Sage Journals: Discover world-class research

Abstract

For over 30 years, detailed personality-phenotype association profiles based on the NEO Personality Inventory (NEO PI) have been published, yet the phenotype profiles have been rarely linked across studies as has been done in brain imaging and genetics. Here, we apply a similar method to explore the similarities, differences, and healthfulness of various phenotypes regarding their NEO PI correlates. We systematically searched for profiles of NEO PI-phenotype correlations across various phenotype domains. Our initial dataset comprised 311 profiles from 90 papers. After standardising the metrics and meta-analysing similar profiles, the data was consolidated into 162 distinct phenotypes. Cluster analysis resulted in five broad cross-domain clusters, characterised by mental health strains, psychopathology dimensions, educational attainment, and health-related behaviours. Most clusters showed moderate deviation from an expert-rated ‘healthy’ personality profile. Furthermore, we replicated ‘personality correlations’, exploring the similarities between uncontrolled eating and addictive phenotypes, accounting for sample sizes of the profiles as well as the correlations among facets. Our findings demonstrate that comparing personality-phenotype profiles across studies offers novel insights and possibilities for theory-testing and exploratory comparisons.

Keywords

profile comparison addiction health NEO-PI

Introduction

Personality psychologists have long published personality trait correlation profiles for various phenotypes. For example, many phenotypes have replicable associations with the Big Five personality traits (Soto, 2019) that sometimes predict outcomes decades into the future (Beck & Johnson, 2022). Often, these associations tend to be rather similar, with many health and success indicators having negative associations with Neuroticism and positive correlations with Conscientiousness. However, the profiles become much more detailed and varied when the Big Five are broken into narrower traits. For instance, the NEO Personality Inventory-Revised (NEO-PI-R) that assesses the 30 facets of the Big Five allows characterising phenotype personality trait correlates in greater detail (Mõttus et al., 2017a). As a result, nuanced differences between seemingly similar profiles can arise, enabling a better understanding of behavioural differences and similarities among different phenotypes. That is, distinct facet combinations can correspond to diverse outcomes, even when corresponding to the same combination of Big Five domains. These nuanced differences might exhibit important practical implications, such as for matching people to occupations that suit their personality profile (Anni et al., 2025) or for uncovering mechanisms underlying specific pathologies through detailed trait patterns (e.g., Krueger & Markon, 2006; Soodla et al., 2025).

Although hundreds of NEO PI facet personality profiles are publicly available, few have attempted to systematically compare them. There have been studies comparing personality-phenotype correlation profiles when studying spousal similarity, development, psychopathology, person-environment fit (Furr, 2010), or for cross-cultural comparisons (Allik et al., 2017; Vainik, Kööts-Ausmees & Mõttus, 2019). However, broader personality profile comparisons offer great value in translating theory-driven constructs into empirically testable frameworks. Methods such as California Adult Q-set (CAQ) prototype-matching - used to assess how closely an individual’s personality matches theoretically derived prototype profiles (Letzring et al., 2005) - and profile agreement indices (McCrae et al., 2001) have demonstrated the power of matching personality patterns to theoretically or clinically derived prototypes. McCrae and colleagues (2001) demonstrated that profile agreement indices correlate with both self- and clinical-rated personality disorder symptoms, highlighting personality profiles’ utility in generating hypotheses and informing interventions.

Two recent studies have directly compared personality-phenotype profiles. Vainik and colleagues (2020) used profile correlations to investigate the long-held theoretical suggestion that obesity has a similar behavioural basis to addiction and psychiatric disorders, which was, up until then, mainly proposed based on a handful of selected observations (Michaud et al., 2017). They correlated NEO Personality Inventory Revised/3 (NEO PI-R/3) (McCrae et al., 2005) profiles previously assessed in different samples to investigate the behavioural overlap between addiction phenotypes (alcohol, gambling, opiate use, smoking), psychiatric phenotypes and obesity phenotype, providing a framework for quantifying the behavioural similarities across many phenotypes. They found that obesity was behaviourally related to addiction phenotypes, but to a lesser extent than uncontrolled eating or addictions themselves. This work demonstrated that addictive features explain only part of the behavioural repertoire leading to obesity, which in turn gives practical information for targeting interventions aimed at reducing obesity or overeating. Another example is Revelle and colleagues (2021), who compared empirical and personality profile correlations within a single large-scale dataset. Applying techniques analogous to genome-wide association studies (GWAS) identifying single-nucleotide polymorphisms (SNPs) related to specific outcomes of interest, they demonstrated how large-scale investigation of personality items can independently enhance predicting of real-world behavioural outcomes beyond aggregated scales. However, both Revelle and Vainik addressed limited sets of phenotypes. To date, there have been no attempts to atheoretically link a wider range of published personality profiles.

Conversely, other research fields have hugely benefitted from cross-study comparisons of phenotypes mapped to common reference panels. For instance, genetic correlations have become a standard method used to identify genetic overlaps between various complex traits and diseases mapped with genome-wide association studies in very different datasets (Bulik-Sullivan et al., 2015; epigraphdb.org; Liu et al., 2021). This has provided etiological insights that have helped to prioritise likely causal relationships between phenotypes (e.g., genetic correlation between anorexia nervosa and schizophrenia, autism and education) (Bulik-Sullivan et al., 2015). Similarly, in brain imaging, brain profiles of numerous psychiatric disorders and normal development have been compared to investigate the possible development and psychopathology (Markello et al., 2022; Cao et al., 2022; neurorynth.org; Yarkoni et al., 2011). This has revealed new disorder-specific and transdiagnostic patterns of neurodevelopmental disorders. These insights in genetics and brain imaging depended on systematically comparing data from multiple datasets through a standardised framework. There may be similar value in comparing personality profiles across a standardised reference panel – a shared questionnaire.

Here, we expand this approach to a broad range of personality trait correlation profiles of phenotypes measured in different datasets over decades of personality research. Based on a systematic literature search for published NEO PI personality profiles using atheoretical hierarchical cluster analysis, we mapped whether and how different phenotypes tend to cluster based on shared trait profiles. To understand how these personality profile clusters associate with a desirable (‘healthy’) personality, we further mapped the clusters to an expert-consensus healthy personality profile (Bleidorn et al., 2020).

In this process, we also advanced the statistical methodology for applying personality profile comparisons. Various metrics have been used to report personality profiles, which we aimed to standardise to a single and easily interpretable metric. Furr (2010) has emphasised that profile similarity comprises distinct components (elevation, scatter, shape) that should be considered for distinguished comparisons, yet concluding that Pearson correlation proves the best when primary interest lies in comparing shape or pattern of traits across profiles. Since we aimed to give a broad overview of the similarities in personality profile patterns, we followed this approach by standardising all the profiles to a correlation coefficient, as has been done in previous attempts at profile comparisons as well (Revelle et al., 2021; Vainik et al., 2020). However, the standard errors/p-values of the correlations need to consider the varying sample sizes of the datasets from which different profiles have been obtained. This has not been previously done in attempts at personality correlations. Moreover, the intercorrelations between NEO PI-R facets (Costa & McCrae, 1992) have not been accounted for either. Here, we sought to overcome these limitations and reanalyse data from Vainik et al. (2020) to compare the obesity and addictive phenotype profiles, adjusting for these statistical considerations.

In summary, we present a systematic scan of the most prominent phenotypes in the personality psychology literature across decades, exploratively clustering these profiles and comparing them to an expert-rated ‘healthy’ personality profile and replicate the profile comparisons of Vainik and colleagues (2020) using more robust statistical approaches.

Methods

Personality profiles of phenotypes

NEO Personality Inventory Revised (NEO PI-R) and its updated version, NEO PI-R 3 (Costa & McCrae, 1992; McCrae et al., 2005), are among the most widely used personality instruments that comprehensively measure the Five Factor Model of personality. It consists of 30 eight-item facet scales for each of the five personality factors (Xie & Cobb, 2020): Neuroticism (N) describes the general tendency to experience negative affect, Extraversion (E) describes the general tendency to stimulation and excitement, Openness (O) describes the tendency to explore novel ideas and intellectual curiosity, Agreeableness (A) is primarily a dimension of interpersonal tendencies, and Conscientiousness (C) describes the level of self-control and goal-directed behaviour. The NEO-PI personality profile of a phenotype presents the pattern of associations this phenotype (e.g., smoking status) has with these 30 facets. Personality has been reliably associated with many life outcomes (e.g., subjective well-being, religious beliefs, risky behaviour etc.) (Roberts et al., 2007; Soto, 2019), and detailed personality-phenotype association profiles using NEO PI profiles have been published for over 30 years, describing the way different phenotypes relate to the Big Five traits or more specialised facets of personality.

Search for personality profiles, inclusion and exclusion criteria

To create a dataset of phenotypes including facet-based information, we systematically collected NEO PI personality profiles from different papers across research domains using two databases (Google Scholar and Web of Knowledge in June 2018). From Web of Knowledge, we searched for papers citing the original manuals of NEO PI-R and NEO PI-3 and filtered based on the word “facet*” in topic. Similar approach could not be used in Google Scholar, as it led to too many results. Instead, we used a specific facet name from each of the five domains. Facet choice was based on the least amount of hits, which should lead to the most specific NEO PI-related results. For instance, “Anxiety” would have been a very unspecific search term. The narrower keywords for each facet that we ended up with were: “Angry hostility” (N2), “Gregariousness” (E2), “Aesthetics” (O2), “Tender-mindedness” (A6), and “Dutifulness” (C3). The personality profiles were then extracted to a pre-prepared sheet (available in OSF page). Both YW and UV coded the first 10 profiles, and any differences in coding were discussed between the authors. After that, YW coded the rest of the papers. As a result, our initial dataset included 311 profiles from 90 papers.

After preprocessing the data, some profiles were removed from further analysis for one of the following reasons: a) in case there were two similar distinct phenotype profiles, we chose the theoretically more informative and representative phenotype for further analysis and excluded the less informative ones (e.g., this was the case for different subscales assessing the same construct; then, to pick the most representative profile, we used correlations among these subscales in deciding which profile to preserve - for instance, out of several forgiveness scales and subscales providing very similar NEO PI profiles, we chose the profile of Forgiveness Likelihood Scale (Brose et al., 2005) as it summarised the profiles of the other subscales); b) profiles not based on empirical correlations with criteria (e.g. expert-rated profiles); c) International Personality Item Pool NEO (IPIP-NEO) profiles: while this questionnaire is similar to NEO PI-R, the facets have slightly different meanings, which may make the personality correlations noisier (Mõttus et al., 2006). After these exclusions, the dataset consisted of 188 phenotype profiles from 75 papers with sample sizes ranging from 21 to 14,848, with the combined sample size of our profile dataset being 99,655 across all phenotypes (see Table 1) (Anglim et al., 2018, Arnholt & Evans, 2021, Aguilar-Vafaie & Moghanloo, 2008, Bagby et al., 2007, Batey et al., 2010, Blickle, 1997, Blickle et al., 1998, Buczkowski et al., 2017, Cerasa et al., 2016, Costa et al., 2014, Dalpé et al., 2019, De Bolle et al., 2011, Del Re, 2013, DeShong et al., 2017, Dobewall et al., 2013, Dunkley & Kyparissis, 2008, Dunkley et al., 1997, Ekehammar & Akrami, 2007, Furnham, 2017; Furnham and Crump, 2014, 2015, 2016, 2017, Galler et al., 2013, Gaughan et al., 2009, Hayward et al., 2013, Helle et al., 2017, Hill et al., 1997, Horwood et al., 2015; Klimstra et al., 2011, 2013, Kwapil et al., 2002; Levallius et al., 2015, 2016, Lievens et al., 2009, Luchetti et al., 2018, Marušić et al., 2011, Madsen et al., 2006, Mann et al., 2017, Mullins-Sweatt et al., 2013, Mõttus et al., 2012, Mõttus et al., 2017c, Nestadt et al., 2008, O’Boyle et al., 2015, Oltmanns et al., 2016; Rector et al., 2002, 2012, Reid et al., 2011, Roose et al., 2012, Ross et al., 2004a, Ross et al., 2004b, Samuels et al., 2004, Schouwenburg & Lay, 1995, Shepherd et al., 2017, Shim et al., 2014, Siegler & Brummett, 2000, Saroglou & Muñoz-García, 2008; Sutin et al., 2010, 2013, Terracciano & Costa, 2004; Terracciano et al., 2009, 2013, 2014, Tybur et al., 2011, Vainik, Dagher, et al., 2019, Vassend et al., 2013, Vedel et al., 2015; Velting, 1999a, 1999b, Walton et al., 2018, Wood et al., 2009, Zhang et al., 2017). The dataset and accompanying codebook are publicly available on our project’s OSF webpage.

Table 1.

Personality profiles initial dataset characteristics

	N
Papers	75
Profiles	188
Total sample size	99,655
Minimum sample size	21
Maximum sample size	14,848
Age (M, SD)	28.5 (±6.5)
Female percentage	46%

Conversion of the statistics

Personality profiles of the phenotypes are either obtained by group comparison – for instance, comparing mean NEO PI scores for phenotype of interest group (e.g., smokers) to those of a control group (e.g., non-smokers) – or by associating the NEO PI scores with a continuous trait (e.g., body mass index, BMI). The papers in our initial dataset used various statistics to represent statistical associations (see Table 2). To compare them, we converted each profile to a standard metric, the Pearson correlation statistic r, using specific statistical conversion methods and R packages (e.g., BSDA, compute.es) for each statistic (see Table 2).

Table 2.

Conversion methods of the statistics

Original statistic	Profiles n	Method
M, SD/SE	47	Run t-test between phenotype and control profiles^a and then calculate r-statistic based on obtained t-statistic^b
β/b	9	Interpreted as an equal estimate to Pearson r-statistic
t	1	Calculated r-statistic based on t-statistic^a
d	5	Calculated r-statistic based on d-statistic^b
HR	2	Calculated d-statistic based on formula log(HR) × π6
HR	2	then converted d-statistic to r-statistic^b
OR	3	Calculated r-statistic from OR-statistic through d-statistic^c
r	118	—

^aFunction tsum.test from R package BSDA was used.

^bFunctions tes and des from R package compute.es was used. This package uses recommended formulas as described in The Handbook of Research Synthesis and Meta-Analysis (Cooper et al., 2019).

^cFunction oddsratio_tor from R package effectsize was used. This package implements formulas based on Borenstein et al. (2021). Converting among effect sizes. Introduction to meta-analysis, 45-49.

Nineteen papers had reported personality measures in M and SD/SE statistic, based on which we calculated r-statistics in one of the three ways. If the paper already included a control group (n = 10 papers) for the phenotype of interest, we calculated the r-statistic based on a t-test between the two profiles. This means that some profiles from the initial dataset were collapsed into one profile based on their comparison (e.g., initially two profiles of CEOs and CEOs working abroad were transformed into one profile ‘CEOs working abroad’ since the crucial aspect of this personality profile is to compare the personality differences in terms of working abroad or not). In other cases, the t-tests were run between the different phenotypes of interest presented in the original paper (e.g., former, current and never smokers all against each other) (n = 4 papers). In case the paper included only one phenotype information (n = 5), we calculated the t-statistic comparing the personality data to the standard norms or raw norm scores of NEO PI-R (Barto et al., 2011; Costa & McCrae, 1992). The effective sample size for all t-test-based studies was estimated according to the formula 2 x (N1 x N2)/(N1 + N2). We derived this formula from the property of additivity of variances of independent variables, and the fact that sample variance is inversely related to its sample size N. This approach avoids inflation of sample sizes, especially in the case where we contrasted phenotypes with NEO PI-R norms, where often a small case N1 was contrasted against the standard norm of N2 = 1000 control group.

Meta-analysis of similar profiles

As some profiles are more represented in the personality literature, our literature-wide search resulted in several profiles of the same phenotype from multiple papers: Alzheimer’s disease, obsessive-compulsive disorder, major depressive disorder, smokers (former, current, never), post-traumatic stress disorder, eating disorder (anorectic and non-anorectic), social phobia, (pathological and non-pathological) gamblers and personality disorder. In all cases, we meta-analytically merged them into one meta-profile, using the same methods as describe previously in Vainik et al. (2020).

Cluster analysis and cluster comparisons

After preprocessing the profiles’ dataset (see Figure 1), we were left with 162 unique phenotype profiles from 64 papers (which can be found in Supplemental Materials on OSF page). To explore potential underlying similarities of the most-represented phenotypes in the personality literature, we ran a hierarchical cluster analysis as an alternative to factor analysis, which requires the number of observations (in our case, the personality facets) to exceed the number of items (in our case, the phenotypes). For example, hierarchical cluster analysis has proven to be an effective method for grouping items (Revelle, 1979). It achieves this by iteratively reassessing the reliability of each cluster using alpha (the mean split-half correlation) and beta (the worst split-half correlation) coefficients as estimates of internal consistency and general factor saturation of the test. We ran the cluster analysis using the ‘iclust’ function from the “psych” R package (Revelle, 2023), implementing default alpha and beta values (3 and 1, respectively).

Figure 1.

Flowchart of the preprocessing of personality profiles dataset. Abbreviations: ALZ = Alzheimer’s disease; OCD = obsessive-compulsive disorder; GMB = gamblers; MDD = major depressive disorder; SMK = smokers; PTSD = post-traumatic stress disorder; PD = panic disorder; Ed = eating disorder; SP = social phobia

To better describe each cluster, we computed an average NEO PI personality profile of each cluster by calculating average values for each facet within the cluster (see Supplemental Table 3). Before averaging, we scaled each profile to have a mean of 0 and an SD of 1, to avoid profiles with a larger average effect size (e.g., personality-depressive r = 0.23) dominating other profiles with a smaller average effect size (e.g., personality-underweight r = 0.02). Since clusters also included profiles with negative loadings, we reversed these profiles before the averaging.

To compare the most-represented personality profile clusters to an expert-rated ‘healthy’ personality profile (Bleidorn et al., 2020), we used the facet-based personality data from their paper, calculated Cohen’s d effect size for each facet by comparing the data from the paper on normative vs expert-rated ‘healthy’ profile, and then correlated those with cluster averages in each of the personality facets.

Statistics of profile comparison

A further challenge is accurately calculating appropriate standard errors and p-values while also considering the NEO PI intercorrelations. The first issue is that correlations across 30 facets implies that sample size n = 30, whereas the original personality profile may be based on either a few dozen or thousands of participants. Analyses comparing profiles from different datasets have ignored this source of variance in accuracy. The second issue is that personality facets are not independent of one another, whereas a key assumption in estimating the uncertainty in correlations is independence of observations. For instance, facets of Neuroticism are often intercorrelated, as are facets within other Big Five domains. Neuroscience have countered that by developing the ‘spin test’: when comparing brain profiles that are spatially correlated, they are bootstrapped on a sphere that approximates the anatomical realities of the brain (Alexander-Bloch et al., 2018).

To overcome these problems, we compared the empirical correlations between profiles with those emerging from random datasets mimicking the correlational structure of NEO PI-R facets (we simulated 5000 of these datasets). At the first step, 30 random normally distributed variables (using the R base function ‘rnorm’) were created with a given sample size (corresponding to the sample size in the original study from which the profile was obtained) to represent the personality facets. At the second step, the transposed data matrix was pre-multiplied by the transpose of the upper triangulation factor of the Cholesky decomposition (obtained with the R base function ‘chol’) of the correlation matrix of the NEO PI-R facets (Costa & McCrae, 1992), and transposed the result. This results in a data matrix with a correlation structure approaching a given structure. The above-described method based on Cholesky decomposition was first described by Scheuer and Stoller (1962) and is nowadays routinely used for generating random data with a given correlation structure (e.g., in the R package rmvtnorm, Genz & Bretz, 2009). In a third step, correlations between each pair of profiles were compared against the distribution of correlations of 5000 randomly generated profiles, and a p-value was obtained as the proportion of times the simulated value (taken to represent the null distribution) was larger or equal to the empirical correlation.

We ran profile comparisons using these statistically more robust approached to replicate Vainik and Colleagues’ (2020) profile comparisons on the same exact data on obesity, uncontrolled eating, and addictive phenotypes, with education and gender as control phenotypes.

Data availability statement

The datasets generated during the current study, and the code for running all the analyses described here are publicly available on the project’s OSF page.

Results

Cluster analysis and cluster comparisons

After pre-processing and filtering the data, we were left with 162 distinct phenotypes which showed a high degree of clustering (Figure 2). The cluster analysis grouped them into 8 distinct clusters. Since hierarchical item-cluster analysis assigns each item to a single cluster, some output clusters were relatively small, containing only 4 to 7 phenotypes, while largest cluster encompassed 56 phenotypes.

Figure 2.

Personality profiles correlation matrix of all the standardised phenotypes (n = 162). Note: The phenotypes here are ordered according to their cluster loadings as in the Supplemental Table 1. A PDF version of this figure, optimised for readability, is available in the Supplemental Materials

The five largest clusters comprised 144 profiles, capturing the majority of the NEO PI personality profiles literature – 89% of our dataset. Interestingly, these biggest clusters can all be described by very high variation in Neuroticism levels (both within the domain and between the clusters), while having low variation in Agreeableness levels (see Figure 3). We hereby provide some examples of phenotypes included in each cluster, but a comprehensive list of all the phenotypes and their cluster loadings can be found in the supplements (see Supplemental Table 1 and Supplemental Table 2).

Figure 3.

Average facet-based personality profiles of the larger clusters. Note: Y-axis represents correlations that are standardised within each cluster to have a mean of 0 and an SD of 1 and then averaged across phenotypes in this cluster. See section Cluster analysis and cluster comparisons for more details. Number of phenotypes in each cluster: Cluster 1 = 56; Cluster 2 = 31; Cluster 3 = 22; Cluster 4 = 19; Cluster 5 = 16

The largest cluster (n = 56 profiles) included most of the psychopathological phenotypes (such as depression, OCD, panic disorder, PTSD, anxiety disorder, social phobia, agoraphobia, eating disorders, alcohol and substance abuse, self-injury and suicidal ideation) as well as other phenotypes related to negative affect, detachment, distrust, and passive-aggressiveness (e.g., chronic illness burden, child molesters etc.). We summarised these phenotypes as a ‘burdensome and distressed’ cluster as it also exhibits the highest values for Neuroticism, especially Anxiety, Depression and Vulnerability facets. The second-largest cluster (n = 31 profiles) included phenotypes such as egocentric and antagonistic factors of personality disorders, narcissism, grandiose, fearless dominance, manipulative and mischievous psychopathic factors, sexually transmitted disease history, argumentativeness and verbal aggressiveness, but also creativity and autonomy of personal wellbeing scale. Religious phenotypes, sexual disgust and forgiveness phenotypes were among the ones with negative loadings to this cluster. This cluster of mainly ‘egocentric impulsivity’ was characterised by high scores in Angry hostility, Impulsiveness, Assertiveness and Excitement seeking, but also in most of the Openness facets, while having the lowest Agreeableness levels. The third cluster (n = 22) represents Machiavellianism-like phenotypes as well as impulsive and callous and antisocial features of personality disorder, drug abusers, heavy drinkers, smokers, gamblers, metabolic syndrome patients, solitude depression and procrastination. It differs from the previous clusters mainly in terms of its antisociality and includes most of the addictive phenotypes. This ‘antisocial and addictive’ cluster scores highest in Excitement seeking and Impulsivity facets and has the lowest Conscientiousness scores. These three clusters comprised 67% of the phenotypes in our dataset.

Cluster 4 (n = 19 profiles) comprised phenotypes of compulsive and paranoid factors of personality disorder, medical school persisters’ GPA, pathogen disgust, multiple perfectionism scales, avoidance and anxiety behaviours in friendship, prejudice. This cluster of ‘distressed perfectionists’ was described by low levels of Warmth, Positive emotions, and high levels of Anxiety, Vulnerability, Dutifulness, and Deliberation. Cluster 5 (n = 16 profiles) differs most from other larger clusters as it can be described by very low levels of Neuroticism, but high levels on Extraversion and Conscientiousness. This cluster includes profiles of CEOs, education, high GPA, commitment-making identity dimensions, and purpose in life. Interestingly, it also includes argumentativeness phenotype. We summarised this as a cluster of ‘purposeful achievement-strivers’ with especially high average values of Assertiveness, Ideas, Competence, Achievement striving, and Self-discipline facets.

Small clusters comprised only four to seven phenotypes and are hence harder to generalise. Cluster 6 (n = 7 profiles) included a phenotype of high hypnotic susceptibility, but also health habits and health behaviour aspects of hypochondria, political studies students’ higher GPA, exploration in depth identity dimension, and gender differences. The average profile of this cluster was characterised by lower levels of Neuroticism, especially Angry hostility, and higher levels of Conscientiousness, especially Order and Self-discipline. This cluster had an extremely low Excitement seeking score and could be labelled ‘self-disciplined and health-aware’. Cluster 7 (n = 7 profiles) encompasses leisure-time activity categories of challenge skills, intrinsic rewards, social interactions, feel good and active engagement from one paper (Barnett, 2013), and interestingly, a BMI phenotype. So, these leisure-time phenotype measures differ enough from other phenotypes in the literature and comprise their own ‘leisure time activities’ cluster that is especially high in Positive emotions, Altruism, Impulsiveness, and Achievement striving, but rather low in Values facet. Cluster 8 (n = 4 profiles) includes positive loadings of odour identification scales, and negative loadings of cold pressor pain intensity and former smokers’ phenotype (Figure 4).

Figure 4.

Average facet-based personality profiles of the smaller clusters. Note: Y-axis represents correlations that are standardised within each cluster to have a mean of 0 and an SD of 1 and then averaged across phenotypes in this cluster. See section Cluster analysis and cluster comparisons for more details. Number of phenotypes in each cluster: Cluster 6 = 7; Cluster 7 = 5; Cluster 8 = 4

Correlations between clusters and a healthy personality profile

These clusters also correlated, with clusters 1 and 6 being the most intercorrelated. This further suggests that many phenomena that NEO PI has been used to characterise are rather similar at the personality profile level. The strongest was the negative correlation between the biggest cluster of ‘burdensome and distressed’ phenotypes to the cluster of ‘purposeful achievement-strivers’. This implies the opposing tendencies of the phenotypes included in those clusters that could be explained by their general desirability in society: while cluster 1 included most of the pathological phenotypes, cluster 5 included mainly educational attainment-related as well as purpose in life phenotypes. Conversely, three of the biggest clusters had moderate correlations with each other, indicating a common tendency for the representation of undesired personality patterns of the phenotypes involved.

Half of the clusters showed negative correlations with the expert-rated ‘healthy’ personality profile – three of these were moderately negative, while the others showed only a weak or negligible correlation (see Figure 5). Two clusters showed moderate positive correlations with the ‘healthy’ profile (clusters 5 and 6), with the ‘health-aware’ cluster having the highest correlation (r = 0.54).

Figure 5.

NEO PI personality literature cluster correlations with expert-rated ‘healthy’ personality profile. Note: No p-value is estimated here. Cluster names: Cluster 1 – Burdensome & distressed; Cluster 2 – Egocentric & impulsive; Cluster 3 – Antisocial & addictive; Cluster 4 – Distressed perfectionists; Cluster 5 – Purposeful & achievement-striving; Cluster 6 – Health-aware; Cluster 7 – Leisure; Cluster 8 – Misc

Re-analysis of Vainik 2020 facet matrix

We then reanalysed the facet profile correlation matrix of Vainik et al. (2020) on obesity and uncontrolled eating behavioural similarity to addictive phenotypes, using more robust statistical methods to take different sample sizes and personality facet intercorrelations into account. The main results replicated, so that obesity shows moderate and significant correlations with addictive phenotypes, but is not related to the profiles of phenotypes intended as controls (see Figure 6).

Figure 6.

Obesity and Uncontrolled Eating phenotype comparisons to addictive phenotypes. Note: Crosses stand for non-significant p-values obtained using the simulation method. Horizontal lines highlight the addictive phenotypes. Abbreviations: OB = obesity; UE = uncontrolled eating; ALC = alcohol; SMK = smoking; GMB = gambling; GMB. A = gambling with attention deficit hyperactivity disorder; OPI = opioid abuse; THC = cannabis; EDU = education; GEN = gender

Discussion

Here, we provided a literature-wide characterisation of NEO PI personality profiles. 89% of 162 distinct phenotypes clustered into five major groups, indicating notable consistency in how different phenotypes align with NEO PI personality patterns. The remaining 11% of the phenotypes formed three smaller clusters, highlighting meaningful variation that may reflect underrepresented or unique phenotypes in the literature. The clusters’ uniqueness was outlined by their differing associations with an expert-defined ‘healthy’ personality profile (Bleidorn et al., 2020), illustrating how certain traits systematically diverge from or align with socially and psychologically adaptive patterns. In addition, we introduced and tested a robust statistical method for comparing phenotypic personality profiles, successfully replicating earlier findings. This empirical mapping of the landscape of personality-phenotype associations, combined with our methodological framework, allows for new possibilities of both theory-driven hypothesis testing (e.g., as a given example of obesity and addictions) and exploratory insights (e.g., personality profiles as latent factors explaining a cluster of phenotypes).

Clusters of the most represented profiles in the literature

Almost 70% of the phenotypes in our dataset clustered into three large groups that diverged most from a desired ‘healthy’ personality profile (Bleidorn et al., 2020). The largest cluster includes a range of burdensome phenotypes described by detached, passive-aggressive tendencies – suggesting negative affectivity and vulnerability as core dimensions across many maladaptive behaviours. The second-largest cluster is characterised mainly by egocentric and impulsive excitement-seeking tendencies, while the third-largest cluster captured mainly addictive phenotypes with antisocial features. Therefore, most of the published phenotypes in the personality literature over the decades have mainly focused on either internalising or externalising problems, confirming the work by Krueger and Markon (2006). In contrast, a range of healthy or more normative phenotypes have been less represented in the literature. A more nuanced representation and understanding of health-related personality profiles could aid in developing and targeting personalised interventions for behaviour change, particularly in health or occupational settings (O’Connor, 2020; Wright & Jackson, 2023; Takahashi et al., 2013; Anni et al., 2024). A similar suggestion was made by Ahola et al. (2024), who identified five distinct personality profiles related to different patterns of physical activity and sedentary behaviour, highlighting the potential of tailoring health interventions – such as selecting appropriate behaviour change strategies – based on underlying personality patterns. Since facet- and item-level profiles allow for more nuanced differentiation between phenotypes (Mõttus et al., 2017b), future research should focus more on measuring and investigating a broader range of facet- and item-based healthier phenotypes as well (e.g., clustering of health behaviour or habit-based profiles, occupations, political views, values, volunteer- and leisure time tendencies, etc.).

Cluster relations to a healthy ‘desired’ personality profile

Only two clusters showed moderate positive correlations to the expert-defined healthy personality profile. Among the largest clusters, the only one similar to the ‘healthy’ profile was that of ‘purposeful and assertive achievement-strivers’ (cluster 5), which included phenotypes of educational attainment and CEOs. However, the cluster most strongly aligned with the healthy profile overall, was that of ‘self-disciplined and health-aware’ phenotypes (cluster 6) who excel in self-management while avoiding risky behaviour. This cluster included gender differences phenotype, obtained from a paper (de Bolle et al., 2015) representing the differences in personality scales based on males subtracted from females, so a female-positive profile. Based on this, females generally seem to display a healthier personality profile than male. Notably, this cluster also included the only health behaviour-related phenotypes in our dataset: health habits and health behaviours subscales of hypochondria assessed by the Illness Attitudes Scale (Cox et al., 2000). This may explain this clusters’ higher correlations with the ‘healthy’ personality, and highlights a need for more versatile health behaviour-related profiles. The strong negative correlation between clusters 1 and 5 points to an underlying axis of social functioning, underscoring how personality profiles may reflect broader dimensions of adaptation and desirability.

Profile comparisons with robust statistical framework

Applying Cholesky decomposition, we overcame the problems of varying sample sizes and intercorrelations of facets when comparing phenotypes from different samples. This more robust statistical method successfully replicated the core findings from Vainik et al. (2020), showing moderate but significant correlations between obesity and addictive phenotypes, but nonsignificant findings between obesity and control phenotypes. In the future, similar robust profile comparisons can be done to compare any of the phenotypes of interest from our profiles dataset to any other publicly available personality profiles from the literature, to validate or test an observational or theoretical idea. Further, the approach should also be applicable to item-level profiles, e.g. Revelle et al. (2021).

Limitations

Our dataset was constructed based on a systematic search for personality profiles in 2018, thus excluding more recent prominent profiles in the literature. However, since we aimed to give a broad overview of the most represented phenotypes studied in the field over many decades, this limitation is unlikely to significantly affect the main conclusions. Still, we encourage updating the dataset by renewed search for phenotype profiles, which is made possible by our open-access data, publicly available on the OSF platform, which allows anyone to rerun the analysis with an updated dataset.

Additionally, this work includes only personality profiles based on the NEO inventory, providing a somewhat skewed estimate of the broader landscape of personality profiles. We are further limited to profiles that previous researchers have chosen to study. However, as this work represents the first attempt in systematically mapping and comparing pre-published profiles, the proposed framework can be readily extended to other personality measures in future research.

We also acknowledge that the smaller clusters identified in our analysis are highly specific and hence not too generalisable to the literature. Since item-cluster analysis forces each phenotype to exactly one cluster, these phenotypes in small clusters could be interpreted as dataset outliers. On the other hand, this can give some suggestions on the more distinguishable profiles represented in the field and what type of phenotypes are underrepresented, and aid future research ideas.

Finally, our profile analysis focused on facets. While offering greater specificity than domains, facets are less detailed than nuances (Stewart et al., 2022). As nuance-based profiles become more widely available, our framework can be used to compare them as well, likely providing even more detailed insights.

Summary

We hereby expanded the efforts to compare personality profiles by compiling the most prominent publicly available NEO PI-based profiles into one dataset and converting them to the same metric. This framework provides a new tool for personality psychology research, allowing for theory-driven hypothesis testing (e.g., confirming shared underpinnings of vulnerable phenotypes) as well as explorative analyses of common latent pathways, as has been done in genetics and brain imaging research. By doing this, we also gave an overview of the most represented phenotypes in the last decades of personality psychology research, concluding that a high majority of the profiles are strongly opposite to the expert-rated ‘healthy’ profile. This highlights the personality literature’s focus on pathological and problematic phenotypes and the lack of more nuanced and versatile representations of healthier or normative phenotypes. We also introduced and tested a robust statistical approach for accurate profile comparisons that can be used in future research comparing phenotypes from any datasets. The methods and steps we outlined and tested here constitute a novel approach for applying large-scale personality correlations to any phenotypes of interest that have publicly available facet-based personality profiles.

Supplemental Material

Supplemental Material - Unity and Diversity of NEO-PI Personality Profiles of 162 Phenotypes

Supplemental Material for Unity and Diversity of NEO-PI Personality Profiles of 162 Phenotypes by Kerli Ilves, Yueh en Wang, Kenn Konstabel, Ants Adamson, Kersten Kõrge, Aire Raidvee, Alain Dagher, René Mõttus, Uku Vainik in Personality Science.

Footnotes

Author Note

Cristian Zanon was the handling editor.

Acknowledgements

Parts of this work were written at writing retreats organized by The University of Tartu Institute of Genomics and The Estonian Doctoral School for Natural and Agricultural Sciences (2021-2027.4.04.24-0003), co-funded by the European Union as well as at writing retreat by Estonian Centre of Excellence for Well-Being Sciences, funded by grant TK218 from the Estonian Ministry of Education and Research.

Author Contributions

Not applicable.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work has been funded by Estonian Research Council’s personal research funding return grants MOBTP94, start-up grants PSG656 and PSG759, and Estonian Research Council’s team grant PRG2190.

Declaration of Conflicting Interests

Not applicable.

ORCID iD

Kerli Ilves

Data Availability Statement

The data and analysis scripts used for this project are publicly accessible on the Open Science Framework page of this project ().

Supplemental Material

Supplemental material for this article is available online. Depending on the article type, these usually include a Transparency Checklist, a Transparent Peer Review File, and optional materials from the authors.

Notes

References

* Aguilar-Vafaie

M. E.

Moghanloo

(2008). Domain and facet personality correlates of religiosity among Iranian college students. Mental Health, Religion & Culture, 11(5), 461–483. https://doi.org/10.1080/13674670701539114

Ahola

Kekäläinen

Chastin

Rantalainen

Kinnunen

M.-L.

Pulkkinen

Kokko

(2024). Do personality profiles contribute to patterns of physical activity and sedentary behavior in adulthood? A prospective cohort study. International Journal of Behavioral Nutrition and Physical Activity, 21(1), 107. https://doi.org/10.1186/s12966-024-01662-y

Alexander-Bloch

A. F.

Shou

Liu

Satterthwaite

T. D.

Glahn

D. C.

Shinohara

R. T.

Vandekar

S. N.

Raznahan

(2018). On testing for spatial correspondence between maps of human brain structure and function. NeuroImage, 178, 540–551. https://doi.org/10.1016/j.neuroimage.2018.05.070

Allik

Church

A. T.

Ortiz

F. A.

Rossier

Hřebíčková

De Fruyt

Realo

McCrae

R. R.

(2017). Mean profiles of the NEO personality inventory. Journal of Cross-Cultural Psychology, 48(3), 402–420. https://doi.org/10.1177/0022022117692100

* Anglim

Bozic

Little

Lievens

(2018). Response distortion on personality tests in applicants: Comparing high-stakes to low-stakes medical settings. Advances in Health Sciences Education: Theory and Practice, 23(2), 311–321. https://doi.org/10.1007/s10459-017-9796-8

Anni

Vainik

Mõttus

(2025). Personality profiles of 263 occupations. Journal of Applied Psychology, 110(4), 481–511. https://doi.org/10.1037/apl0001249

Arnholt

Evans

(2021). BSDA: Basic statistics and data analysis (R package version 1.2.1). https://CRAN.R-project.org/package=BSDA

* Bagby

R. M.

Vachon

D. D.

Bulmash

E. L.

Toneatto

Quilty

L. C.

Costa

P. T.

(2007). Pathological gambling and the five-factor model of personality. Personality and Individual Differences, 43(4), 873–880. https://doi.org/10.1016/j.paid.2007.02.011

* Barnett

L. A.

(2013). What people want from their leisure: The contributions of personality facets in differentially predicting desired leisure outcomes. Journal of Leisure Research, 45(2), 150–191. https://doi.org/10.18666/jlr-2013-v45-i2-3010

10.

Barto

Chappelle

King

R. E.

Ree

M. J.

Teachout

M. S.

(2011). The NEO-PI-R as a premorbid baseline measure. Air Force School of Aerospace Medicine.

11.

* Batey

Chamorro-Premuzic

Furnham

(2010). Individual differences in ideational behavior: Can the big five and psychometric intelligence predict creativity scores? Creativity Research Journal, 22(1), 90–97. https://doi.org/10.1080/10400410903579627

12.

Beck

E. D.

Jackson

J. J.

(2022). A mega-analysis of personality prediction: Robustness and boundary conditions. Journal of Personality and Social Psychology, 122(3), 523–553. https://doi.org/10.1037/pspp0000386

13.

Bleidorn

Hopwood

C. J.

Ackerman

R. A.

Witt

E. A.

Kandler

Riemann

Samuel

D. B.

Donnellan

M. B.

(2020). The healthy personality from a basic trait perspective. Journal of Personality and Social Psychology, 118(6), 1207–1225. https://doi.org/10.1037/pspp0000231

14.

* Blickle

(1997). Argumentativeness and the facets of the big five. Psychological Reports, 81(3_suppl), 1379–1385. https://doi.org/10.2466/pr0.1997.81.3f.1379

15.

* Blickle

Habasch

Senft

(1998). Verbal aggressiveness: Conceptualization and measurement a decade later. Psychological Reports, 82(1), 287–298. https://doi.org/10.2466/pr0.1998.82.1.287

16.

Borenstein

Hedges

L. V.

Higgins

J. P.

Rothstein

H. R.

(2021). Introduction to meta-analysis. John Wiley & Sons.

17.

* Brose

L. A.

Rye

M. S.

Lutz-Zois

Ross

S. R.

(2005). Forgiveness and personality traits. Personality and Individual Differences, 39(1), 35–46. https://doi.org/10.1016/j.paid.2004.11.001

18.

* Buczkowski

Basinska

Ratajska

Lewandowska

Luszkiewicz

Sieminska

(2017). Smoking status and the five-factor model of personality: Results of a cross-sectional study conducted in Poland. International Journal of Environmental Research and Public Health, 14(2), 126. https://doi.org/10.3390/ijerph14020126

19.

Bulik-Sullivan

Finucane

H. K.

Anttila

Gusev

Day

F. R.

Loh

P. R.

ReproGen Consortium Psychiatric Genomics Consortium Genetic Consortium for Anorexia Nervosa of the Wellcome Trust Case Control Consortium 3 Duncan

Perry

J. R. B.

Patterson

Robinson

E. B.

Daly

M. J.

Price

A. L.

Neale

B. M.

(2015). An atlas of genetic correlations across human diseases and traits. Nature Genetics, 47(11), 1236–1241. https://doi.org/10.1038/ng.3406

20.

Cao

Cupertino

R. B.

Ottino-Gonzalez

Murphy

Pancholi

Juliano

Chaarani

Albaugh

Yuan

Schwab

Stafford

Goudriaan

A. E.

Hutchison

Luijten

Groefsema

Momenan

Schmaal

Sinha

ENIGMA Addiction Working Group . (2022). Cortical profiles of numerous psychiatric disorders and normal development share a common pattern. Molecular Psychiatry, 28(2), 1–12. https://doi.org/10.1038/s41380-022-01855-6

21.

* Cerasa

Lombardo

Tripodi

Stillitano

Sarica

Gramigna

Martino

Pullera

Tigani

De Carlo

Idone

Scaglione

Ziarelli

Vasta

Donzuso

Rizzo

Zucaro

D. L.

(2016). Five-factor personality traits in priests. Personality and Individual Differences, 95(2016), 89–94. https://doi.org/10.1016/j.paid.2016.02.040

22.

Cooper

Hedges

L. V.

Valentine

J. C.

(Eds.), (2019). The handbook of research synthesis and meta-analysis. Russell Sage Foundation.

23.

Costa

P. T.

McCrae

R. R.

(1992). Neo personality inventory-revised (NEO PI-R): Psychological Assessment Resources. https://personality-project.org/r/psych/help/neo.htm

24.

* Costa

P. T.

Weiss

Duberstein

P. R.

Friedman

Siegler

I. C.

(2014). Personality facets and all-cause mortality among medicare patients aged 66 to 102 years: A Follow-on study of Weiss and Costa (2005). Psychosomatic Medicine, 76(5), 370–378. https://doi.org/10.1097/PSY.0000000000000070

25.

* Cox

B. J.

Borger

S. C.

Asmundson

G. J. G.

Taylor

(2000). Dimensions of hypochondriasis and the five-factor model of personality. Personality and Individual Differences, 29(1), 99–108. https://doi.org/10.1016/S0191-8869(99)00180-4

26.

* Dalpé

Demers

Verner-Filion

Vallerand

R. J.

(2019). From personality to passion: The role of the big five factors. Personality and Individual Differences, 138(2019), 280–285. https://doi.org/10.1016/j.paid.2018.10.021

27.

* De Bolle

De Clercq

Pham-Scottez

Mels

Rolland

J.-P.

Guelfi

J. D.

Braet

De Fruyt

(2011). Personality pathology comorbidity in adult females with eating disorders. Journal of Health Psychology, 16(2), 303–313. https://doi.org/10.1177/1359105310374780

28.

* De Bolle

De Fruyt

McCrae

R. R.

Löckenhoff

C. E.

Costa

P. T.

Aguilar-Vafaie

M. E.

Ahn

Alcalay

Allik

Avdeyeva

T. V.

Bratko

Brunner-Sciarra

Cain

T. R.

Chan

Chittcharat

Crawford

J. T.

Fehr

Ficková

Terracciano

(2015). The emergence of sex differences in personality traits in early adolescence: A cross-sectional, cross-cultural study. Journal of Personality and Social Psychology, 108(1), 171–185. https://doi.org/10.1037/a0038497

29.

Del Re

A. C.

(2013). Compute.es: Compute effect sizes (R package version 0.2-2). https://cran.r-project.org/package=compute.es

30.

* DeShong

H. L.

Helle

A. C.

Lengel

G. J.

Meyer

Mullins-Sweatt

S. N.

(2017). Facets of the dark triad: Utilizing the five-factor model to describe machiavellianism. Personality and Individual Differences, 105(2017), 218–223. https://doi.org/10.1016/j.paid.2016.09.053

31.

* Dobewall

Realo

Allik

Esko

Metspalu

(2013). Self-other agreement in happiness and life-satisfaction: The role of personality traits. Social Indicators Research, 114(2), 479–492. https://doi.org/10.1007/s11205-012-0157-y

32.

* Dunkley

D. M.

Blankstein

K. R.

Flett

G. L.

(1997). Specific cognitive-personality vulnerability styles in depression and the five-factor model of personality. Personality and Individual Differences, 23(6), 1041–1053. https://doi.org/10.1016/S0191-8869(97)00079-2

33.

* Dunkley

D. M.

Kyparissis

(2008). What is DAS self-critical perfectionism really measuring? Relations with the five-factor model of personality and depressive symptoms. Personality and Individual Differences, 44(6), 1295–1305. https://doi.org/10.1016/j.paid.2007.11.022

34.

* Ekehammar

Akrami

(2007). Personality and prejudice: From big five personality factors to facets. Journal of Personality, 75(5), 899–926. https://doi.org/10.1111/j.1467-6494.2007.00460.x

35.

* Furnham

(2017). Personality differences in managers who have, and have not, worked abroad. European Management Journal, 35(1), 39–45. https://doi.org/10.1016/j.emj.2016.10.002

36.

* Furnham

Crump

(2014). A bright side, facet analysis of schizotypal personality disorder: The relationship between the HDS imaginative factor, the NEO-PI-R personality trait facets in a large adult sample. Thinking Skills and Creativity, 11(2014), 42–47. https://doi.org/10.1016/j.tsc.2013.10.001

37.

* Furnham

Crump

(2015). A big five facet analysis of a paranoid personality disorder: The validity of the HDS sceptical scale of subclinical paranoia. Journal of Individual Differences, 36(4), 199–204. https://doi.org/10.1027/1614-0001/a000174

38.

* Furnham

Crump

(2016). A big five facet analysis of a psychopath: The validity of the HDS mischievous scale of sub-clinical psychopathy. Scandinavian Journal of Psychology, 57(2), 117–121. https://doi.org/10.1111/sjop.12260

39.

* Furnham

Crump

(2017). Personality correlates of passive-aggressiveness: A NEO-PI-R domain and facet analysis of the HDS leisurely scale. Journal of Mental Health, 26(6), 496–501. https://doi.org/10.3109/09638237.2016.1167853

40.

Furr

R. M.

(2010). The double-entry intraclass correlation as an index of profile similarity: Meaning, limitations, and alternatives. Journal of Personality Assessment, 92(1), 1–15. https://doi.org/10.1080/00223890903379134

41.

* Galler

J. R.

Bryce

C. P.

Zichlin

M. L.

Waber

D. P.

Exner

Fitzmaurice

G. M.

Costa

P. T.

(2013). Malnutrition in the first year of life and personality at age 40. The Journal of Child Psychology and Psychiatry and Allied Disciplines, 54(8), 911–919. https://doi.org/10.1111/jcpp.12066

42.

* Gaughan

E. T.

Miller

J. D.

Pryor

L. R.

Lynam

D. R.

(2009). Comparing two alternative measures of general personality in the assessment of psychopathy: A test of the NEO PI-R and the MPQ. Journal of Personality, 77(4), 965–996. https://doi.org/10.1111/j.1467-6494.2009.00571.x

43.

Genz

Bretz

(2009). Computation of multivariate normal and t probabilities, series lecture notes in statistics. Springer-Verlag.

44.

* Hayward

R. D.

Taylor

W. D.

Smoski

M. J.

Steffens

D. C.

Payne

M. E.

(2013). Association of five-factor model personality domains and facets with presence, onset, and treatment outcomes of major depression in older adults. American Journal of Geriatric Psychiatry: Official Journal of the American Association for Geriatric Psychiatry, 21(1), 88–96. https://doi.org/10.1016/j.jagp.2012.11.012

45.

* Helle

A. C.

Trull

T. J.

Widiger

T. A.

Mullins-Sweatt

S. N.

(2017). Utilizing interview and self-report assessment of the five-factor model to examine convergence with the alternative model for personality disorders. Personality disorders, 8(3), 247–254. https://doi.org/10.1037/per0000174

46.

* Hill

R. W.

McIntire

Bacharach

V. R.

(1997). Perfectionism and the big five factors. Journal of Social Behavior & Personality, 12(1), 257.

47.

* Horwood

Anglim

Tooley

(2015). Type D personality and the five-factor model: A facet-level analysis. Personality and Individual Differences, 83(2015), 50–54. https://doi.org/10.1016/j.paid.2015.03.041

48.

* Klimstra

T. A.

Luyckx

Goossens

Teppers

De Fruyt

(2013). Associations of identity dimensions with big five personality domains and facets. European Journal of Personality, 27(3), 213–221. https://doi.org/10.1002/per.1853

49.

* Klimstra

T. A.

Luyckx

Teppers

Goossens

Fruyt

F. D.

(2011). Congruence between adolescent personality types based on the big five domains and the 30 NEO-PI-3 personality facets. Journal of Research in Personality, 45(5), 513–517. https://doi.org/10.1016/j.jrp.2011.07.004

50.

Krueger

R. F.

Markon

K. E.

(2006). Reinterpreting comorbidity: A model-based approach to understanding and classifying psychopathology. Annual Review of Clinical Psychology, 2(1), 111–133. https://doi.org/10.1146/annurev.clinpsy.2.022305.095213

51.

* Kwapil

T. R.

Wrobel

M. J.

Pope

C. A.

(2002). The five-factor personality structure of dissociative experiences. Personality and Individual Differences, 32(3), 431–443. https://doi.org/10.1016/S0191-8869(01)00035-6

52.

Letzring

T. D.

Block

Funder

D. C.

(2005). Ego-control and ego-resiliency: Generalization of self-report scales based on personality descriptions from acquaintances, clinicians, and the self. Journal of Research in Personality, 39(4), 395–422. https://doi.org/10.1016/j.jrp.2004.06.003

53.

* Levallius

Clinton

Bäckström

Norring

(2015). Who do you think you are? - Personality in eating disordered patients. Journal of Eating Disorders, 3(1), 3. https://doi.org/10.1186/s40337-015-0042-6

54.

* Levallius

Roberts

B. W.

Clinton

Norring

(2016). Take charge: Personality as predictor of recovery from eating disorder. Psychiatry Research, 246, 447–452. https://doi.org/10.1016/j.psychres.2016.08.064

55.

* Lievens

Ones

D. S.

Dilchert

(2009). Personality scale validities increase throughout medical school. Journal of Applied Psychology, 94(6), 1514–1535. https://doi.org/10.1037/a0016137

56.

Liu

Elsworth

Erola

Haberland

Hemani

Lyon

Zheng

Lloyd

Vabistsevits

Gaunt

T. R.

(2021). EpiGraphDB: A database and data mining platform for health data science. Bioinformatics, 37(9), 1304–1311. https://doi.org/10.1093/bioinformatics/btaa961

57.

* Luchetti

Sutin

A. R.

Delitala

Stephan

Fiorillo

Marongiu

Masala

Schlessinger

Terracciano

(2018). Personality traits and facets linked with self-reported alcohol consumption and biomarkers of liver health. Addictive Behaviors, 82, 135–141. https://doi.org/10.1016/j.addbeh.2018.02.034

58.

* Madsen

Parsons

Grubin

(2006). The relationship between the five-factor model and DSM personality disorder in a sample of child molesters. Personality and Individual Differences, 40(2), 227–236. https://doi.org/10.1016/j.paid.2005.06.023

59.

Mann

F. D.

Engelhardt

Briley

D. A.

Grotzinger

A. D.

Patterson

M. W.

Tackett

J. L.

Strathan

D. B.

Heath

Lynskey

Slutske

Martin

N. G.

Tucker-Drob

E. M.

Harden

K. P.

(2017). Sensation seeking and impulsive traits as personality endophenotypes for antisocial behavior: Evidence from two independent samples. Personality and Individual Differences, 105(2017), 30–39. https://doi.org/10.1016/j.paid.2016.09.018

60.

Markello

R. D.

Hansen

J. Y.

Liu

Z. Q.

Bazinet

Shafiei

Suárez

L. E.

Blostein

Seidlitz

Baillet

Satterthwaite

T. D.

Chakravarty

M. M.

Raznahan

Misic

(2022). Neuromaps: Structural and functional interpretation of brain maps. Nature Methods, 19(11), 1472–1479. https://doi.org/10.1038/s41592-022-01625-w

61.

* Marušić

Kamenov

Ž.

Jelić

(2011). Personality and attachment to friends. Društvena Istraživanja, 20(114), 1119–1137. https://doi.org/10.5559/di.20.4.10

62.

McCrae

R. R.

Costa

P. T.

Jr. Martin

T. A.

(2005). The NEO–PI–3: A more readable revised NEO personality inventory. Journal of Personality Assessment, 84(3), 261–270. https://doi.org/10.1207/s15327752jpa8403_05

63.

McCrae, R. R., Yang, J., Costa, Jr, P. T., Dai, X., Yao, S., Cai, T., & Gao, B. (2001). Personality profiles and the prediction of categorical personality disorders. Journal of Personality, 69(2), 155–174. https://doi.org/10.1111/1467-6494.00140

64.

Michaud

Vainik

Garcia-Garcia

Dagher

(2017). Overlapping neural endophenotypes in addiction and obesity. Frontiers in Endocrinology, 8, 127. https://doi.org/10.3389/fendo.2017.00127

65.

Mõttus

Bates

T. C.

Condon

D. M.

Mroczek

Revelle

(2017b). Your personality data can do more: Items provide leverage for explaining the variance and co-variance of life outcomes. psyarxiv.com/4q9gv.

66.

Mõttus

Kandler

Bleidorn

Riemann

McCrae

R. R.

(2017a). Personality traits below facets: The consensual validity, longitudinal stability, heritability, and utility of personality nuances. Journal of Personality and Social Psychology, 112(3), 474–490. https://doi.org/10.1037/pspp0000100

67.

Mõttus

Pullmann

Allik

(2006). Toward more readable big five personality inventories. European Journal of Psychological Assessment, 22(3), 149–157. https://doi.org/10.1027/1015-5759.22.3.149

68.

* Mõttus

Realo

Allik

Esko

Metspalu

(2012). History of the diagnosis of a sexually transmitted disease is linked to normal variation in personality traits. The Journal of Sexual Medicine, 9(11), 2861–2867. https://doi.org/10.1111/j.1743-6109.2012.02891.x

69.

* Mõttus

Realo

Vainik

Allik

Esko

(2017c). Educational attainment and personality are genetically intertwined. Psychological Science, 28(11), 1631–1639. https://doi.org/10.1177/0956797617719083

70.

* Mullins-Sweatt

S. N.

Lengel

G. J.

Grant

D. M.

(2013). Non-suicidal self-injury: The contribution of general personality functioning. Personality and Mental Health, 7(1), 56–68. https://doi.org/10.1002/pmh.1211

71.

* Nestadt

Costa

P. T.

Hsu

F.-C.

Samuels

Bienvenu

O. J.

Eaton

W. W.

(2008). The relationship between the five-factor model and latent diagnostic and statistical manual of mental disorders, fourth edition personality disorder dimensions. Comprehensive Psychiatry, 49(1), 98–105. https://doi.org/10.1016/j.comppsych.2007.05.015

72.

* 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(6), 644–664. https://doi.org/10.1111/jopy.12126

73.

O’Connor

D. B.

(2020). The future of health behaviour change interventions: Opportunities for open science and personality research. Health Psychology Review, 14(1), 176–181. https://doi.org/10.1080/17437199.2019.1707107

74.

* Oltmanns

J. R.

DeShong

H. L.

Sanders

C. A.

Kurtz

J. E.

(2016). Validation of the five-factor model personality disorder prototype counts with the personality assessment inventory. Journal of Psychopathology and Behavioral Assessment, 38(4), 590–599. https://doi.org/10.1007/s10862-016-9549-7

75.

* Rector

N. A.

Bagby

R. M.

Huta

Ayearst

L. E.

(2012). Examination of the trait facets of the five-factor model in discriminating specific mood and anxiety disorders. Psychiatry Research, 199(2), 131–139. https://doi.org/10.1016/j.psychres.2012.04.027

76.

* Rector

N. A.

Hood

Richter

M. A.

Michael Bagby

(2002). Obsessive-compulsive disorder and the five-factor model of personality: Distinction and overlap with major depressive disorder. Behaviour Research and Therapy, 40(10), 1205–1219. https://doi.org/10.1016/S0005-7967(02)00024-4

77.

* Reid

R. C.

D. S.

Lopez

Collard

Parhami

Karim

Fong

(2011). Exploring facets of personality and escapism in pathological gamblers. Journal of Social Work Practice in the Addictions, 11(1), 60–74. https://doi.org/10.1080/1533256X.2011.547071

78.

Revelle

(1979). Hierarchical cluster analysis and the internal structure of tests. Multivariate Behavioral Research, 14(1), 57–74. https://doi.org/10.1207/s15327906mbr1401_4

79.

Revelle

(2023). psych: Procedures for personality and psychological research (Version 2.3.3.) [R package]. Northwestern University. https://CRAN.R-project.org/package=psych

80.

Revelle

Dworak

E. M.

Condon

D. M.

(2021). Exploring the persome: The power of the item in understanding personality structure. Personality and Individual Differences, 169(2021), 109905. https://doi.org/10.1016/j.paid.2020.109905

81.

Roberts

B. W.

Kuncel

N. R.

Shiner

Caspi

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: A Journal of the Association for Psychological Science, 2(4), 313–345. https://doi.org/10.1111/j.1745-6916.2007.00047.x

82.

* Roose

Bijttebier

Claes

Lilienfeld

S. O.

De Fruyt

Decuyper

(2012). Psychopathic traits in adolescence and the five factor model of personality. Journal of Psychopathology and Behavioral Assessment, 34(1), 84–93. https://doi.org/10.1007/s10862-011-9243-8

83.

* Ross

S. R.

Kendall

A. C.

Matters

K. G.

Mark

Rye

M. S. R.

Wrobel

T. A.

(2004a). A personological examination of Self- and other-forgiveness in the five factor model. Journal of Personality Assessment, 82(2), 207–214. https://doi.org/10.1207/s15327752jpa8202_8

84.

* Ross

S. R.

Lutz

C. J.

Bailley

S. E.

(2004b). Psychopathy and the five factor model in a noninstitutionalized sample: A domain and facet level analysis. Journal of Psychopathology and Behavioral Assessment, 26(4), 213–223. https://doi.org/10.1023/B:JOBA.0000045337.48535.a5

85.

* Samuels

Bienvenu

O. J.

Cullen

Costa

P. T.

Eaton

W. W.

Nestadt

(2004). Personality dimensions and criminal arrest. Comprehensive Psychiatry, 45(4), 275–280. https://doi.org/10.1016/j.comppsych.2004.03.013

86.

* Saroglou

Muñoz-García

(2008). Individual differences in religion and spirituality: An issue of personality traits and/or values. Journal for the Scientific Study of Religion, 47(1), 83–101. https://doi.org/10.1111/j.1468-5906.2008.00393.x

87.

Scheuer

E. M.

Stoller

D. S.

(1962). On the generation of normal random vectors. Technometrics, 4(2), 278–281. https://doi.org/10.1080/00401706.1962.10490011

88.

* Schouwenburg

H. C.

Lay

C. H.

(1995). Trait procrastination and the Big-five factors of personality. Personality and Individual Differences, 18(4), 481–490. https://doi.org/10.1016/0191-8869(94)00176-S

89.

* Shepherd

Hautus

M. J.

Urale

P. W. B.

(2017). Personality and perceptions of common odors. Chemosensory Perception, 10(1–2), 23–30. https://doi.org/10.1007/s12078-016-9220-4

90.

* Shim

J.-Y.

Lee

Sung

Y.-A.

Kim

H.-N.

Kim

H.-L.

(2014). Association between extraversion personality and abnormal glucose regulation in young Korean women. Acta Diabetologica, 51(3), 421–427. https://doi.org/10.1007/s00592-013-0529-z

91.

* Siegler

I. C.

Brummett

B. H.

(2000). Associations among NEO personality assessments and well-being at midlife: Facet-level analyses. Psychology and Aging, 15(4), 710–714. https://doi.org/10.1037/0882-7974.15.4.710

92.

Soodla

H. L.

Lehto

Kõiv

Vainik

Akkermann

Mõttus

(2025). Assessing the overlap of personality traits and internalizing psychopathology using multi-informant data: Two sides of the same coin? Journal of Psychopathology and Clinical Science, 134(4), 400–413. https://doi.org/10.1037/abn0000967

93.

Soto

C. J.

(2019). How replicable are links between personality traits and consequential life outcomes? The life outcomes of personality replication project. Psychological Science, 30(5), 711–727. https://doi.org/10.1177/0956797619831612

94.

Stewart

R. D.

Mõttus

Seeboth

Soto

C. J.

Johnson

(2022). The finer details? The predictability of life outcomes from big five domains, facets, and nuances. Journal of Personality, 90(2), 167–182. https://doi.org/10.1111/jopy.12660

95.

* Sutin

A. R.

Costa

P. T.

Uda

Ferrucci

Schlessinger

Terracciano

(2010). Personality and metabolic syndrome. Age, 32(4), 513–519. https://doi.org/10.1007/s11357-010-9153-9

96.

* Sutin

A. R.

Zonderman

A. B.

Ferrucci

Terracciano

(2013). Personality traits and chronic disease: Implications for adult personality development. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 68(6), 912–920. https://doi.org/10.1093/geronb/gbt036

97.

Takahashi

Edmonds

G. W.

Jackson

J. J.

Roberts

B. W.

(2013). Longitudinal correlated changes in conscientiousness, preventative health-related behaviors, and self-perceived physical health. Journal of Personality, 81(4), 417–427. https://doi.org/10.1111/jopy.12007

98.

* Terracciano

Costa

P. T.

(2004). Smoking and the five-factor model of personality. Addiction, 99(4), 472–481. https://doi.org/10.1111/j.1360-0443.2004.00687.x

99.

* Terracciano

Iacono

O’Brien

R. J.

Troncoso

J. C.

Sutin

A. R.

Ferrucci

Zonderman

A. B.

Resnick

S. M.

(2013). Personality and resilience to Alzheimer’s disease neuropathology: A prospective autopsy study. Neurobiology of Aging, 34(4), 1045–1050. https://doi.org/10.1016/j.neurobiolaging.2012.08.008

100.

* Terracciano

Sutin

A. R.

O’Brien

R. J.

Ferrucci

Zonderman

A. B.

Resnick

S. M.

(2014). Personality and risk of Alzheimer’s disease: New data and meta-analysis. Alzheimer's and Dementia: The Journal of the Alzheimer's Association, 10(2), 179–186. https://doi.org/10.1016/j.jalz.2013.03.002

101.

* Terracciano

Sutin

A. R.

McCrae

R. R.

Deiana

Ferrucci

Schlessinger

Uda

Costa

P. T.

(2009). Facets of personality linked to underweight and overweight. Psychosomatic Medicine, 71(6), 682–689. https://doi.org/10.1097/PSY.0b013e3181a2925b

102.

* Tybur

J. M.

Bryan

A. D.

Lieberman

Caldwell Hooper

A. E.

Merriman

L. A.

(2011). Sex differences and sex similarities in disgust sensitivity. Personality and Individual Differences, 51(3), 343–348. https://doi.org/10.1016/j.paid.2011.04.003

103.

Vainik

Kööts-Ausmees

Mõttus

(2019b). Personality loosely mirrors the geography of the world. In Realo

(Ed.), In Praise of an inquisitive mind: A festschrift in honor of Jüri Allik occasion of his 70th birthday. University of Tartu Press.

104.

Vainik

Misic

Zeighami

Michaud

Mõttus

Dagher

(2020). Obesity has limited behavioural overlap with addiction and psychiatric phenotypes. Nature Human Behaviour, 4(1), 27–35. https://doi.org/10.1038/s41562-019-0752-x

105.

* Vainik

Dagher

Realo

Colodro-Conde

Mortensen

E. L.

Jang

Juko

Kandler

Sørensen

T. I. A.

Mõttus

(2019). Personality-obesity associations are driven by narrow traits: A meta-analysis. Obesity Reviews: An Official Journal of the International Association for the Study of Obesity, 20(8), 1121–1131. https://doi.org/10.1111/obr.12856

106.

* Vassend

Røysamb

Nielsen

C. S.

(2013). Five-factor personality traits and pain sensitivity: A twin study. Pain, 154(5), 722–728. https://doi.org/10.1016/j.pain.2013.01.010

107.

* Vedel

Thomsen

D. K.

Larsen

(2015). Personality, academic majors and performance: Revealing complex patterns. Personality and Individual Differences, 85(2015), 69–76. https://doi.org/10.1016/j.paid.2015.04.030

108.

* Velting

D. M.

(1999a). Personality and negative expectancies: Trait structure of the beck hopelessness scale. Personality and Individual Differences, 26(5), 913–921. https://doi.org/10.1016/S0191-8869(98)00194-9

109.

* Velting

D. M.

(1999b). Suicidal ideation and the five-factor model of personality. Personality and Individual Differences, 27(5), 943–952. https://doi.org/10.1016/S0191-8869(99)00046-X

110.

* Walton

K. E.

Pantoja

McDermut

(2018). Associations between lower order facets of personality and dimensions of mental disorder. Journal of Psychopathology and Behavioral Assessment, 40(3), 465–475. https://doi.org/10.1007/s10862-017-9633-7

111.

* Wood

A. M.

Joseph

Maltby

(2009). Gratitude predicts psychological well-being above the big five facets. Personality and Individual Differences, 46(4), 443–447. https://doi.org/10.1016/j.paid.2008.11.012

112.

Wright

A. J.

Jackson

J. J.

(2023). Do changes in personality predict life outcomes? Journal of Personality and Social Psychology, 125(6), 1495–1518. https://doi.org/10.1037/pspp0000472

113.

Xie

Cobb

C. L.

(2020). Revised NEO personality inventory (NEO-PI-R) (335). The Wiley Encyclopedia of Personality and Individual Differences: Measurement and Assessment.

114.

Yarkoni

Poldrack

R. A.

Nichols

T. E.

Van Essen

D. C.

Wager

T. D.

(2011). Large-scale automated synthesis of human functional neuroimaging data. Nature Methods, 8(8), 665–670. https://doi.org/10.1038/nmeth.1635

115.

* Zhang

Wang

Shen

Zhang

Wang

Chen

Wang

(2017). Relationship between hypnosis and personality trait in participants with high or low hypnotic susceptibility. Neuropsychiatric Disease and Treatment, 13, 1007–1012. https://doi.org/10.2147/NDT.S134930

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.52 MB