7

The Five-Factor Model of Personality Disorders

Joshua D. Miller and Thomas A. Widiger

Trait-based, dimensional approaches to personality disorders (PDs) have finally entered mainstream psychiatric nosology with their inclusion in Section III of the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5; APA, 2013) and the upcoming release of ICD-11. The adoption of such approaches did not proceed smoothly or swiftly, however, as such models have been offered as empirically grounded approaches for well over 20 years (e.g., Costa & McCrae, 1992a; Frances, 1993; Livesley, Jackson, & Schroeder, 1992; Widiger & Trull, 1992). The DSM-5 Section III model, which assesses personality pathology via a combination of signs of personality dysfunction (Criterion A) and personality disorder traits (Criterion B), was included as an emerging model in Section III, in part, due to concerns with its empirical grounding. Although this model’s empirical grounding could have been better articulated and defended (Miller & Lynam, 2013), it is possible that inclusion in Section III was the best outcome available (Widiger, 2013), given its chilly reception by scholars and clinicians wed to more traditional personality disorder constructs and diagnostic approaches (e.g., Gunderson, 2010, 2013).

Substantial portions of this new diagnostic model of personality disorders – the use of pathological traits with clear ties to general or “normal” personality traits and the recreation of traditional PDs via a summation of traits – have ties to several empirical traditions, perhaps most notably work on the Five-Factor Model of PD (Widiger & Costa, 2012). In this chapter, we review the robust literature on the Five-Factor Model of PD and demonstrate its “power” with regard to efficiency, parsimony, validity, and utility. Importantly, throughout this chapter we take steps to demonstrate why this literature is so relevant and foundational to the changes being made to the major psychiatric taxonomies with regard to the conceptualization, assessment, and diagnosis of personality disorders and why a failure to adequately attend to it in the lead-up to the DSM-5 left the door open for substantial criticisms regarding the validity of the Section III diagnostic model of PD.

The Five-Factor Model of Personality: An Introduction

The Five-Factor Model (FFM) is the predominant model of personality structure (John, Naumann, & Soto, 2008). It has its roots in the “lexical” tradition: what is of most importance, interest, or meaning to persons will be naturally encoded within the language. The most important domains of personality are those with the greatest number of terms to describe and differentiate their various manifestations and nuances, and the structure of personality is provided by the empirical relationship among these trait terms (Goldberg, 1993). Lexical research in the English language, and all other languages examined, have converged well onto the Big Five or FFM.

The FFM includes the domains of neuroticism (or negative affectivity), extraversion (versus introversion), openness (or unconventionality), agreeableness (versus antagonism), and conscientiousness (or constraint). Each broad domain can be broken down into more specific components. For example, the domain of agreeableness versus antagonism includes compassion vs. callousness, morality vs. immorality, modesty vs. arrogance, affability vs. combativeness, and trust vs. distrust (Crowe, Lynam, & Miller, 2018). Each domain includes both adaptive and maladaptive personality traits (Widiger, Gore, Crego, Rojas, & Oltmanns, 2017). Consider again the domain of agreeableness versus antagonism. Most of the traits of agreeableness are adaptive (e.g., trusting, honest, generous, cooperative, and humble) but there are also maladaptive variants of these traits (e.g., gullible, guileless, selflessly sacrificial, subservient, and self-denigrating, respectively). Most of the traits of antagonism are maladaptive (e.g., cynical-suspicious, manipulative, boastful, and callous) but there are also adaptive variants of these traits (e.g., cautious-skeptical, savvy, confident, and tough-minded).

One of the compelling attributes of the FFM is its robustness, which is a natural consequence of accounting for virtually every trait term within the language as a result of its developmental ties to the lexical hypothesis. Other dimensional models of general personality are well understood in terms of the domains and facets of the FFM (O’Connor, 2017). For example, Neuroticism and Extraversion within Eysenck’s PEN model (Eysenck & Eysenck, 1970) can be considered isomorphic with Extraversion and Neuroticism from the FFM, whereas Eysenck’s Psychoticism dimension can be considered a blend of FFM Antagonism and low Conscientiousness (Costa & McCrae, 1995).

The FFM has amassed a considerable body of empirical support, including childhood antecedents (Mervielde, De Clercq, De Fruyt, & Van Leeuwen, 2005), multivariate behavior genetics with respect to its structure (Jarnecke & South, 2017), temporal stability across the lifespan (Roberts & DelVecchio, 2000), and cross-cultural replication (Allik & Realo, 2017). The FFM is associated with a wide array of important life outcomes, both positive and negative, including diverse forms of psychopathology (Bagby, Uliaszek, Gralnick, & Al-Dajani, 2017), mortality, divorce, and occupational attainment (Roberts, Kuncel, Shiner, Caspi, & Goldberg, 2007), as well as subjective well-being, social acceptance, criminality, and interpersonal conflict (Ozer & Benet-Martinez, 2006).

Describing PDs as Configurations of General Traits: Expert Ratings

Explicit descriptions of PDs as general trait configurations from the perspective of the FFM can be traced back, in part, to translations of DSM-III-R (APA, 1987) and IV (APA, 1994) PDs into the language of the FFM by Widiger and colleagues (Widiger, Trull, Clarkin, Sanderson, & Costa, 1994). These experts in PD rated the 30 facets of the FFM on their degree of relevance (low or high levels) to each DSM PD based on the DSM symptoms and text-based descriptions, as well as the empirical literature on each. For instance, antisocial PD was described by the following set of FFM facets: angry hostility (high), excitement seeking (high), straightforwardness (low), altruism (low), compliance (low), tendermindedness (low), dutifulness (low), self-discipline (low), and deliberation (low). Widiger and Lynam (1998) used the same approach to describe psychopathic PD, as assessed by Hare’s Psychopathy Checklist–Revised (Hare, 2003), and, like with antisocial PD described above, showed that facets/traits from the domains of agreeableness and conscientiousness were most critical, with more complex (i.e., highs and lows) relations for traits from the domains of neuroticism and extraversion.

Since this initial work, there have been more comprehensive studies in which a greater number of ratings were collected, some of which used academicians and others used practicing clinicians as raters. For instance, expert raters (i.e., individuals who have published a study or more on the PD they were asked to rate) were asked to describe individuals considered prototypical of psychopathy (Miller, Lynam, Widiger, & Leukefeld, 2001) and the 10 DSM-IV PDs (Lynam and Widiger, 2001) using the 30 facets of the FFM. Unlike the initial studies where a more dichotomous rating approach was used, here raters were asked to describe a prototypical case of each disorder using a 1 (prototypical case would be extremely low on given facet) to 5 scale (prototypical case would be extremely high on given facet).

Across the 11 PDs (psychopathy and 10 DSM-IV PDs), data were collected from an average of 17 experts; these raters proved to be relatively reliable and consistent with the initial Widiger et al. (1994) ratings. Given the reasonable agreement among raters, these ratings were then averaged to create an FFM prototype for each PD construct. As an example, the expert-rated FFM prototype for narcissistic personality disorder was characterized by elevations (i.e., mean expert ratings of a 4 or higher) on angry hostility, assertiveness, and excitement seeking, as well as low scores (i.e., mean expert ratings of 2 or lower) on all six facets of agreeableness (e.g., modesty, straightforwardness), dutifulness, self-discipline, deliberation, warmth, anxiety, depression, self-consciousness, and vulnerability. The academician ratings for psychopathy and the DSM-IV PDs can be found in Table 7.1. A second set of FFM PD prototypes were developed using the same approach except employing clinicians and a larger number of raters (mean of 31; Samuel & Widiger, 2004). As with the academician ratings, the clinical ratings manifested good inter-rater agreement for all the PDs (see Table 7.1).

Table 7.1FFM PD prototypes

	Psychopathy		Paranoid			Schizoid			Schizotypal			Antisocial
	AR	MA	AR	CR	MA	AR	CR	MA	AR	CR	MA	AR	CR	MA
Neuroticism
Anxiety	1.47	–.15	3.60	4.25	.27	2.23	3.06	.13	4.25	3.85	.27	1.82	2.00	.00
Angry Hostility	3.87	.29	4.00	4.39	.41	2.54	2.84	.19	3.08	3.42	.29	4.14	3.93	.27
Depression	1.40	.05	3.30	3.64	.35	3.15	3.42	.28	3.58	3.62	.39	2.45	2.70	.12
Self-consciousness	1.07	–.09	3.30	2.94	.29	3.31	3.37	.23	4.00	3.69	.32	1.36	1.63	.02
Impulsiveness	4.53	.24	2.90	3.17	.15	2.08	2.03	.00	3.17	3.16	.17	4.73	4.22	.27
Vulnerability	1.47	.00	3.60	3.36	.22	3.31	2.97	.14	3.75	3.96	.25	2.27	2.07	.04
Extraversion
Warmth	1.73	–.20	1.30	1.61	–.28	1.08	1.19	–.42	1.58	1.58	–.28	2.14	2.00	–.13
Gregariousness	3.67	.03	1.70	1.89	–.20	1.00	1.06	–.48	1.58	1.62	–.25	3.32	3.48	.02
Assertiveness	4.47	.16	2.90	3.25	–.08	1.54	1.90	–.22	2.17	2.04	–.13	4.23	4.07	.06
Activity	3.67	.07	2.90	3.19	–.08	1.92	2.00	–.25	2.25	2.23	–.13	4.00	4.00	.02
Excitement Seek	4.73	.31	2.20	2.42	–.01	1.38	1.71	–.21	2.17	2.12	–.04	4.64	4.30	.25
Positive Emotions	2.53	–.10	2.20	2.08	–.27	1.23	1.55	–.38	1.92	1.65	–.26	2.86	3.52	–.09
Openness
Fantasy	3.07	.05	2.90	3.14	.00	3.23	2.81	–.05	3.83	4.00	.14	2.82	3.48	.10
Aesthetics	2.33	–.01	2.20	2.54	–.05	2.77	2.42	–.06	3.17	3.31	.07	2.36	2.78	.00
Feelings	1.80	–.10	2.40	2.46	–.02	1.31	1.52	–.17	2.17	2.31	.03	2.27	2.41	–.02
Actions	4.27	.09	2.00	2.37	–.10	1.62	2.13	–.13	2.42	2.81	–.06	4.23	4.07	.10
Ideas	3.53	.03	3.50	3.29	–.03	3.38	3.45	.00	4.33	4.38	.09	2.91	3.26	.04
Values	2.87	.00	1.90	1.69	–.05	2.31	2.42	–.05	2.42	2.81	.01	3.00	3.48	.08
Agreeableness
Trust	1.73	–.34	1.00	1.19	–.45	2.38	1.68	–.28	2.08	2.04	–.31	1.45	1.70	–.22
Straightforwardness	1.13	–.61	2.00	1.89	–.24	2.77	2.42	–.09	3.00	2.46	–.16	1.41	1.41	–.37
Altruism	1.33	–.41	1.90	1.86	–.21	2.38	2.29	–.19	2.75	2.50	–.15	1.41	1.41	–.24
Compliance	1.33	–.48	1.40	1.92	–.27	3.00	2.77	–.08	2.50	2.65	–.13	1.77	1.81	–.32
Modesty	1.00	–.31	2.40	2.53	–.06	3.31	3.48	.08	3.08	3.27	.05	1.68	1.70	–.17
Tendermindedness	1.27	–.31	1.80	2.14	–.18	2.38	2.58	–.11	3.00	2.88	–.05	1.27	1.52	–.19
Conscientiousness
Competence	4.20	–.17	3.30	3.53	–.13	2.85	3.00	–.13	2.33	2.85	–.18	2.09	2.52	–.21
Order	2.60	–.17	3.70	3.56	.00	3.08	3.19	–.02	2.00	2.58	–.06	2.41	2.74	–.18
Dutifulness	1.20	–.32	3.40	3.39	–.10	3.00	3.16	–.08	2.50	2.77	–.10	1.41	1.52	–.29
Achievement Strive.	3.07	–.11	3.00	3.08	–.07	2.38	2.68	–.13	2.25	2.35	–.13	2.09	2.33	–.19
Self-discipline	1.87	–.22	3.50	3.19	–.14	3.15	3.10	–.12	2.67	2.77	–.18	1.81	1.85	–.25
Deliberation	1.60	–.38	3.80	3.56	–.09	3.23	3.71	–.02	2.67	3.73	–.10	1.64	1.96	–.38
Profile rs: AR-CR			.95*			.91*			.91*			.97*
AR-MA/CR-MA	.77*		.71*	.75*		.73*	.81*		.80*	.79*		.80*	.79*

AR = Academician Ratings (Lynam & Widiger, 2001; Miller et al., 2001); CR = Clinician Ratings (Samuel & Widiger, 2004); MA = Meta-analytically derived rs (Decuyper et al., 2009; Samuel & Widiger, 2008). Expert rated items rated 2 or lower are underlined; items rated 4 or higher are bolded. Correlations > .20 are italicized. *p < .01. OCPD = Obsessive-Compulsive PD.

Table 7.1 (cont.)

	Borderline			Histrionic			Narcissistic			Avoidant			Dependent			OCPD
	AR	CR	MA	AR	CR	MA	AR	CR	MA	AR	CR	MA	AR	CR	MA	AR	CR	MA
Neuroticism
Anxiety	4.04	4.25	.38	3.42	4.07	.00	2.33	2.71	.02	4.76	4.34	.41	4.32	4.46	.39	4.00	4.49	.16
Angry Hostility	4.75	4.56	.48	3.42	3.55	.08	4.08	3.90	.23	2.81	2.90	.29	2.42	2.95	.18	3.00	3.24	.10
Depression	4.17	4.03	.50	2.68	3.27	–.06	2.42	2.75	.03	3.95	3.72	.53	3.63	4.03	.41	3.18	3.76	.09
Self-consciousness	3.17	2.94	.35	2.00	2.45	–.11	1.50	1.67	–.03	4.67	4.45	.56	4.16	4.42	.42	3.29	3.86	.13
Impulsiveness	4.79	4.38	.34	4.32	4.16	.17	3.17	3.57	.14	1.62	2.14	.14	2.32	2.49	.17	1.53	2.18	–.07
Vulnerability	4.17	4.03	.39	3.95	3.90	.01	2.92	2.76	–.01	4.52	3.90	.40	4.32	4.64	.43	3.12	3.49	.03
Extraversion
Warmth	3.21	2.69	–.20	3.89	3.50	.26	1.42	2.05	–.07	2.33	2.45	–.35	3.84	3.49	–.03	2.06	2.24	–.07
Gregariousness	2.92	3.28	–.12	4.74	4.32	.35	3.83	3.95	.04	1.29	1.45	–.42	3.26	2.54	–.03	2.18	2.40	–.16
Assertiveness	3.17	3.69	–.09	3.84	3.39	.27	4.67	4.00	.19	1.19	1.52	–.39	1.32	1.46	–.21	3.00	3.03	–.01
Activity	3.29	3.56	–.10	4.16	3.94	.25	3.67	4.14	.09	2.05	2.07	–.29	2.26	2.00	–.12	3.35	3.31	.03
Excitement Seek	3.88	4.06	.06	4.47	4.13	.27	4.17	4.10	.16	1.24	1.55	–.23	2.26	1.69	–.06	1.59	1.88	–.12
Positive Emotions	2.63	3.16	–.26	4.16	3.80	.23	3.33	3.52	–.02	1.67	1.79	–.39	2.53	2.03	–.15	2.41	2.29	–.09
Openness
Fantasy	3.29	4.00	.13	4.37	4.13	.16	3.75	3.82	.11	3.14	3.07	.00	3.05	2.95	.05	2.06	2.52	–.09
Aesthetics	2.96	3.19	.05	3.53	3.60	.10	3.25	3.32	.04	3.05	2.69	–.03	2.89	2.58	.01	2.59	2.56	.01
Feelings	4.00	3.84	.09	4.16	4.13	.18	1.92	2.68	.05	3.43	3.07	–.04	3.74	3.45	.05	1.82	2.22	.01
Actions	4.00	3.78	–.03	4.21	3.70	.12	4.08	3.36	.04	2.00	1.83	–.20	2.21	1.79	–.13	1.53	1.76	–.12
Ideas	3.21	3.69	–.01	3.11	3.30	.04	2.92	3.09	.07	3.19	2.69	–.05	2.84	2.26	–.12	1.76	2.48	.03
Values	2.88	3.00	.05	3.63	3.50	.04	2.67	2.68	–.01	2.57	2.34	–.05	2.89	2.05	–.04	1.76	1.82	–.09
Agreeableness
Trust	2.21	1.69	–.29	4.00	3.39	.05	1.42	1.86	–.20	2.24	2.39	–.29	4.26	3.95	–.07	2.65	2.20	–.08
Straightforwardness	2.08	1.94	–.21	2.32	2.29	–.10	1.83	1.91	–.31	2.90	2.82	–.06	3.11	2.90	.00	3.47	3.06	.04
Altruism	2.46	2.31	–.18	2.21	2.52	.02	1.00	1.73	–.20	2.90	2.93	–.12	3.95	3.85	.03	2.76	2.63	.04
Compliance	2.00	1.81	–.27	2.53	2.90	–.12	1.58	1.77	–.26	3.52	3.21	–.02	4.68	4.50	.10	3.18	2.82	.01
Modesty	2.83	2.56	.03	2.32	2.20	–.16	1.08	1.23	–.37	4.33	3.68	.20	4.26	4.23	.16	3.06	3.17	.02
Tendermindedness	2.79	2.47	–.09	3.05	3.00	.02	1.50	1.77	–.17	3.43	3.43	–.02	3.89	3.79	.09	2.82	2.76	.00
Conscientiousness
Competence	2.71	2.78	–.29	2.37	2.68	–.01	3.25	3.00	.01	3.05	3.45	–.23	2.58	3.28	–.25	4.53	4.41	.19
Order	2.38	2.31	–.10	2.10	2.30	–.05	2.92	3.00	–.03	3.43	3.48	–.03	2.89	3.21	–.06	4.76	4.59	.25
Dutifulness	2.29	2.22	–.22	2.10	2.32	–.08	2.42	2.50	–.10	3.29	3.45	–.09	3.79	3.79	–.08	4.76	4.20	.25
Achievement Strive.	2.50	2.72	–.19	2.68	2.60	.04	3.92	3.18	.02	2.67	2.90	–.19	2.47	2.97	–.16	4.29	4.03	.25
Self-discipline	2.33	2.34	–.29	1.79	2.13	–.04	2.08	2.23	–.09	3.05	3.07	–.22	2.84	3.31	–.23	4.53	4.06	.21
Deliberation	1.88	2.09	–.27	1.74	1.94	–.16	2.25	2.45	–.13	3.43	3.62	–.01	3.00	3.36	–.06	4.59	4.37	.24
Profile rs: AR-CR	.93*			.95*			.95*			.96*			.90*			.94*
AR-MA/CR-MA	.84*	.77*		.86*	.79*		.81*	.87*		.78*	.76*		.60*	.63*		.91*	.92*

FFM PD Meta-Analyses

Following the development of expert ratings for these 11 PDs, several meta-analyses were published that examined the FFM trait correlates of the DSM PDs (Samuel & Widiger, 2008; Saulsman & Page, 2004) and psychopathy (Decuyper, De Pauw, De Fruyt, De Bolle, & De Clercq., 2009; O’Boyle, Forsyth, Banks, Story, & White, 2015). Using data from 18 samples (n = 3207), Samuel and Widiger (2008) documented the facet level FFM relations with DSM PDs. For instance, borderline PD was characterized by significant positive correlations with all six neuroticism facets (e.g., depressiveness, angry hostility) and negative correlations with traits from agreeableness (e.g., trust, compliance), conscientiousness (e.g., competence, self-discipline), and extraversion (e.g., positive emotions). Similarly, Decuyper et al. compiled data from 26 independent samples (n = 6913) to present the meta-analytic correlations between the 30 FFM traits and psychopathy scores. Psychopathy was most consistently negatively correlated with facets from agreeableness (e.g., straightforwardness, compliance) and conscientiousness (e.g., deliberation, dutifulness), and manifested both positive and negative correlations with neuroticism (positive: angry hostility, impulsiveness; negative: anxiety) and extraversion (positive: excitement seeking; negative: warmth). See Table 7.1 for the meta-analytically derived FFM profiles for psychopathy and DSM PDs.

In general, the academic and clinician FFM PD profiles were highly correlated with one another (range: .90–.95; mean r = .94) and converged with the meta-analytic profiles (academic ratings: mean r = .80; clinician ratings: mean r = .80), documenting the robustness of the FFM profiles. Dependent PD was the only case with substantive divergence between expected (i.e., expert) and obtained profiles such that both sets of expert ratings characterized individuals with dependent PD using high levels of agreeableness (e.g., trust, compliance, modesty) whereas the empirical data do not support this relation (see Miller & Lynam, 2008 and Lowe, Edmundson, & Widiger, 2009 for a discussion of these issues). Instead, lower levels of conscientiousness (paired with high neuroticism) may better characterize this PD from the perspective of the FFM.

Prototype Matching for FFM PDs

The previous findings demonstrated that traits can be used to reliably and robustly describe PDs within approaches and across them. We next examined whether personality traits scores on measures of the FFM could be used to assess PDs. In this vein, two scoring techniques were developed utilizing the expert ratings described above. The first strategy was a prototype matching approach in which an individual’s scores on a faceted measure of the FFM was quantitatively compared to one or more of the FFM PD prototypes using a double entry-q intraclass correlation (rICC), which quantifies the absolute similarity of the two set of traits (e.g., individual A’s scores on the 30 FFM facets vs. the FFM PD prototype).

Another means of scoring an individual’s FFM data with regard to the FFM PD prototypes is the additive count procedure (Miller, Bagby, Pilkonis, Reynolds, & Lynam, 2005). In this far simpler approach, an individual’s FFM PD scores are calculated by summing scores on only the FFM facets considered particularly relevant for the PDs of interest. The cutoffs typically used to determine a facet’s relevance are scores of 4 or higher or 2 or lower; facets that are deemed relevant at low levels (e.g., deliberation for antisocial PD) are reversed scored before being summed. The two scoring approaches for the FFM PDs yield scores that are highly correlated with one another (median r for DSM-IV PDs across two samples: .91; Miller et al., 2005). Given the similarity of the resultant scores, the count approach is preferred primarily because the scores are easier to calculate.

Validity of the FFM PD Scores: Convergent Validity

Several studies have examined the correlations between the FFM PD similarity scores and/or counts and DSM-IV PD symptom counts using self-report data. Miller (2019) conducted a meta-analytic review of the convergent validity correlations across these studies (k range: 11 [psychopathy] to 20 [borderline]; N range: 3094 [Cluster A PDs] to 4394 [borderline]). The studies included in the review scored the FFM data based on self-reported data from the NEO PI-R, interview data from the Structured Interview for the Five Factor Model (SIFFM; Trull & Widiger, 1997), and clinician ratings from the Five Factor Model Scoring Sheet (FFMSS; Few et al., 2010). The official DSM PD symptoms were assessed using interviews, expert consensus ratings, and self-report measures. The unweighted mean correlations ranged from .17 (Obsessive-Compulsive PD [OCPD]) to .60 (avoidant PD) with a median of .46. The largest convergent validity correlations for the FFM PDs were for avoidant, borderline, and psychopathy, whereas the smallest convergent validity effect sizes were found for OCPD, schizotypal, and histrionic PDs. In general, OCPD and histrionic demonstrate substantial heterogeneity in effect sizes with regard to their trait correlates (Samuel and Widiger, 2008), which likely affects their convergent correlations when scored using an FFM PD approach. Overall, however, the convergent validity correlations for the FFM PDs are relatively strong when considering that they are similar in size to those found when comparing two explicit measures of DSM PDs with one another (Miller, Few, & Widiger, 2012), are derived from studies that utilize a wide array of methodologies to assess the FFM facets and the PDs, and most of the FFM scores were derived from instruments that were not written to assess pathological variants of these traits.

Discriminant Validity/Comorbidity

Several studies have examined the convergent and discriminant validity of the FFM PD prototypes. In most cases, the FFM PD scores manifest their strongest correlations with the convergent DSM-IV PD (e.g., FFM antisocial and DSM-IV antisocial) and in cases where this is not true, the FFM PD score is typically slightly more correlated with another PD from the same cluster (e.g., FFM dependent and DSM avoidant) or another PD known to be comorbid with the target PD (e.g., schizoid and avoidant; e.g., Zimmerman, Rothschild, & Chelminski, 2005). Of course, finding strong discriminant validity for the FFM PDs scores is difficult as the DSM PDs themselves are notoriously deficient in this domain. One benefit of a trait-based approach to PDs, in fact, is that the co-occurrence found among DSM PDs can be modeled by the degree to which they share underlying FFM traits (i.e., Lynam & Widiger, 2001).

Criterion Validity: Convergence of Empirical Networks

Another, perhaps more important, way of testing the validity of the FFM PD scores is to examine the degree to which they can recreate the nomological networks associated with specific DSM-IV PDs. For example, one can compare the empirical profiles generated by the FFM PD scores with those generated by DSM-IV PD scores. For instance, Miller and colleagues (Miller, Reynolds, & Pilkonis, 2004; Miller et al., 2010) compared the empirical trait profile derived from FFM and DSM PD scores on the 15 traits from Clark’s (1993) Schedule for Nonadaptive and Adaptive Personality (SNAP) and found substantial absolute similarity (mean rICC = .73), suggesting both approaches are capturing the same underlying trait constructs. One can use a similar quantitative approach to compare the two sets of scoring using non-trait based correlates as well. Such an approach has been used to examine the similarity of FFM and DSM-based conceptualizations of borderline PD, which is a particularly important test given that some experts have posited that such trait-based approaches are likely insufficient for the assessment of this important construct (e.g., Gunderson, 2010, 2013). Both, Trull, Widiger, Lynam, and Costa (2003) and Miller and colleagues (Miller, Morse, Nolf, Stepp, & Pilkonis, 2012) demonstrated that FFM borderline PD scores yielded empirical profiles with a number of relevant constructs such as developmental history (e.g., child abuse; parental psychopathology), emotional experiences, pathological personality traits, and functioning, that were very similar to the relations modeled using official DSM borderline scores (i.e., combining data from these two studies: rICC = .87; see Table 7.2). The substantial absolute similarity in relations evinced from FFM and DSM PD scores is all the more noteworthy when considering that the FFM PDs are typically assessed with instruments that have no conceptual ties to the DSM and do not explicitly assess functional impairment. Similar studies offering support for the FFM PD perspective have been conducted for psychopathy (Derefinko & Lynam, 2007; Miller & Lynam, 2003; Miller et al., 2001) and antisocial PD (Gudonis, Miller, Miller, & Lynam, 2008).

Table 7.2Nomological network of DSM-IV and FFM BPD scores in two samples

	Borderline PD
	FFM BPD	DSM-IV BPD
	r	r
Childhood Abuse
Sexual T	.19	.21
Physical T	.20	.20
Parental Psychopathology
Bio. Parent – any disorder T	.26	.20
Bio. Father substance use disorder T	.23	.14
Bio. Father mood disorder T	.09	.10
Bio. Mother substance use disorder T	.05	.06
Bio. Mother mood disorder T	.21	.19
Psychological Distress
Depression M	.56	.58
Anxiety M	.58	.59
Distress M	.65	.54
Affect
Negative M	.56	.41
Positive M	–.44	–.29
Attachment Style
Anxiety M	.60	.48
Avoidance M	.29	.30
Informant-report PDs
Paranoid M	.40	.46
Schizoid M	.23	.27
Schizotypal M	.45	.46
Antisocial M	.43	.54
Borderline M	.53	.53
Histrionic M	.36	.50
Narcissistic M	.28	.36
Avoidant M	.48	.33
Dependent M	.36	.41
Obsessive-Compulsive M	.19	.22
Self-harm
No intent to die M	.39	.45
Intent to die M	.36	.56
Aggression – perpetration
Aggression M	.33	.38
Assault M	.20	.35
Aggression – victimization
Aggression M	.27	.31
Assault M	.21	.23
Interpersonal Functioning
Interpersonal sensitivity M	.73	.55
Interpersonal ambivalence M	.55	.38
Aggression M	.59	.47
Need for approval M	.59	.46
Lack of sociability M	.56	.44
Functioning
Distress M	.64	.62
Romantic M	.39	.44
Parental M	.44	.58
Occupational M	.46	.54
Social M	.45	.54
Distress on others M	.59	.76
Interpersonal T	.53	.39
Global dysfunction T	.52	.39
Profile Similarity (rICC)	.87*

Note. ^T = data from Trull et al. (2003); ^M = data from Miller et al. (2012)

BPD = Borderline PD.

Sex Differences

A useful trait-based model should be able to explain key questions in the field, especially if it is to claim to be a more parsimonious approach. This is true for issues related to comorbidity (Lynam & Widiger, 2001), as noted previously. Another area in which a trait-based approach might prove useful is in explaining commonly found sex differences among the DSM personality disorders, which has been plagued with questions as to whether such differences reflect issues of bias (in terms of criteria included in the DSM or how clinicians differentially apply the official criteria) or genuine differences. Lynam and Widiger (2007) examined whether sex differences for DSM PDs could be accounted for by sex differences on the FFM traits that underlie PDs by correlating sex differences for each FFM PD based on what is known about sex differences in the facets of the FFM with the actual sex differences (ds) found for the DSM PDs compiled via a meta-analytic review. The two sets of data were substantially related (r = .72), indicating that sex differences in DSM PDs may be due, in part, to differences on the general personality traits that comprise each PD. That is, PDs for which low agreeableness is a central component (e.g., antisocial and narcissistic PDs), men will score higher; PDs in which high neuroticism is a key aspects (e.g., borderline and dependent), women will tend to score higher.

Clinical Utility

There are many ways to examine the clinical utility of different approaches to the conceptualization, assessment, and diagnosis of PDs. One common approach (referred to often as a consumer preference approach) is to present clinicians the different models and inquire about their preferences (e.g., DSM vs. FFM-based models). A second approach that speaks more to actual, performance-based utility is to compare models as to their capabilities to provide important, clinically relevant information (e.g., treatment utilization or compliance). Several consumer preference types of surveys have been conducted in which DSM and FFM based conceptualizations are compared (e.g., Rottman, Ahn, Sanislow, & Kim, 2009; Samuel & Widiger, 2006; Spitzer, First, Shedler, Westen, & Skodol, 2008; Sprock, 2003). When comparable methods are used, however, the FFM fares as well or better than the DSM-based approach in terms of preferences (Mullins-Sweatt & Lengel, 2012).

The second and arguably better approach to clinical utility is to compare how well different approaches do in providing clinically relevant information. For example, Miller et al. (2010) examined the relations between the FFM PD prototypes, as measured by a brief clinician rating form, and several indices of impairment and found that the FFM PD counts were significantly related to overall impairment, occupational impairment, social impairment, and distress caused to others; more importantly, the FFM PD counts consistently accounted for greater unique variance in the impairment scores than did DSM-IV PD symptoms. Studies of FFM approaches to borderline PD have demonstrated substantial relations with self-harm behavior and multiple ratings of impairment (Miller et al., 2012; Trull et al., 2003). Similarly, FFM antisocial and borderline PD scores predicted externalizing behaviors across several years (Stepp & Trull, 2007). In fact, data suggest that personality variables are excellent targets of clinical interventions (Presnall, 2012) given (a) that they change with intervention (Roberts et al., 2017) and (b) that changes in PD traits portend change in DSM PD symptoms (e.g., Warner et al., 2004; Wright, Hopwood, & Zanarini, 2015).

FFM PD Scales

It is a strong testament to the validity of understanding the DSM personality disorders from the perspective of the FFM that one can use an existing measure of the FFM, such as the NEO PI-R (Costa & McCrae, 1992b) to provide an assessment of a PD that is just as valid as any existing, direct measure of that PD (Miller et al., 2012). However, there have now been developed a number of self-report measures of respective PDs from the perspective of the FFM, thereby bridging the gap between normal-range and disordered personality (Bagby & Widiger, 2018; Widiger, Lynam, Miller, & Oltmanns, 2012). Eight FFM PD inventories have been developed, including (but not limited to) the Elemental Psychopathy Assessment (EPA; Lynam et al., 2011), the Five Factor Borderline Inventory (FFBI; Mullins-Sweatt et al., 2012), the Five Factor Obsessive-Compulsive Inventory (FFOCI; Samuel, Riddell, Lynam, Miller, & Widiger, 2012), and the Five Factor Narcissism Inventory (FFNI; Glover, Miller, Lynam, Crego, & Widiger, 2012). Each was constructed by first identifying which facets of the FFM appeared to be most relevant for each respective personality disorder. The facet selections were based on researchers’ FFM descriptions of each PD (i.e., Lynam & Widiger, 2001), clinicians’ descriptions (i.e., Samuel & Widiger, 2004), and FFM-PD research (e.g., Samuel & Widiger, 2008). Scales were then constructed to assess the maladaptive variants of each facet that were specific to each personality disorder, yielding scales at both poles of all five domains of the FFM (Widiger et al., 2012).

The FFM PD scales have been validated in part by demonstrating convergence with alternative measures of the respective personality disorder (e.g., Lynam et al., 2011; Samuel, Lynam, Widiger, & Ball, 2012). Each of the scales have also been shown to have incremental validity over a respective NEO PI-R facet scale in accounting for variance within a respective personality disorder (e.g., Glover et al., 2012; Lynam et al., 2011; Mullins-Sweatt et al., 2012). Demonstrating that a maladaptive personality trait scale has incremental validity over a normal personality trait scale in accounting for variance in a measure of personality disorders is not though a particularly striking finding. However, the FFM PD inventories have also demonstrated incremental validity over the more traditional and established measures of the respective personality disorders (e.g., Glover et al., 2012; Mullins-Sweatt et al., 2012; Samuel, Riddell, et al., 2012), due in part to the fact that these self-report measures include quite a number of subscales assessing the more specific components of each heterogeneous PD syndrome.

Perhaps most importantly, the FFM PD scales have also demonstrated strong convergent and discriminant validity with respect to the domains of the FFM. FFM PD scales have even demonstrated convergent and discriminant validity with respect to individual facet scales. It should be noted though that facet-level predictions have not even been attempted for other comparable maladaptive trait scales, such as the Personality Inventory for DSM-5 (PID-5; Krueger, Derringer, Markon, Watson, & Skodol, 2012) and the Computerized Adaptive Test-Personality Disorder-Static Form (CAT-PD-SF; Simms et al., 2011).

Characterization of PDs as Collections of Pathological Traits: DSM-5 Section III

As noted earlier, the recognition of the role of traits in the conceptualization, assessment, and diagnosis of PDs took a major leap forward in DSM-5, as traits now play a fundamental role in the assessment of DSM-5 PDs in Section III of the manual. In what follows, the new alternative model of PD (AMPD) is reviewed, along with initial empirical evidence as to its performance.

Connection to the FFM

The DSM-5 AMPD involves two major components, an evaluation/consideration of the presence of personality dysfunction in two domains (self and interpersonal), as well as documentation of the presence of pathological personality traits. These pathological traits can then be used in place of the previous criteria lists to diagnose six of the ten DSM-IV PDs (schizotypal, antisocial, borderline, narcissistic, avoidant, OCPD) or to generate PD-Trait Specified diagnoses for cases where an individual manifests significant personality dysfunction paired with one or more trait elevations. Although the DSM-5 trait model and its description changed over the course of time (see Widiger, 2013 for a review), the final model comprises five broader domains (i.e., negative affectivity, detachment, antagonism, disinhibition, and psychoticism) and 25 more specific facets (e.g., anxiousness, withdrawal, grandiosity, irresponsibility, unusual beliefs and experiences). The domains of the DSM-5 trait model are explicitly linked to the FFM as they are described as “maladaptive variants of the extensively validated and replicated model of personality known as the ‘Big Five,’ or Five-Factor Model of personality (FFM)” (APA, 2013, p. 773), although this was not initially the case (Miller & Lynam, 2013).

The DSM-5 trait model is typically assessed using the Personality Inventory for DSM-5 (Krueger et al., 2012; see Maples et al., 2015 for a briefer, faceted version as well). Miller (2019) examined the convergence between the DSM-5 and FFM domains across seven studies (n = 2471); convergent correlations ranged from .20 (psychoticism – openness) to .71 (negative affectivity – neuroticism) with a mean correlation of .56. Despite differences in how these models were assessed across these studies, with regard to FFM-based instruments (e.g., NEO PI-R; Five Factor Model Rating Form) and raters (self; clinical ratings), the effect sizes did not vary dramatically except in the case of openness. For this domain, the convergent correlations ranged from −.18 to .46. Not surprisingly, there continues to be an ongoing debate as to the nature of the relations between these dimensions (e.g., Chmielewski, Bagby, Markon, Ring, & Ryder, 2014; Edmundson, Lynam, Miller, Gore, & Widiger, 2011).

Expert Ratings of the DSM-5 Section III Traits Associated with DSM-IV/5 PDs

The DSM-5 Personality and Personality Disorder (P & PD) Work Group specified the manner in which the AMPD traits would be used to diagnose the DSM-IV PDs (see Table 7.3). For instance, the DSM-5 AMPD approach states that antisocial PD is diagnosed using the following traits (in addition to evidence of both self and interpersonal dysfunction): manipulativeness, callousness, deceitfulness, hostility, risk taking, impulsivity, and irresponsibility. In addition to the ratings provided by the DSM-5 P & PD Work Group, independent expert ratings of the relevance of DSM-5 traits to each DSM-IV PD were collected by Samuel, Lynam, Widiger, and Ball (2012). These ratings were collected by asking individuals who had published on the PD for which they provided ratings to rate the relevance of all AMPD traits in relation to a specific PD using a 0 (not at all or very little) to 3 (extremely descriptive) metric.¹ See Table 7.3 for these expert ratings. In general, the DSM-5 P & PD Work Group trait assignments were correlated (traits included in the DSM-5 count were given a “1,” those not included in a given PD diagnosis were given a “0”) with these expert ratings with correlations ranging from .51 (paranoid) to .91 (schizoid) with a median of .73. A comparison of these two sets of ratings demonstrates where the two diverge. For instance, the DSM-5 P & PD Work Group chose only two traits for the diagnosis of narcissistic PD – grandiosity and attention seeking – whereas the Samuel et al. experts rated grandiosity (3.00), manipulativeness (2.38), and callousness (2.07) as being most prototypical of this disorder (attention seeking, which the DSM-5 P & PD Work Group included, was given only a rating of 1.83 by the Samuel et al.). Decisions as to which traits to include in which PD may have overemphasized concerns with discriminant validity at the cost of overall construct validity. Similarly, the experts included in Samuel, Lynam, et al. (2012) chose submissiveness as being relevant to avoidant PD, which was not included by the DSM-5 P & PD Work Group; conversely, the AMPD includes the traits of withdrawal, intimacy avoidance, and anhedonia, which the experts did not feel were as relevant.

Table 7.3Academician and DSM-5 Ratings of PDs using the DSM-5 trait model

	SCT		APD		BPD		NPD		AVD		OCPD
	AR	D5	AR	D5	AR	D5	AR	D5	AR	D5	AR	D5
Submissiveness	0.43		0.09		1.43		0.14		2.07		0.54
Depressivity	0.59		0.35		1.85	x	0.55		1.67		0.95
Separation insecurity	0.43		0.09		2.69	x	0.62		1.07		0.54
Anxiousness	1.64		0.23		1.93	x	0.83		2.43	x	1.62
Emotional lability	0.64		1.36		2.79	x	1.28		0.43		0.33
Suspiciousness	2.50	x	1.57		1.32		1.45		0.79		0.38
Perseveration	0.62		0.14		0.45		0.38		0.29		2.46	x
Restricted affectivity	1.93	x	1.62		0.29		0.86		1.21		1.46	x
Withdrawal	2.50	x	0.93		0.39		0.36		1.79	x	0.93
Intimacy avoidance	2.29		1.10		0.67		1.14		1.79	x	0.92	x
Anhedonia	1.79		0.48		0.57		0.41		1.36	x	0.85
Manipulativeness	0.29		2.95	x	0.95		2.38		0.07		0.85
Deceitfulness	0.29		2.67	x	0.69		1.59		0.07		0.31
Hostility	0.54		2.50	x	1.42	x	1.69		0.07		0.96
Callousness	0.47		2.84	x	1.24		2.07		0.00		0.79
Attention seeking	0.36		1.43		1.10		1.83	x	0.00		0.23
Grandiosity	0.50		2.57		0.55		3.00	x	0.36		1.00
Irresponsibility	0.79		2.76	x	1.12		0.86		0.00		0.15
Impulsivity	0.71		2.62	x	2.48	x	0.93		0.07		0.31
Distractibility	1.36		1.38		1.12		0.17		0.50		0.46
Risk taking	1.50		2.85	x	2.23	x	1.85		0.50		0.62
Rigid Perfectionism	0.74		0.19		0.52		0.83		0.55		3.00	x
Eccentricity	2.79	x	0.24		0.96		0.18		0.04		0.31
Cognitive/perceptual dysregulation	2.00	x	0.05		1.70		0.07		0.07		0.23
Unusual beliefs/experiences	2.90	x	0.17		0.56		0.14		0.04		0.23
Profile rs	0.79*		0.73*		0.82*		0.53*		0.68*		0.73*

Note: AR = Academician Ratings (compiled from Samuel, Lynam, et al., 2012); D5 = DSM-5 Trait assignments; SCT = Schizotypal; APD = Antisocial Personality Disorder; BPD = Borderline; NPD = Narcissistic; AVD = Avoidant; OCPD = Obsessive-Compulsive Data compiled from Samuel, Lynam, et al., 2012. Ratings of 2 or higher are bolded. Profile correlations calculated by replacing “xs” with 1 for the DSM-5 ratings (and traits without an “x” were given a 0).

Validity of the DSM-5 Trait PD Counts: Convergent Validity

Several studies have examined the correlations between the traditional DSM-5 Section II PD counts and the AMPD scores (e.g., Anderson, Snider, Sellbom, Krueger, & Hopwood, 2014; Hopwood, Thomas, Markon, Wright, & Krueger, 2012; Miller, Few, Lynam, & MacKillop, 2015; Samuel, Hopwood, Krueger, Thomas, & Ruggero, 2013). Although there are a variety of ways one might calculate counts (see Samuel et al., 2013 for a review), the procedure that is used most commonly and is consistent with the one used in the FFM literature simply sums scores for each relevant facet. Averaged across these four studies, the convergent validity correlations for the Section III PD counts and the Section II PD scores range from .39 (obsessive-compulsive) to .71 (borderline) with a median of .60 (see Table 7.4). The size of these correlations were relatively similar across the samples despite differences in sample composition (e.g., Miller et al., 2015: community participants in mental health treatment; Anderson et al., 2014; Hopwood et al., 2012; Samuel et al., 2013: undergraduates) and assessment of the Section II and III PDs (Miller et al., 2015: clinical ratings; Anderson et al., 2014; Hopwood et al., 2012; Samuel et al., 2013: self-reports). It is also noteworthy that the average convergent correlations manifested by the Section III PD counts and Section II PD scores were significantly correlated (r = .72) with the number of traits used to assess each PD in the Section III approach (range: 2 [narcissistic] to 7 [antisocial; borderline]), suggesting that greater convergence may be attainable for several PDs if additional traits are added to the Section III DSM-5 PDs.

Table 7.4Convergent correlations among DSM-5 PD trait counts and Section II DSM-5 PDs

	Anderson et al., 2014	Hopwood et al., 2012	Miller et al., 2015	Samuel et al., 2013	Mean r
DSM PD Counts	(N = 397)	(N = 808)	(N = 109)	(N = 1025)
Schizotypal	.50	.70	.56	.71	.63
Antisocial	.54	.68	.81	.61	.67
Borderline	.62	.70	.81	.66	.71
Narcissistic	.45	.60	.53	.58	.54
Avoidant	.50	.57	.55	.60	.56
OCPD	.03	.56	.43	.49	.39

Note: OCPD = Obsessive-Compulsive PD. Anderson et al. (2014) and Hopwood et al. (2012) did not present these exact analyses but were communicated via personal communication (December 6 and 18, 2017, respectively).

Discriminant Validity/Comorbidity

Miller and colleagues (2015) examined the discriminant validity of the DSM-5 PD counts and found that for seven of the ten PDs they manifested their largest correlation (or tied for the largest) with the corresponding DSM-5 Section II PD. For two of the remaining three (paranoid and narcissistic) they manifested slightly higher correlations with PDs from the same cluster (schizotypal and histrionic, respectively). Much like the findings reported previously for the FFM PDs, one would expect that the DSM-5 Section III PD counts would also recreate the comorbidity found among the Section II PDs; Miller et al. (2015) found that this was the case as the patterns of relations among the two sets of PD scores were significantly correlated with one another (r = .78) and that the comorbidity among the DSM-5 Section III PD counts was significantly associated with the number of traits shared among the PDs (r = .76).

Criterion Validity: Convergence of Empirical Networks

Miller et al. (2015) also examined whether the DSM-5 Section II and III PDs manifested similar empirical networks with relation to the 30 general traits from the FFM. To do this, the two sets of DSM-5 PD scores were first correlated with the 30 facets of the FFM and the absolute similarities between the two sets of correlations were tested. The intraclass correlations among these FFM trait profiles for the ten PDs ranged from .59 (obsessive-compulsive) to .98 (borderline) with a mean of .90. While most demonstrated nearly perfectly correlated trait profiles, more moderate overlap was found for the trait correlates of the Section II and III OCPD scores. Here the correlations differed primarily in relation to facets from extraversion with the AMPD OCPD scores manifesting substantially larger negative correlations with traits such as gregariousness, warmth, and positive emotions. It is worth nothing that the original DSM-5 trait based diagnosis of OCPD involved only two facets: perseveration and rigid perfectionism. However, this diagnosis was revised prior to inclusion in the DSM-5 and two other traits were added: restricted affectivity and intimacy avoidance. Miller et al. (2015) demonstrated that removal of these two “new” traits from the Section III OCPD count resulted in an increased convergent correlation with the Section II OCPD scores (r = .56 for two traits vs. .43 for all four traits) and better convergence with the FFM profile generated by the DSM-5 Section II OCPD scores (two traits: r_ICC = .78; four traits: r_ICC = .59). This does not speak to whether these traits belong in the AMPD or not, but simply demonstrates that their inclusion is responsible, in part, for the lower convergence with the traditional, DSM-5 Section II OCPD construct.

Sex Differences

Mirroring the findings for the FFM, it is possible that the sex differences found for DSM PDs might be explained by differences on the AMPD pathological traits. Based on the gender differences found for the DSM-5 traits in the Few et al. (2013) study, men had higher scores on risk taking, restricted affect, and eccentricity, whereas women had higher scores for traits such as emotional lability and depressivity. Similar to the Lynam and Widiger (2007) finding using FFM traits, sex differences on the DSM-5 Section III PD trait counts were significantly correlated with the sex differences reported for the DSM PDs on the basis of Lynam and Widiger’s meta-analytic review (r = .64). As with the FFM data, it seems that sex differences in PDs may be due, at least in part, to differences in pathological personality traits that comprise these disorders.

Clinical Utility

As with the FFM, there are now results that speak to the clinical utility of the DSM-5 Section III approach from the perspective of clinicians’ preferences, as well as data that examine these models in relation to clinically relevant outcomes. Morey, Skodol, and Oldham (2014) compared clinicians’ preference for the DSM-5 Section II and III PD models and found that for five of six outcomes (e.g., ease of use; communicating with patients; communicating with professionals; useful for formulating intervention plans), the DSM-5 Section III trait model was seen as having greater clinical utility than the DSM-5 Section II PD approach.

With regard to a focus on clinically relevant correlates and outcomes, Few and colleagues (2013) demonstrated that the DSM-5 Section III traits were substantially correlated with interview-based ratings of personality impairment including impairments in identity (mean r = .48), self-directness (mean r = .46), empathy (mean r = .44), and intimacy (mean r = .50). The DSM-5 Section III traits were also significantly correlated with DSM-5 Section II PDs, as well as symptoms of anxiety, depression, and overall distress. With regard to incremental validity, the DSM-5 traits provided twice the incremental validity in the impairment variables above and beyond the variance accounted for by the DSM-5 Section II PDs (mean ΔR² = .11) as compared to that provided by the DSM-5 Section II PDs above the Section III trait domains (mean ΔR² = .06). With regard to symptoms of depression and anxiety, as well as general distress, the DSM-5 traits again accounted for additional variance over the Section II PDs (mean ΔR² = .08), although the PDs accounted for additional variance as well (mean ΔR² = .08).²

Conclusions

The inclusion of a trait-model in the DSM-5 represented an important advance towards an empirically based, valid, and clinically useful approach to research and treatment of personality pathology. There is a robust literature on the validity and utility of general trait models, much of which generalizes to the DSM-5 pathological trait model. We believe the existing empirical base could have been used to provide a stronger foundation for the DSM-5 PD proposal (Miller & Lynam, 2013) as much of this pertinent literature was ignored in publications originally put out by the DSM-5 P & PD Work Group (Blashfield & Reynolds, 2012; Lilienfeld, Watts, & Smith, 2012). The best chance that the DSM-5 AMPD, or a similar trait-based model, has of eventually becoming the sole or primary diagnostic approach used in the future iterations of the DSM is to combine the existing literature on the FFM (and other trait models) approaches to PDs with the rapidly growing research on the DSM-5 pathological trait model. The integration of the extant empirical literature on trait models of PDs with the burgeoning research on the AMPD would go far in rebutting claims that the DSM-5 Section III PD model, as least the trait portion, represents a brand new and untested model. Another important task will be to work towards building consensus within the field, a Herculean task, given the substantial criticisms that have been levied against various aspects of the DSM-5 Section II PD approach (Gunderson, 2010, 2013; Livesley, 2012; Shedler et al., 2010). Moving forward, it will be important that objective considerations of the existing data drive decisions as to how PDs are conceptualized, assessed, and diagnosed in the official diagnostic nosology.

References

Allik, J., & Realo, A. (2017). Universal and specific in the five factor model. In T. A. Widiger (Ed.), The Oxford Handbook of the Five Factor Model (pp. 173–190). New York: Oxford University Press.

American Psychiatric Association. (1987). Diagnostic and Statistical Manual of Mental Disorders (revised 3rd ed.). Washington, DC: American Psychiatric Association.

American Psychiatric Association. (1994). Diagnostic and Statistical Manual of Mental Disorders (4th ed.). Washington, DC: American Psychiatric Association.

American Psychiatric Association. (2013). Diagnostic and Statistical Manual of Mental Disorders (5th ed.). Arlington, VA: American Psychiatric Publishing.

Anderson, J., Snider, S., Sellbom, M., Krueger, R., & Hopwood, C. (2014). A comparison of the DSM-5 Section II and Section III personality disorder structures. Psychiatry Research, 216, 363–372.

Bagby, R. M., Uliaszek, A. A., Gralnick, T. M., & Al-Dajani, N. (2017). Axis I disorders. In T. A. Widiger (Ed.), The Oxford Handbook of the Five Factor Model (pp. 479–506). New York: Oxford University Press.

Bagby, R. M., & Widiger, T. A. (2018). Five factor model personality disorder scales: An introduction to a special section on assessment of maladaptive variants of the five factor model. Psychological Assessment, 30, 1–9.

Blashfield, R. K., & Reynolds, S. M. (2012). An invisible college view of the DSM-5 personality disorder classification. Journal of Personality Disorders, 26, 821–829.

Chmielewski, M., Bagby, R. M., Markon, K., Ring, A. J., & Ryder, A. G. (2014). Openness to experience, intellect, schizotypal personality disorder, and psychoticism: Resolving the controversy. Journal of Personality Disorders, 28, 483–499.

Clark, L. A. (1993). Manual for the Schedule for Nonadaptive and Adaptive Personality (SNAP). Minneapolis, MN: University of Minnesota Press.

Costa, P. T., & McCrae, R. R. (1992a). The five-factor model of personality and its relevance to personality disorders. Journal of Personality Disorders, 6, 343–359.

Costa, P. T., & McCrae, R. R. (1992b). Revised NEO Personality Inventory (NEO-PI-R) and NEO Five-Factor Inventory (NEO-FFI) Professional Manual. Odessa, FL: Psychological Assessment Resources.

Costa, P. T., & McCrae, R. R. (1995). Primary traits of Eysenck’s P-E-N model: Three- and five-factor solutions. Journal of Personality and Social Psychology, 69, 308–317.

Crowe, M. L., Lynam, D. R., & Miller, J. D. (2018). Uncovering the structure of agreeableness from self-report measures. Journal of Personality, 86, 771–787.

Decuyper, M., De Pauw, S., De Fruyt, F., De Bolle, M., & De Clercq, B. J. (2009). A meta‐analysis of psychopathy‐, antisocial PD‐ and FFM associations. European Journal of Personality, 23, 531–565.

Derefinko, K. J., & Lynam, D. R. (2007). Using the FFM to conceptualize psychopathy: A test using a drug abusing sample. Journal of Personality Disorders, 21, 638–656.

Edmundson, M., Lynam, D. R., Miller, J. D., Gore, W. L., & Widiger, T. A. (2011). A five-factor measure of schizotypal personality traits. Assessment, 18, 321–334.

Eysenck, S. G., & Eysenck, H. J. (1970). Crime and personality: An empirical study of the three-factor theory. British Journal of Criminology, 10, 225–239.

Few, L. R., Miller, J. D., Morse, J. Q., Yaggi, K. E., Reynolds, S. K., & Pilkonis, P. A. (2010). Examining the reliability and validity of clinician ratings on the Five-Factor Model score sheet. Assessment, 17, 440–453.

Few, L. R., Miller, J. D., Rothbaum, A., Meller, S., Maples, J., Terry, D., … MacKillop, J. (2013). Examination of the Section III DSM-5 diagnostic system for personality disorders in an outpatient clinical sample. Journal of Abnormal Psychology, 22, 1057–1069.

Frances, A. (1993). Dimensional diagnosis of personality: Not whether, but when and which. Psychological Inquiry, 4, 110–111.

Glover, N., Miller, J. D., Lynam, D. R., Crego, C., & Widiger, T. A. (2012). The Five-Factor Narcissism Inventory: A five-factor measure of narcissistic personality traits. Journal of Personality Assessment, 94, 500–512.

Goldberg, L. R. (1993). The structure of phenotypic personality traits. American Psychologist, 48, 26–34.

Gudonis, L. C., Miller, D. J., Miller, J. D., & Lynam, D. R. (2008). Conceptualizing personality disorders from a general model of personality functioning: Antisocial personality disorder and the five-factor model. Personality and Mental Health, 2, 249–264.

Gunderson, J. G. (2010). Revising the borderline diagnosis for DSM-V: An alternative proposal. Journal of Personality Disorders, 24, 694–708.

Gunderson, J. G. (2013). Seeking clarity for future revisions of the personality disorders in DSM-5. Personality Disorders: Theory, Research, and Treatment, 4, 368–376.

Hare, R. D. (2003). The Psychopathy Checklist–Revised. Multi-Health Systems, Toronto, Ontario, Canada.

Hopwood, C. J., Thomas, K. M., Markon, K. E., Wright, A. G., & Krueger, R. F. (2012). DSM-5 personality traits and DSM-IV personality disorders. Journal of Abnormal Psychology, 121, 424–432.

Jarnecke, A. M., & South, S. C. (2017). Behavior and molecular genetics of the five-factor model. In T. A. Widiger (Ed.), The Oxford Handbook of the Five Factor Model (pp. 301–318). New York: Oxford University Press.

John, O. P., Naumann, L. P., & Soto, C. J. (2008). Paradigm shift to the integrative Big Five trait taxonomy: History, measurement, and conceptual issues. In O. P. John, R. R. Robins, & L. A. Pervin (Eds.), Handbook of Personality: Theory and Research (3rd. ed., pp. 114–158). New York: Guilford Press.

Krueger, R. F., Derringer, J., Markon, K., Watson, D., & Skodol, A. (2012). Initial construction of a maladaptive personality trait model and inventory for DSM-5. Psychological Medicine, 42, 1879–1890.

Lilienfeld, S. O., Watts, A. L., & Smith, S. F. (2012). The DSM revision as a social psychological process: A commentary on Blashfield and Reynolds. Journal of Personality Disorders, 26, 830–834.

Livesley, J. (2012). Tradition versus empiricism in the current DSM‐5 proposal for revising the classification of personality disorders. Criminal Behaviour and Mental Health, 22, 81–90.

Livesley, W. J., Jackson, D. N., & Schroeder, M. L. (1992). Factorial structure of traits delineating personality disorders in clinical and general population samples. Journal of Abnormal Psychology, 101, 432–440.

Lowe, J. R., Edmundson, M., & Widiger, T. A. (2009). Assessment of dependency, agreeableness, and their relationship. Psychological Assessment, 21, 543–553.

Lynam, D. R., Gaughan, E. T., Miller, J. D., Miller, D. J., Mullins-Sweatt, S., & Widiger, T. A. (2011). Assessing the basic traits associated with psychopathy: Development and validation of the Elemental Psychopathy Assessment. Psychological Assessment, 23, 108–124.

Lynam, D. R., & Widiger, T. A. (2001). Using the five-factor model to represent the DSM-IV personality disorders: An expert consensus approach. Journal of Abnormal Psychology, 110, 401–412.

Lynam, D. R., & Widiger, T. A. (2007). Using a general model of personality to understand sex differences in the personality disorders. Journal of Personality Disorders, 21, 583–602.

Maples, J. L., Carter, N. T., Few, L. R., Crego, C., Gore, W. L., Samuel, D. B., … Miller, J. D. (2015). Testing whether the DSM-5 personality disorder trait model can be measured with a reduced set of items: An item response theory investigation of the Personality Inventory for DSM-5. Psychological Assessment, 27, 1195–1210.

Mervielde, I., De Clercq, B., De Fruyt, F., & Van Leeuwen, K. (2005). Temperament, personality, and developmental psychopathology as childhood antecedents of personality disorders. Journal of Personality Disorders, 19, 171–201.

Miller, J. D. (2019). Personality disorders as collections of traits. In D. B. Samuel & D. L. Lynam (Eds.), Using Basic Personality Research to Inform Personality Pathology (pp. 40–69). New York: Oxford University Press.

Miller, J. D., Bagby, R. M., Pilkonis, P. A., Reynolds, S. K., & Lynam, D. R. (2005). A simplified technique for scoring the DSM-IV personality disorders with the five-factor model. Assessment, 12, 404–415.

Miller, J. D., Few, L. R., Lynam, D. R., & MacKillop, J. (2015). Pathological personality traits can capture DSM-IV personality disorder types. Personality Disorders: Theory, Research, and Treatment, 6, 32–40.

Miller, J. D., Few, L. R., & Widiger, T. A. (2012). Assessment of personality disorders and related traits: Bridging DSM-IV-TR and DSM-5. In T. A. Widiger (Ed.), The Oxford Handbook of Personality Disorders (pp. 108–140). New York: Oxford University Press.

Miller, J. D., & Lynam, D. R. (2003). Psychopathy and the five factor model of personality: A replication and extension. Journal of Personality Assessment, 81, 168–178.

Miller, J. D., & Lynam, D. R. (2008). Dependent personality disorder: Comparing an expert generated and empirically derived five-factor model personality disorder count. Assessment, 15, 4–15.

Miller, J. D., & Lynam, D. R. (2013). Missed opportunities in the DSM-5 section III personality disorder model. Personality Disorders: Theory, Research, and Treatment, 4, 365–366.

Miller, J., Lynam, D., Widiger, T., & Leukefeld, C. (2001). Personality disorders as extreme variants of common personality dimensions: Can the five factor model adequately represent psychopathy? Journal of Personality, 69, 253–276.

Miller, J. D., Maples, J., Pryor, L. R., Morse, J. Q., Yaggi, K., & Pilkonis, P. A. (2010). Using clinician-rated five-factor model data to score the DSM-IV personality disorders. Journal of Personality Assessment, 92, 296–305.

Miller, J. D., Morse, J. Q., Nolf, K., Stepp, S. D., & Pilkonis, P. A. (2012). Can DSM-IV borderline personality disorder be diagnosed via dimensional personality traits? Implications for the DSM-5 personality disorder proposal. Journal of Abnormal Psychology, 121, 944–950.

Miller, J. D., Reynolds, S. K., & Pilkonis, P. A. (2004). The validity of the five-factor model prototypes for personality disorders in two clinical samples. Psychological Assessment, 16, 310–322.

Miller, J. D., Sleep, C. E., & Lynam, D. R. (2018). DSM-5 alternative model of personality disorder: Testing the trait perspective captured in criterion B. Current Opinion in Psychology, 21, 50–54.

Morey, L. C., Skodol, A. E., & Oldham, J. M. (2014). Clinician judgments of clinical utility: A comparison of DSM-IV-TR personality disorders and the alternative model for DSM-5 personality disorders. Journal of Abnormal Psychology, 123, 398–405.

Mullins-Sweatt, S. N., Edmundson, M., Sauer-Zavala, S., Lynam, D. R., Miller, J. D., & Widiger, T. A. (2012). Five-factor measure of borderline personality traits. Journal of Personality Assessment, 94, 475–487.

Mullins‐Sweatt, S. N., & Lengel, G. J. (2012). Clinical utility of the five‐factor model of personality disorder. Journal of Personality, 80, 1615–1639.

O’Boyle, E. H., Forsyth, D. R., Banks, G. C., Story, P. A., & White, C. D. (2015). A meta‐analytic test of redundancy and relative importance of the dark triad and five‐factor model of personality. Journal of Personality, 83, 644–664.

O’Connor, B. P. (2017). Robustness. In T. A. Widiger (Ed.), The Oxford Handbook of the Five Factor Model (pp. 151–172). New York: Oxford University Press.

Ozer, D. J., & Benet-Martinez, V. (2006). Personality and the prediction of consequential outcomes. Annual Review of Psychology, 57, 401–421.

Presnall, J. R. (2012). Disorders of personality: Clinical treatment from a five-factor perspective. In T. A. Widiger & P. T Costa (Eds.), Personality Disorders and the Five-Factor Model of Personality (3rd ed., pp. 409–432). Washington, DC: APA.

Roberts, B. W., & DelVecchio, W. F. (2000). The rank-order consistency of personality traits from childhood to old age: A quantitative review of longitudinal studies. Psychological Bulletin, 126, 3–25.

Roberts, B. W., Kuncel, N. R., Shiner, R., Caspi, A., & Goldberg, L. R. (2007). The power of personality: The comparative validity of personality traits, socioeconomic status, and cognitive ability for predicting important life outcomes. Perspectives on Psychological Science, 2, 313–345.

Roberts, B. W., Luo, J., Briley, D. A., Chow, P. I., Su, R., & Hill, P. L. (2017). A systematic review of personality trait change through intervention. Psychological Bulletin, 143, 117–141.

Rottman, B., Ahn, W. K., Sanislow, C., & Kim, N. (2009). Can clinicians recognize DSM-IV personality disorders from five-factor model descriptions of patient cases? American Journal of Psychiatry, 166, 427–433.

Samuel, D. B., Hopwood, C. J., Krueger, R. F., Thomas, K. M., & Ruggero, C. J. (2013). Comparing methods for scoring personality disorder types using maladaptive traits in DSM-5. Assessment, 20, 353–361.

Samuel, D. B., Lynam, D. R., Widiger, T. A., & Ball, S. A. (2012). An expert consensus approach to relating the proposed DSM-5 types and traits. Personality Disorders: Theory, Research, and Treatment, 3, 1–16.

Samuel, D. B., Riddell, A. D. B., Lynam, D. R., Miller, J. D., & Widiger, T. A. (2012). A five-factor measure of obsessive-compulsive personality traits. Journal of Personality Assessment, 94, 456–465.

Samuel, D. B., & Widiger, T. A. (2004). Clinicians’ personality descriptions of prototypic personality disorders. Journal of Personality Disorders, 18, 286–308.

Samuel, D. B., & Widiger, T. A. (2006). Clinicians’ judgments of clinical utility: A comparison of the DSM-IV and five-factor models. Journal of Abnormal Psychology, 115, 298–308.

Samuel, D. B., & Widiger, T. A. (2008). A meta-analytic review of the relationships between the five-factor model and DSM-IV-TR personality disorders: A facet level analysis. Clinical Psychology Review, 28, 1326–1342.

Saulsman, L. M., & Page, A. C. (2004). The five-factor model and personality disorder empirical literature: A meta-analytic review. Clinical Psychology Review, 23, 1055–1085.

Shedler, J., Beck, A., Fonagy, P., Gabbard, G. O., Gunderson, J., Kernberg, O., … Westen, D. (2010). Personality disorders in DSM-5. American Journal of Psychiatry, 167, 1026–1028.

Simms, L. J., Goldberg, L. R., Roberts, J. E., Watson, D., Welte, J., & Rotterman, J. H. (2011). Computerized adaptive assessment of personality disorder: Introducing the CAT-PD project. Journal of Personality Assessment, 93, 380–389.

Spitzer, R. L., First, M. B., Shedler, J., Westen, D., & Skodol, A. E. (2008). Clinical utility of five dimensional systems for personality diagnosis: A “consumer preference” study. Journal of Nervous and Mental Disease, 196, 356–374.

Sprock, J. (2003). Dimensional versus categorical classification of prototypic and nonprototypic cases of personality disorder. Journal of Clinical Psychology, 59, 991–1014.

Stepp, S., & Trull, T. J. (2007). Predictive validity of the five-factor model prototype scores for antisocial and borderline personality disorders. Personality and Mental Health, 1, 27–39.

Trull, T. J., & Widiger, T. A. (1997). Structured Interview for the Five-Factor Model of Personality (SIFFM): Professional Manual. Odessa, FL: Psychological Assessment Resources.

Trull, T. J., Widiger, T. A, Lynam, D. R., & Costa, P. T. (2003). Borderline personality disorder from the perspective of general personality functioning. Journal of Abnormal Psychology, 112, 193–202.

Warner, M. B., Morey, L. C., Finch, J. F., Gunderson, J. G., Skodol, A. E., Sanislow, C. A., … Grilo, C. M. (2004). The longitudinal relationship of personality traits and disorders. Journal of Abnormal Psychology, 113, 217–227.

Widiger, T. A. (2013). A postmortem and future look at the personality disorders in DSM-5. Personality Disorders: Theory, Research, and Treatment, 4, 382–387.

Widiger, T. A., & Costa, P. T. (Eds.) (2012). Personality Disorders and the Five-Factor Model of Personality (3rd ed.). Washington, DC: APA.

Widiger, T. A., Gore, W. L, Crego, C., Rojas, S. L., & Oltmanns, J. R. (2017). Five factor model and personality disorder. In T. A. Widiger (Ed.), The Oxford Handbook of the Five Factor Model (pp. 449–478). New York: Oxford University Press.

Widiger, T. A., & Lynam, D. R. (1998). Psychopathy and the five-factor model of personality. In T. Millon & E. Simonsen (Eds.), Psychopathy: Antisocial, Criminal, and Violent Behavior (pp. 171–187). New York, NY: Guilford Press.

Widiger, T. A., Lynam, D. R., Miller, J. D., & Oltmanns, T. F. (2012). Measures to assess maladaptive variants of the five factor model. Journal of Personality Assessment, 94, 450–455.

Widiger, T. A., & Trull, T. J. (1992). Personality and psychopathology: An application of the five‐factor model. Journal of Personality, 60, 363–393.

Widiger, T. A., Trull, T. J., Clarkin, J. F., Sanderson, C. J., & Costa, P. T. (1994). A description of the DSM-III-R and DSM-IV personality disorders with the five-factor model of personality. In P. T. Costa, Jr. & T. A. Widiger (Eds.), Personality Disorders and the Five-Factor Model of Personality (pp. 41–56). Washington, DC: APA.

Wright, A. G., Hopwood, C. J., & Zanarini, M. C. (2015). Associations between changes in normal personality traits and borderline personality disorder symptoms over 16 years. Personality Disorders: Theory, Research, and Treatment, 6, 1–11.

Zimmerman, M., Rothschild, L., & Chelminski, I. (2005). The prevalence of DSM-IV personality disorders in psychiatric outpatients. American Journal of Psychiatry, 162, 1911–1919.

¹These ratings were conducted on the original 37 traits put forth by the DSM-5 P & PD Work Group but can be translated to the official 25-trait model following Krueger et al. (2012).

²It is worth noting that the trait component of the AMPD generally accounts for substantially more unique variance in Section II PDs than does the impairment component (Criterion A), raising some questions about the necessity and utility of Criterion A (see Miller, Sleep, & Lynam, 2018 for a review).