality Development, ed. N Eisenberg, W Damon, RM Lerner, pp. 99–166. Hoboken, NJ: Wiley. 6thed.

Rudolph KD, Klein DN. 2009. Exploring depressive personality traits in youth: origins, correlates, and de-velopmental consequences. Dev. Psychopathol. 21:1155–80

Ryder AG, Quilty LC, Vachon DD, Bagby RM. 2010. Depressive personality and treatment outcome in majordepressive disorder. J. Personal. Disord. 24:392–404

Schneider K. 1958. Psychopathic Personalities. London: CassellScollon CN, Diener E. 2006. Love, work, and changes in extraversion and neuroticism over time. J. Personal.

Soc. Psychol. 91:1152–65Shea MT, Leon AC, Mueller TI, Solomon DA, Warshaw MG, Keller MB. 1996. Does major depression

result in lasting personality change? Am. J. Psychiatry 153:1404–10Smith DJ, Duffy L, Stewart ME, Muir WJ, Blackwood DHR. 2005. High harm avoidance and low self-

directedness in euthymic young adults with recurrent, early-onset depression. J. Affect. Disord. 87:83–89Smits DJM, Boeck PD. 2006. From BIS/BAS to the Big Five. Eur. J. Personal. 20:255–70

Finds thatantidepressantsinfluenced personalityindependent of theireffects on depression ina placebo-controlledtrial.

Tang TZ, DeRubeis RJ, Hollon SD, Amsterdam J, Shelton R, Schalet B. 2009. Personality changeduring depression treatment: a placebo-controlled trial. Arch. Gen. Psychiatry 66:1322–30

Vachon DD, Sellbom M, Ryder AG, Miller JD, Bagby RM. 2009. A five-factor model description of depressivepersonality disorder. J. Personal. Disord. 23:447–65

Van den Akker A, Dekovic M, Prinzie P. 2010. Transitioning to adolescence: how changes in child personalityand overreactive parenting predict adolescent adjustment problems. Dev. Psychopathol. 22:151–63


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.


van Os J, Jones PB. 1999. Early risk factors and adult person–environment relationships in affective disorder.Psychol. Med. 29:1055–67

van Os J, Jones P, Lewis G, Wadsworth M, Murray R. 1997. Developmental precursors of affective illness ina general population birth cohort. Arch. Gen. Psychiatry 54:625–31

Verstraeten K, Vasey MW, Raes F, Bijttebier P. 2009. Temperament and risk for depressive symptoms inadolescence: mediation by rumination and moderation by effortful control. J. Abnorm. Child Psychol.37:349–61

Watson D. 2000. Mood and Temperament. New York: GuilfordWatson D. 2009. Differentiating the mood and anxiety disorders: a quadripartite model. Annu. Rev. Clin.

Psychol. 5:221–47Watson D, Kotov R, Gamez W. 2006. Basic dimensions of temperament in relation to personality and

psychopathology. In Personality and Psychopathology, ed. R Krueger, J Tackett, pp. 7–38. New York:Guilford

Examines mediators andmoderators of relationbetween personality anddepressive symptoms inadolescents.

Wetter EK, Hankin BL. 2009. Mediational pathways through which positive and negative emotionalitycontribute to anhedonic symptoms of depression: a prospective study of adolescents. J. Abnorm.Child Psychol. 37:507–20

Wichers MC, Myin-Germeys I, Jacobs N, Peeters F, Kenis G, et al. 2007. Evidence that moment-to-momentvariation in positive emotions buffer genetic risk for depression: a momentary assessment twin study.Acta Psychiatr. Scand. 115:451–57

Zinbarg RE, Uliaszek AA, Adler JM. 2008. The role of personality in psychotherapy for anxiety and depression.J. Personal. 76:1649–88

Zuroff DC, Mongrain M, Santor DA. 2004. Conceptualizing and measuring personality vulnerability todepression: comment on Coyne and Whiffen (1995). Psychol. Bull. 130:489–511


Ann

u. R

ev. C

lin. P

sych

ol. 2

011.

7:26

9-29

5. D

ownl

oade

d fro

m w

ww

.ann

ualre

view

s.org

by S

tate

Uni

vers

ity o

f New

Yor

k - S

tony

Bro

ok o

n 03

/31/

11. F

or p

erso

nal u

se o

nly.