Examining an obsessive-compulsive core dimensions model: Structural validity of harm avoidance and incompleteness

https://doi.org/10.1016/j.jocrd.2014.01.003Get rights and content

Highlights

  • A model of OCD posits harm avoidance and incompleteness as core motivational dimensions.

  • Our four studies examine the model׳s structural validity in clinical and nonclinical samples.

  • We found support across the method of assessment, level of generality, and population.

  • Results provide support for the core dimensions model.

Abstract

Building upon work by Rasmussen and Eisen, our group has proposed a model comprising two core motivational dimensions underlying obsessive-compulsive symptoms: harm avoidance and incompleteness. The model has received increasing attention; however the structural soundness and divergence of its factors are yet to be investigated fully, either as symptom-specific motivations for clinical OCD symptoms or as stylistic traits in the nonclinical population. This paper presents four studies designed to investigate the structural validity of harm avoidance and incompleteness in clinical and nonclinical samples. Results yielded support across the method of assessment (interview, questionnaire), level of generality (symptom-specific state, trait), and population (clinical, nonclinical). Evidence was also found of the model׳s method invariance, with both factors strongly self-associated across method forms when ascertained as symptom-specific motivations. The results provide support for key assertions of the core dimensions model and also point to the utility of the interviewer-rated and questionnaire measures developed during this work: the Obsessive-Compulsive Core Dimensions Interview (OC-CDI) and Core Dimensions Questionnaire (OC-CDQ). Clinical and theoretical implications and challenges for future research are discussed.

Introduction

The search for causal factors in obsessive-compulsive disorder (OCD) hinges upon determination of its phenotype, a task yet to be conclusively resolved (Rasmussen, Eisen, & Greenberg, 2013). The hunt for “the” OCD phenotype has instead converged on the likelihood that OCD is a heterogeneous condition, with different manifestations, potentially having different etiologies. The evidence for this was sufficiently compelling to prompt the OCD working group for the newest edition of the Diagnostic and Statistical Manual (DSM-5; American Psychiatric Association [APA], 2013) to give serious consideration to including heterogeneity (of symptoms) into its recommendations for changes to the diagnostic criteria for OCD (see Leckman et al., 2010, Phillips, 2009). In the end, in DSM-5 OCD is both the same and different in ways that will likely make it a watershed for research. OCD, minus hoarding, remains a unitary diagnostic entity, but is no longer considered an anxiety disorder and instead has been reclassified into the new category of obsessive-compulsive and related disorders. The issue of heterogeneity remains unresolved but is acknowledged by references to different symptom themes and dimensions throughout the OCD section.

The idea that overt symptoms are a meaningful basis for understanding heterogeneity in OCD has been the focus of considerable research attention over the past decade or so. Numerous studies have used data-reduction statistical methods like factor analysis to examine the structural characteristics of OCD symptoms, and have generally found variations of between three and five factors, comprising symmetry and ordering, contamination and cleaning, obsessions and checking (potentially further divisible into harm-related obsessions and checking, and other obsessions), and hoarding. Alongside this heightened research activity, however, has been growing awareness of the many conceptual and methodological challenges and limitations of this approach to classifying OCD (see Clark, 2005, McKay et al., 2004, Radomsky and Taylor, 2005 for reviews). Examples include findings of substantial overlap among ostensibly discrete symptom “subtypes” (e.g., Baer, 1994, Summerfeldt et al., 1999), variable and often sub-threshold loadings of symptoms within their predicted factors (e.g., Pinto et al., 2008, Summerfeldt et al., 1999), and joint predictions of the same symptoms by different symptom factors (e.g., Summerfeldt et al., 2004, Storch et al., 2007, Storch et al., 2008), all of which converge to suggest that OCD symptoms have commonalities and differences that cannot be understood with overt symptom data alone.

Indeed, a fundamental limitation of symptom-based classification is that its exclusive focus on the topographic aspects of OCD ignores potentially more meaningful underlying features. This is because it is based upon “what the individual does, rather than why the individual does it” (Summerfeldt et al., 2004, p. 1464). Dissimilar symptoms, for example, may have markedly similar motivational underpinnings (e.g., checking, like washing, may serve to alleviate anxious apprehension and allay fears about potential harm). In both cases, the harm avoidant, preventative, or neutralizing form of the two profiles is similar, though the content differs. Conversely, behaviors that on the surface appear similar can be characterized by qualitatively different subjective experiences. Someone may clean to eliminate germs and prevent harm or, alternatively, to preserve the pristine “just right” state of belongings and so regain a sense of satisfaction or inner completeness, with a little sense of threat (Tallis, 1996), although symptom-based classification strategies would collapse both into the “cleaning” category.

These empirical and conceptual points converge to suggest that underlying the many overt symptom expressions in OCD may be broader core dimensions. A categorical framework, such as the DSM, conceptualizes disorders as discrete nonoverlapping entities. In contrast, a dimensional perspective focuses on varying expressions of functional processes that cross diagnostic boundaries and manifest symptom types, each representing a continuum ranging from normal adaptive behavior to pathology (see Clark, 2005, Widiger, 1992, Widiger and Clark, 2000, Widiger and Samuel, 2005). Dimensional models were under unprecedented intense debate during the recent revision of the DSM (Kraemer, 2007; see Krueger et al., 2005, Regier, 2007). The potential supplemental value of transdiagnostic dimensions or “underlying vulnerabilities,” particularly for explaining heterogeneity within and commonalities across ostensibly discrete diagnoses, is explicitly acknowledged in the preface to the DSM-5 (APA, 2013, p. xli), and was the impetus behind the new organizational structure of diagnostic chapters and codes. In the case of OCD this resulted in a compromise: it finds its new home with functionally similar conditions like trichotillomania and body dysmorphic disorder, but this is placed adjacent to the Anxiety Disorders chapter, in acknowledgment of evidence showing that OCD has elements of, or “close relationships with,” both (APA, 2013, p. 235).

Many indicators point to the particular suitability of OCD to a dimensional perspective, including its remarkable heterogeneity (see McKay et al., 2004, Taylor, 2005), its proneness to diagnostic “boundary disputes” (Widiger & Samuel, 2005, p. 494), that is, overlap between its features and those of other disorders (Phillips et al., 2003, Phillips et al., 2010), evidence that for the most part neither its symptoms nor its putatively central cognitive features are taxonomic (Haslam et al., 2005, Olatunji et al., 2008), a tradition of analogue research showing continuity of its features and correlates into the nonclinical population (Gibbs, 1996), and its high rates and variability of co-occurrence with other symptoms and disorders (see Phillips et al., 2010, Stein et al., 2010). What underlying vulnerabilities, then, might plausibly cut across overt symptoms in OCD, across its clinical and nonclinical expressions, and perhaps across its boundaries with other disorders?

Two distinct affective-motivational themes can be readily identified in OCD. One, similar to that observed in the anxiety disorders, is characterized by the primary role of anxious apprehension, sensitivity to potential threat, and excessive avoidance of perceived harm. The idea that such features jointly express a common underlying vulnerability to anxiety disorders is conceptually well developed and supported (e.g., Barlow, 2000, Brown and Barlow, 1992, Cloninger, 1986, Zinbarg and Barlow, 1996). This aspect of OCD has most influenced up-to-the-present North American views on its diagnosis and classification (see Stein et al., 2010) and psychological formulation and treatment (e.g., Carr, 1974, McFall and Wollersheim, 1979, Rachman and Hodgson, 1980, Foa et al., 1985, Rachman, 1997, Salkovskis, 1985). Many expressions of OCD, however, do not fit this anxious-avoidant profile. Though “distress” (i.e., negative affect) prevention or reduction without specifiable threat has been recognized in the diagnostic criteria for compulsions since DSM-IV (APA, 2000), phenomenological accounts suggest a more specific affective-motivational picture. As described in detail elsewhere (Summerfeldt, 2004), the affective component is characterized by a tormenting sense of dissatisfaction or discomfort with one׳s current state. Motivationally what dominates is the drive to quell profound feelings of imperfection – “not just right[ness]” (Coles et al., 2003, Leckman et al., 1994) – connected with the sense that actions, intentions and perceptions have been incompletely achieved. It is often experienced as a sensory-affective disturbance (see da Silva Prado et al., 2008). This aspect of OCD was documented over a century ago by Janet (1903), who labeled it (in translation) as incompleteness (Pitman, 1987b), and is newly acknowledged in the DSM-5 as one of the “affective responses” seen in OCD (APA, 2013, p. 239). Incompleteness is not uncommon. In studies with clinical samples using different operationalizations, Ferrão et al. (2012), Leckman et al., 1994, Leckman et al., 1994/1995, Miguel et al. (2000), and Storch et al. (2010) all found that well over half of the participants endorsed that it was associated with their need to perform compulsions. Incompleteness seems to represent the extreme end of a continuum of obsessive-compulsive, or pathologically perfectionistic, personality traits (see Coles et al., 2008, Ecker et al., 2013) and have parallels in several obsessive-compulsive “spectrum” conditions, such as tic disorders (Leckman et al., 1994, Miguel et al., 2000) and body dysmorphic disorder (Veale et al., 1996).

Direct and indirect empirical support has accumulated from multiple sources not only for the value of differentiating overt obsessive-compulsive symptoms from their underlying motivations, but also for the possibility that such underlying motivations may take harm avoidance and incompleteness forms. Examples include clinical accounts and case studies (e.g., Rasmussen and Eisen, 1988, Rasmussen and Eisen, 1990, Rasmussen and Eisen, 1992, Summerfeldt, 2004, Summerfeldt, 2006, Summerfeldt, 2008, Tallis, 1996), and studies with nonclinical samples using both correlational (Pietrefesa and Coles, 2008, Ghisi et al., 2010) and lab-based behavioral analog designs (e.g., Cougle et al., 2013, Pietrefesa and Coles, 2009). In one of the few quantitative studies with clinical participants, Ecker and Gönner (2008) analyzed data from a large sample (n=202) and multiple symptom measures to test hypotheses regarding whether harm avoidance and incompleteness underlie and cross over overt symptoms. In line with their general predictions, some symptoms (symmetry, harm-related and repugnant obsessions) were associated with only one dimension, whereas others (i.e., checking) were motivationally diverse.

These two motivational themes were first incorporated into a classificatory model of OCD by Rasmussen and Eisen, 1988, Rasmussen and Eisen, 1990, Rasmussen and Eisen, 1992, who on the basis of theoretically-informed clinical observation posited that OCD may be subtyped according to three “core features”: abnormal risk assessment, pathologic doubt, and incompleteness, which cut across overt symptom manifestations and are associated with distinct features, comorbidities, vulnerabilities, and causal factors. Despite its heuristic appeal, this model has weaknesses. Foremost is that it is implicitly categorical; although the core features are ostensibly dimensional, paramount emphasis is placed upon the distinct subtypes that derive from them. This raises conceptual and methodological drawbacks true of categorical models in general (see Widiger, 1992). At what point, for example, does an individual stop belonging to one subtype and enter into the domain of another? As yet, there is no known pathophysiology for the subtypes, and the distinction is made on the basis of behavioral, motivational and emotional features. These are continuous variables with gradations of severity, however, and could plausibly appear in “blends” in the same individual. The core features would be better depicted as discrete continuous dimensions. Individual profiles might then be predicted from their interaction, or the predominance of one, relative to the other. Relatedly, there is little basis for the distinctness of pathological doubt. This has been seen as the essential characteristic of OCD in general (Beech, 1974, Reed, 1985), and might be more parsimoniously represented as the intersection of the other two dimensions – a possibility once alluded to by Rasmussen and Eisen (1990). In an effort to address these points and make the model more amenable to quantitative research, our group proposed a revision to Rasmussen and Eisen׳s model, comprising harm avoidance and incompleteness as its continuous core dimensions. We see the two as orthogonal; that is, an individual׳s “placement” on one dimension has no direct causal bearing on placement on the other, and all combinations of levels are possible in the general population. In OCD, harm avoidance and incompleteness can be conceptualized as axes defining a dimensional space wherein particular symptom configurations are most likely to develop. In a case study in Summerfeldt (2004), for example, the combination of high incompleteness with normatively “typical” harm avoidance was associated with obsessions involving symmetry and the need to know or remember details, mental rituals, and re-reading, repeating, and ordering compulsions.

The model rests on the idea that core motivations in OCD are content-independent, and exist at a different level of analysis than overt behaviors or symptoms. This presents unique challenges for its quantification because measures of OCD symptoms, following longstanding psychometric conventions (see Allport, 1937, Cattell, 1957), rest on the premise that a single behavior has a single meaning. Examples can be seen in existing attempts to capture incompleteness-related OCD symptoms. The most cited measure, Coles and colleagues׳ well-validated Not Just Right Experiences (NJRE) Questionnaire (Coles et al., 2003), uses behavioral anchors (e.g. getting dressed, locking the door); however it is not intended as a self-standing classificatory measure as it does not assess a full range of symptoms or other potential motivations. At present, two well-recognized multidimensional measures of OCD symptoms include items tapping the Incompleteness construct: the Vancouver Obsessive Compulsive Inventory (VOCI; Thordarson et al., 2004) in its “Just Right” subscale, and the symptom checklist of the revised Yale-Brown Obsessive-Compulsive Scale (Y-BOCS II; Goodman, Rasmussen, Price, & Storch, 2006), in single obsession and compulsion items. In both instruments, however, it is treated as a distinct symptom type of OCD, which presupposes that it can exist “in itself” rather than as an affective-motivational context out of which diverse manifest symptoms may emerge. The compulsion item on the Y-BOCS II symptom checklist, for example – “doing something until ‘just right’ – is imbedded alongside manifest symptoms like compulsive counting and ordering. In the case of the VOCI, as Ecker and Gönner (2008) have noted, the subscale contains items from multiple levels of evaluation. Most reference specific symptom content plausibly linked with incompleteness (some paralleling other VOCI subscales, e.g., “I am strongly compelled to count things”), but a few tap both content and form (e.g., “I often have trouble getting things done because I try to do everything exactly right”). It is difficult to reliably assess harm avoidance and incompleteness without confounding act with underlying motivation.

Despite these obstacles, there has been considerable interest in the core dimensions model. It is being increasingly cited in the literature, often in studies also utilizing the Obsessive-Compulsive Core Dimensions Questionnaire (OC-CDQ), an unpublished measure we developed to operationalize it (e.g., Coles et al., 2005, Cougle et al., 2013, Ecker and Gönner, 2008, Ecker et al., 2013, Pietrefesa and Coles, 2008, Pietrefesa and Coles, 2009, Sarig et al., 2012). Before advancing any substantive inferences about the explanatory value of this model, however, it is important to establish its structural validity. The model posits two distinct dimensions – harm avoidance and incompleteness – each characterized by a coherent set of subjective experiences. These are structural properties, and require empirical validation. This step has been somewhat bypassed. Our group׳s cited empirical research examining the structural validity of the model-closely linked to the development of the OC-CDQ – largely appears in conference proceedings (e.g., Summerfeldt, Kloosterman, Parker, Antony, & Swinson, 2001) so is not widely accessible. In the one existing published study, Pietrefesa and Coles (2008) investigated the “separability” of harm avoidance and incompleteness, operationalized as stylistic (i.e., non context-specific) traits, in a college student sample, using the OC-CDQ. Though a very high correlation was found between the two factors (r=.76) confirmatory factor analyses found support for a two-factor model.

The research by Pietrefesa and Coles (2008), though informative, does not touch on some crucial foundational issues. Most importantly, the core dimensions model and its predecessor by Rasmussen and Eisen were developed to account for the motivations behind obsessive-compulsive symptoms, so clearly the coherence and divergence of the two factors as they relate specifically to symptoms, that is, as a state, need to be established, and in the clinical population. The empirical grounds for extrapolating these dimensions into assumedly cross-situationally stable traits, assessable via self-report questionnaire, as reported in several often-cited studies using the OC-CDQ including Pietrefesa and Coles, 2008, Pietrefesa and Coles, 2009, and Ecker and Gönner (2008), also need to be established.

The present study seeks to fill this gap so that research on the core dimensions model can proceed on firmer empirical footings. In it, we describe our efforts to operationalize the model with careful attention to its structural validity, with data gathered using different methods (semi-idiographic interview and standardized self-report questionnaire), in clinical OCD and nonclinical samples, and with its two dimensions ascertained as states (i.e., symptom-specific) and traits. Dimensional models are based upon the idea of an unbroken continuum between healthy diversity and predisposition to illness, with disorder seen as an aberration of otherwise adaptive functional processes. It is not surprising then that what evolved was a search for measurement variables sensitive enough to harm avoidance and incompleteness that they could be identifiable irrespective of the population, the method of assessment, or the level of generality (state vs. trait) at which they were assessed.

The research comprised four incremental studies, and resulted in the final version of the OC-CDQ. Studies 1, 2 and 3 involved model specification and testing using semi-idiographic interview then self-report questionnaire data, with the two dimensions ascertained first as states (symptom specific) in clinical samples then as traits in a nonclinical sample. Study 4 involved model testing and replication with self-report questionnaire data from independent nonclinical (trait) and clinical (state) samples. As a final analysis, we examined the evidence for method invariance of the model using data from a subset of the clinical sample that had completed both interview and self-report state measures.

Section snippets

Clinical sample

Clinical participants were 289 individuals with OCD who had been sequentially evaluated and diagnosed at a Canadian hospital-based outpatient anxiety clinic. Exclusion criteria were age below 16 or current psychotic disorder. There were 104 men and 185 women, ranging in age from 16 to 65 (Mean=34.41±11.26). Half were either married (41.6%) or cohabitating (7.7%), 45.3% were single, and 5.5% were separated or divorced. The majority were well educated (19.9% had some postsecondary education,

Measures

Several measurement challenges needed to be met in the development of a reliable quantitative method for operationalizing the core dimensions model. Primarily, the model maintains that topographically similar OCD symptoms may be associated with different underlying motivations or subjective experiences. This renders quantitative assessment difficult, for reasons noted earlier in this paper. Secondly, there is great variability in the number of OCD symptoms reported by people with the disorder,

Model development

Neither of the two a priori models provided a good fit to the OC-CDI data. None of the fit indices attained criterion values for either the unidimensional single factor model (GFI=.809, AGFI=.047, SRMR=.191, CFI=.632) or the two-factor “obsessions and compulsions” model (GFI=.809, AGFI=−.906, SRMR=.191, CFI=.628). We therefore proceeded to the EFA, which yielded a 2-factor solution. Each of the four OC-CDI items loaded on one of two factors jointly accounting for 86.08% of the total variance.

Results and discussion

Factor loadings from all exploratory factor analyses are shown in Table A1 in appendix. Inspection of the rotated factor loadings indicated that all 31 items loaded significantly (>.33) on one of two clear well-defined factors, jointly accounting for 56.69% of the total variance. All 17 INC items loaded on the first factor extracted, with an eigenvalue of 11.84 and accounting for 38.19% of the total variance. Item loadings for this factor ranged from .52 to .87, with 10 items (59%) loading

Results

The simple structure two-factor HA and INC trait model found in Study 3 was replicated in the CFA of data from the nonclinical sample. Parameter estimates for each of the 20 items on the two latent factors are shown in Table 2. For both factors, parameter estimates on average were moderate with 60% of the items loading ≥.70, indicating factorial stability. The correlation between factors was .70. This model provided very good fit to the data according to all criteria (see Table 1).

The two

General discussion

The present series of studies set out to investigate the structural validity of a model which posits that underlying obsessive-compulsive symptoms are two core affective-motivational dimensions: harm avoidance and incompleteness. The research began by evaluating the model using a semi-idiographic interview to measure the two dimensions as they apply to current symptoms experienced by individuals with clinical OCD. It progressed to the use of a nomothetic self-report questionnaire for the same

Acknowledgments

This research was supported in part by funding from the Ontario Mental Health Foundation (OMHF), in the form of post-doctoral and New Investigator fellowships to Laura J. Summerfeldt. Portions of this research were presented at the meeting of the Canadian Psychological Association in Montreal, June 2005.The authors thank James Parker for his comments on drafts of the manuscript.

References (123)

  • A. Gentsch et al.

    Dysfunctional forward model mechanisms and aberrant sense of agency in obsessive-compulsive disorder

    Biological Psychiatry

    (2012)
  • M. Ghisi et al.

    In search of specificity: “not just right experiences” and obsessive-compulsive symptoms in non-clinical and clinical Italian individuals

    Journal of Anxiety Disorders

    (2010)
  • N.A. Gibbs

    Nonclinical populations in research on obsessive-compulsive disorder: A critical review

    Clinical Psychology Review

    (1996)
  • A.M. Graybiel et al.

    Toward a neurobiology of obsessive-compulsive disorder

    Neuron

    (2000)
  • N. Haslam et al.

    Subtyping obsessive-compulsive disorder: A taxometric analysis

    Behavior Therapy

    (2005)
  • P.H. Kloosterman et al.

    The obsessive-compulsive trait of incompleteness in parents of children with autism spectrum disorders

    Journal of Obsessive-Compulsive and Related Disorders

    (2013)
  • J.C. Lee et al.

    Perfectionism and sensory phenomena: Phenotypic components of obsessive-compulsive disorder

    Comprehensive Psychiatry

    (2009)
  • D. McKay

    Treating disgust reactions in contamination-based obsessive-compulsive disorder

    Journal of Behavior Therapy and Experimental Psychiatry

    (2006)
  • D. McKay et al.

    A critical evaluation of obsessive-compulsive disorder subtypes: Symptoms versus mechanisms

    Clinical Psychology Review

    (2004)
  • A.S. Pietrefesa et al.

    Moving beyond an exclusive focus on harm avoidance in obsessive compulsive disorder: Considering the role of incompleteness

    Behavior Therapy

    (2008)
  • A.S. Pietrefesa et al.

    Moving beyond an exclusive focus on harm avoidance in obsessive-compulsive disorder: Behavioral validation for the separability of harm avoidance and incompleteness

    Behavior Therapy

    (2009)
  • A. Pinto et al.

    Further development of Y-BOCS dimensions in the OCD collaborative genetics study: Symptoms vs. categories

    Psychiatry Research

    (2008)
  • R.K. Pitman

    A cybernetic model of obsessive-compulsive psychopathology

    Comprehensive Psychiatry

    (1987)
  • S. Rachman

    A cognitive theory of obsessions

    Behaviour Research and Therapy

    (1997)
  • A.S. Radomsky et al.

    Subtyping OCD: Prospects and problems

    Behavior Therapy

    (2005)
  • S.A. Rasmussen et al.

    The epidemiology and clinical features of obsessive compulsive disorder

    Psychiatric Clinics of North America

    (1992)
  • S.A. Rasmussen et al.

    Toward a neuroanatomy of obsessive-compulsive disorder revisited

    Biological Psychiatry

    (2013)
  • P.M. Salkovskis

    Obsessional–compulsive problems: A cognitive-behavioral analysis

    Behaviour Research and Therapy

    (1985)
  • S. Sarig et al.

    Obsessive-compulsive tendencies are related to indecisiveness and reliance on feedback in a neutral color judgment task

    Journal of Behavior Therapy and Experimental Psychiatry

    (2012)
  • E.A. Storch et al.

    Associations between miscellaneous symptoms and symptom dimensions: An examination of pediatric obsessive-compulsive disorder

    Behaviour Research and Therapy

    (2007)
  • E.A. Storch et al.

    Psychometric analysis of the Yale-Brown obsessive-compulsive scale second edition symptom checklist

    Journal of Anxiety Disorders

    (2010)
  • G.W. Allport

    Personality: A psychological interpretation

    (1937)
  • American Psychiatric Association

    Diagnostic and statistical manual of mental disorders

    (2000)
  • American Psychiatric Association

    Diagnostic and statistical manual of mental disorders

    (2013)
  • J.C. Anderson et al.

    The effect of sampling error on convergence, improper solutions, and goodness-of-fit indices for maximum likelihood confirmatory factor analysis

    Psychometrika

    (1984)
  • G. Andrews

    Anxiety, personality, and anxiety disorders

    International Review of Psychiatry

    (1991)
  • L. Baer

    Factor analysis of symptom subtypes of obsessive compulsive disorder and their relation to personality and tic disorders

    Journal of Clinical Psychiatry

    (1994)
  • D.H. Barlow

    Anxiety and its disorders: The nature and treatment of anxiety and panic

    (1988)
  • D.H. Barlow

    Unraveling the mysteries of anxiety and its disorders from the perspective of emotion theory

    American Psychologist

    (2000)
  • H.R. Beech

    Obsessional states

    (1974)
  • P.M. Bentler

    Comparative fit indices in structural models

    Psychological Bulletin

    (1990)
  • O.J. Bienvenu et al.

    Personality and anxiety disorders: A review

    Journal of Personality Disorders

    (2003)
  • T.A. Brown et al.

    Comorbidity among anxiety and mood disorders: Implications for treatment and DSM-IV

    Journal of Consulting and Clinical Psychology

    (1992)
  • A.T. Carr

    Compulsive neurosis: A review of the literature

    Psychological Bulletin

    (1974)
  • R.B. Cattell

    Personality and motivation: Structure and measurement

    (1957)
  • L.A. Clark

    Temperament as a unifying basis for personality and psychopathology

    Journal of Abnormal Psychology

    (2005)
  • C.R. Cloninger

    A unified biosocial theory of personality and its role in the development of anxiety states

    Psychiatric Developments

    (1986)
  • M.E. Coles et al.

    Initial data characterizing the progression from obsessions and compulsions to full-blown obsessive compulsive disorder

    Cognitive Therapy and Research

    (2012)
  • W.H. Cooper

    Ubiquitous halo

    Psychological Bulletin

    (1981)
  • L. Cronbach et al.

    Construct validity in psychological tests

    Psychological Bulletin

    (1955)
  • Cited by (0)

    View full text