History of Neuroticism: From Eysenck to the Big Five
Author: Ukrainian Psychological Hub · Published: September 30, 2026 · Editorial Policy
The modern history of neuroticism is not a straight line from an old psychiatric diagnosis to a Big Five score. It is the history of several research traditions converging on a broad dimension of personality. Hans J. Eysenck’s work in the 1940s made neuroticism a measurable, continuous dimension of individual differences. A separate lexical tradition asked which broad personality dimensions recur when people describe one another in ordinary language. During the 1980s and 1990s, these lines of research increasingly converged, and neuroticism became one of the standard domains of the Big Five and Five-Factor Model.
That distinction matters because the word carries historical baggage. Contemporary personality science uses neuroticism as a dimensional trait construct, not as a diagnosis. It is also not interchangeable with the historical clinical category neurosis, with the everyday label neurotic, with current anxiety symptoms, or with an anxiety disorder. For the modern construct itself, see Neuroticism: What It Is in the Big Five Personality Model. For the historical clinical distinction, see Neurosis vs Neuroticism: Historical Disorder Concept vs Big Five Personality Trait.
The central historical point is this: Eysenck was crucial to the modern psychometric development of neuroticism, but he did not invent the Big Five, and Big Five neuroticism is not simply the Eysenck scale transplanted unchanged into a five-trait model. The Big Five emerged from a broader body of factor-analytic, lexical, questionnaire, and cross-observer research. Understanding that history explains why today’s instruments can use the labels Neuroticism, Emotional Stability, or Negative Emotionality for closely related territory while still differing in items, facets, scoring, and interpretation.
History of Neuroticism in One Timeline
In 1936, Gordon Allport and Henry Odbert published a large psycho-lexical study of English trait names. Their project did not establish modern neuroticism by itself, but it helped create the lexical research program: if socially important personality differences become encoded in language, then systematic analysis of trait words may reveal broad dimensions of personality. Allport and Odbert’s study became a major starting point for later lexical taxonomies.
In 1943, Raymond Cattell reduced and organized trait terms into clusters and used factor-analytic methods to search for a more manageable structure. His study of personality descriptors was part of the route that eventually led to the recurring five-factor pattern. This lexical route developed independently of Eysenck’s early two-dimensional theory.
In 1947, Eysenck published Dimensions of Personality. The book treated neuroticism and introversion–extraversion as broad dimensions rather than simple diagnostic types. Eysenck’s analyses were rooted in a historically specific clinical research setting: the book describes an original population of 1,000 male service patients, reduced to 700 after exclusions, and factor-analytic work on ratings, symptoms, and personality characteristics. The 1947 book is the landmark source for Eysenck’s early dimensional formulation.
During the 1950s and 1960s, Eysenck translated the theory into questionnaires. The measurement line ran through the Maudsley instruments, the Eysenck Personality Inventory, and later the Eysenck Personality Questionnaire. A modern historical analysis of the EPI summarizes this sequence and shows how Eysenck’s Neuroticism scale remained a broad factor while its item content reflected the theory and measurement practices of its era; see Chapman, Weiss, Barrett, and Duberstein.
Meanwhile, lexical researchers kept finding broad factors that resembled one another across samples. Donald Fiske’s 1949 work found recurring structure across ratings from different sources. Tupes and Christal’s 1961 Air Force technical report, republished in 1992, identified five recurrent factors, including Emotional Stability. Warren Norman’s 1963 peer-nomination studies again produced five relatively orthogonal factors. These studies supplied a separate route toward what would later be called the Big Five.
In 1981, John Digman and Naomi Takemoto-Chock reanalyzed six major lexical studies and concluded that five broad factors described the data well. In the mid-1980s, Robert McCrae and Paul Costa directly compared questionnaire-based NEO scales with lexical five-factor measures. Their work helped show that questionnaire Neuroticism and lexical Emotional Stability belonged to the same broad structural territory, although the scoring direction and specific measures differed.
By 1990, the five-factor structure had become much harder to dismiss as a local artifact of one dataset or one instrument. Lewis Goldberg demonstrated a highly replicable Big Five structure across large sets of trait adjectives, while Digman reviewed the emergence of the five-factor model. In 1992, McCrae and Oliver John published a major synthesis explicitly describing the Five-Factor Model in terms of Extraversion, Agreeableness, Conscientiousness, Neuroticism, and Openness to Experience.
In the 1990s and beyond, neuroticism became a standard broad domain across many personality research programs, while instruments continued to operationalize it differently. The NEO tradition used Neuroticism and a six-facet structure. Lexical Big Five measures often used the reverse-pole label Emotional Stability. The BFI-2, introduced in 2017, labels the corresponding domain Negative Emotionality and measures the facets Anxiety, Depression, and Emotional Volatility. These are related models of the same broad territory, not interchangeable score systems.
Before Eysenck: Clinical Language and the Rise of Trait Measurement
The resemblance between neuroticism and neurosis can create a misleading historical story. The older word neurosis belonged to changing medical, psychiatric, and psychoanalytic classifications. Modern neuroticism belongs to personality measurement. Eysenck’s contribution was to make neuroticism a broad psychometric dimension that could vary across people, rather than to preserve neurosis as a categorical diagnosis.
Benjamin Lahey’s review of the public-health significance of neuroticism makes this modern boundary explicit. It describes neuroticism as relatively stable individual differences in negative emotional responding to threat, frustration, or loss and notes that the contemporary construct is defined psychometrically rather than through older theories of unconscious conflict. Lahey’s review is useful because it connects the historical terminology with the modern dimensional construct without treating them as the same thing.
This shift from category to dimension was part of a wider transformation in personality psychology. Researchers increasingly used correlations and factor analysis to ask whether apparently different tendencies clustered into broader traits. A person could therefore be located somewhere along a continuum instead of being assigned to a personality “type.” Eysenck’s early neuroticism was one of the most influential examples of this approach.
The historical boundary remains important today. A high neuroticism score on a personality inventory is not evidence that someone has “neurosis,” and it does not establish any contemporary psychiatric disorder. If the question is about the everyday adjective rather than the construct, Neurotic vs Neuroticism: Everyday Label vs Big Five Personality Trait owns that distinction.
1947: Eysenck Makes Neuroticism a Broad Personality Dimension
Eysenck’s 1947 Dimensions of Personality is a natural starting point for the modern history because it put Neuroticism, conventionally abbreviated N, beside Introversion–Extraversion as a major axis of personality. The book’s chapter structure explicitly included neuroticism, theories of neuroticism, and questionnaire study of neuroticism. The project combined ratings, clinical observations, questionnaires, experimental measures, and factor analysis in an attempt to identify broad dimensions beneath many narrower characteristics.
The sample context matters. Eysenck’s original factor-analytic work was not a twenty-first-century general-population Big Five survey. It drew heavily on male service patients with reactive mental illnesses in a wartime and postwar clinical setting. That context helped shape both the questions asked and the language used. It would be a mistake to treat those early distributions as modern norms or to assume that the original item content defines neuroticism for every later instrument.
At the same time, Eysenck’s dimensional move was durable. Neuroticism became a trait that could be measured in everyone, with higher scores representing greater emotional instability or susceptibility to negative emotion and lower scores representing greater emotional stability. This was a conceptual separation from a yes-or-no diagnosis, even though Eysenck’s research remained closely connected to psychopathology and his terminology reflected its clinical era.
Eysenck also wanted personality dimensions to be explanatory, not merely descriptive. He proposed biological and learning-theory accounts of individual differences. Those hypotheses became an important part of the Eysenckian program, but they are historically separable from the later Big Five taxonomy. Modern Big Five neuroticism does not require accepting Eysenck’s particular biological mechanisms, and a neuroticism score is not a biological diagnosis.
The Eysenck Measurement Line: From Maudsley Scales to the EPQ
A trait becomes scientifically influential partly through measurement. Eysenck’s neuroticism spread because it was repeatedly operationalized in practical questionnaires, producing a family of measures that researchers could administer in large samples.
The Maudsley Medical Questionnaire appeared early in this sequence, followed by the Maudsley Personality Inventory in 1959. A 1958 paper described a short questionnaire for measuring two dimensions of personality, neuroticism and extraversion, in 1,600 adults divided across sex, age, and social-class groups; see Eysenck’s short-questionnaire study. The goal was increasingly standardized quantitative assessment.
In 1964, Hans and Sybil Eysenck introduced the Eysenck Personality Inventory. A contemporary paper described it as an improved short questionnaire for measuring extraversion and neuroticism; the bibliographic record is available through PubMed. The EPI also included a Lie scale intended to help evaluate response tendencies.
The later Eysenck Personality Questionnaire formalized the better-known PEN framework: Psychoticism, Extraversion, and Neuroticism, with a Lie scale alongside them. This matters because Eysenck’s mature taxonomy remained a three-factor system, not a preliminary version of the Big Five waiting for two more traits to be added. Eysenck continued to argue that three very broad dimensions captured the highest-order structure more parsimoniously.
Modern analyses also show why an old EPI Neuroticism score should not be treated as numerically equivalent to a modern Big Five score. Chapman and colleagues found that the EPI Neuroticism domain could itself be decomposed into narrower components and noted that its content was especially focused on anxiety-related material relative to multifaceted Five-Factor Model Neuroticism. Their 2013 analysis illustrates a general measurement principle: two instruments can target the same broad domain while sampling different lower-level content.
A Parallel Route: The Lexical Tradition and Emotional Stability
The Big Five did not descend solely from Eysenck’s questionnaires. Its other major historical route came from the lexical hypothesis and factor analyses of trait-descriptive language.
Allport and Odbert’s 1936 catalog of English trait names supplied an enormous linguistic domain for personality research. Cattell’s 1943 work reduced and clustered this vocabulary to make factor analysis more tractable. The key idea was methodological: rather than beginning with one theorist’s preferred dimensions, researchers could begin with a broad sample of ordinary personality descriptors and ask which higher-order factors emerged from ratings.
Donald Fiske’s 1949 study analyzed personality ratings from different sources and found similar factors across psychologists, teammates, and self-ratings. One of the recurring factors was labeled Emotional Control. Fiske’s paper did not yet deliver the canonical Big Five, but it strengthened the idea that broad personality structure could recur across observers.
The decisive five-factor evidence came from Ernest Tupes and Raymond Christal. Their 1961 Air Force report analyzed ratings on 35 personality traits in eight samples and found five recurrent factors. When the work was republished in Journal of Personality in 1992, the factors were listed as Surgency, Agreeableness, Dependability, Emotional Stability, and Culture. The republished study is historically important because Emotional Stability occupies the broad region that later models often describe in the opposite scoring direction as Neuroticism.
Warren Norman’s 1963 work then replicated a five-factor structure in peer-nomination ratings. Norman’s study helped establish that the five-factor pattern could recur beyond a single military dataset. The factors were not yet standardized under today’s exact names, and researchers disagreed about interpretation, rotation, hierarchy, and the best labels. But the broad architecture was becoming visible.
This parallel history explains a common modern puzzle: some Big Five tests report Emotional Stability rather than Neuroticism. That is not simply a euphemistic renaming invented later. Emotional Stability has deep roots in the lexical five-factor tradition. In many models it is the opposite pole or reverse-scored direction of the broad domain now commonly called neuroticism. For the modern comparison, see Neuroticism vs Emotional Stability: Two Ends of the Same Big Five Dimension?.
1981–1987: The Convergence of Lexical and Questionnaire Traditions
By the early 1980s, researchers had decades of trait-rating studies, but the literature still contained competing labels and factor solutions. Digman and Takemoto-Chock’s 1981 reanalysis was a major integrative step. They reexamined six studies of the natural language of personality and reported that five factors described the data well, with high correspondence across analyses. Their paper helped revive interest in the five-factor solution after a period in which it had not dominated personality psychology.
At roughly the same time, Costa and McCrae were developing the NEO approach. The acronym originally emphasized Neuroticism, Extraversion, and Openness. This is another reason the history cannot be reduced to “the Big Five appeared all at once.” The NEO program itself expanded as evidence accumulated for Agreeableness and Conscientiousness as additional broad domains.
A particularly important bridge appeared in 1985. McCrae and Costa compared lexical five-factor dimensions with questionnaire scales and reported substantial correspondence between lexical Neuroticism and Extraversion factors and the corresponding NEO scales. Their 1985 study was not merely another replication of five factors; it helped connect two research traditions that had used different starting points and different kinds of measures.
In 1987, McCrae and Costa tested the five-factor model across both instruments and observers. They identified Neuroticism, Extraversion, Openness, Agreeableness, and Conscientiousness in peer ratings and showed convergence with questionnaire measures. The cross-instrument, cross-observer validation strengthened the case that the five domains were not artifacts of one questionnaire format.
The historical lesson is convergence, not simple succession. Eysenck’s N did not disappear and then reappear as Big Five N. Lexical Emotional Stability and questionnaire Neuroticism were increasingly recognized as alternative orientations toward a closely corresponding broad domain, while the five-factor taxonomy grew around that convergence.
1990–1992: Neuroticism Becomes a Standard Big Five Domain
The years around 1990 were a consolidation point for the Big Five. Lewis Goldberg’s 1990 paper tested unusually comprehensive sets of trait terms. Across 1,431 trait adjectives organized into 75 clusters, multiple analyses produced closely similar five-factor structures; a second study using common terms and both self- and peer ratings again recovered the same broad structure. Goldberg’s study gave the lexical Big Five a powerful modern empirical demonstration.
Also in 1990, John Digman reviewed the field under the title Personality Structure: Emergence of the Five-Factor Model. Digman’s review documented how several once-separated lines of personality research were converging on five broad dimensions.
In 1992, McCrae and John published one of the most influential summaries of the Five-Factor Model. Their formulation explicitly defined the five basic dimensions as Extraversion, Agreeableness, Conscientiousness, Neuroticism, and Openness to Experience and argued that evidence from both natural-language adjectives and theoretically based questionnaires supported the model. Their 1992 synthesis is a key historical marker for neuroticism’s status as a standard Five-Factor Model domain.
The same period also saw continued debate. Costa and McCrae argued that the five factors were basic dimensions supported by longitudinal, cross-observer, cross-cultural, and other evidence; see Four Ways Five Factors Are Basic90236-I). Eysenck responded that three higher-order factors offered a more theoretically grounded account and criticized the Five-Factor Model on hierarchy and theory; see Four Ways Five Factors Are Not Basic90237-J).
That exchange is historically useful because it prevents a retrospective simplification. Personality science did not discover that Eysenck was “wrong” and then merely replace three traits with five. Researchers were debating different levels of trait hierarchy, different methods for defining broad dimensions, and different expectations about whether a taxonomy should also specify biological causes. Modern personality science retains evidence for hierarchical structure: broad domains can be organized above narrower facets and, in some models, beneath still broader higher-order factors.
The NEO-PI-R and the Facet Expansion of Neuroticism
The Five-Factor Model became especially influential because it was paired with detailed instruments. The NEO Personality Inventory was first published in 1985 and revised in 1992 as the NEO-PI-R; the APA Dictionary entry for the NEO Personality Inventory summarizes that publication history.
The NEO-PI-R assessed all five domains and six facets within each domain. For Neuroticism, the classic NEO-PI-R facets were Anxiety, Angry Hostility, Depression, Self-Consciousness, Impulsiveness, and Vulnerability. Costa, McCrae, and Dye’s work on revising the inventory described the extension of facet measurement across Agreeableness and Conscientiousness while retaining and modifying the existing Neuroticism, Extraversion, and Openness facets; see their 1991 paper90177-D).
This facet architecture was historically influential, but it is instrument-specific. It should not be projected backward onto the EPI or forward onto every later Big Five measure. An EPI Neuroticism scale, a NEO-PI-R Neuroticism domain, a NEO-FFI Neuroticism domain, and a BFI-2 Negative Emotionality domain all belong to overlapping personality territory, yet their items, facet representation, score ranges, and normative systems differ.
That measurement history is one reason modern readers should resist statements such as “a neuroticism score of 30 is high.” A raw score has meaning only within a specified instrument, scoring procedure, reference group, and population. A percentile also depends on the relevant norms. Historical continuity of a construct does not create numerical interchangeability between tests.
Neuroticism, Emotional Stability, and Negative Emotionality: Why the Labels Differ
Three labels now appear frequently around the same broad personality territory.
Neuroticism is the conventional label in Eysenck’s inventories and in the NEO/Five-Factor Model tradition. It emphasizes the tendency toward negative emotional reactivity and vulnerability to distress.
Emotional Stability is often used for the opposite pole, especially in lexical Big Five traditions. Higher Emotional Stability therefore usually corresponds to lower Neuroticism within the same conceptual dimension. The exact scoring direction depends on the instrument, so the labels should be read together with the scale documentation.
Negative Emotionality is the domain label used in the BFI-2. The BFI-2 is a 60-item instrument with 15 facets across five domains. Its Negative Emotionality domain includes Anxiety, Depression, and Emotional Volatility; see Soto and John’s BFI-2 validation and the Berkeley Personality Lab’s BFI-2 documentation.
These labels show both continuity and evolution. The broad domain is recognizable across traditions, but each instrument decides how to sample it. Negative Emotionality is therefore not a universal replacement word that makes every historical neuroticism scale identical, and Emotional Stability is not always obtained by a simple arithmetic transformation unless the particular test defines it that way.
The same caution applies to nearby concepts. Negative affect can refer to current, recent, or characteristic unpleasant emotional experience. Trait anxiety is narrower and centers on anxiety proneness. Psychiatric symptoms are clinical phenomena whose duration, severity, context, and impairment may matter diagnostically. Neuroticism is a broad personality disposition. Correlation among these constructs does not erase their conceptual and measurement boundaries.
What Big Five Neuroticism Inherited from Eysenck
Modern Big Five neuroticism clearly inherited several important ideas from the Eysenckian program.
First, it is dimensional. People vary along a continuum rather than belonging to a “neurotic” versus “non-neurotic” type.
Second, it is broad. Neuroticism is meant to summarize covariance among multiple forms of negative emotional tendency rather than to represent one isolated symptom.
Third, it is psychometric. The construct is operationalized through patterns of responses and validated through relations among items, scales, observers, and outcomes.
Fourth, it has a stable-individual-differences interpretation. The construct concerns characteristic tendencies across time and situations, even though personality can change and momentary emotional states fluctuate.
These points help explain why Eysenck remains central to the history even though the Big Five did not originate as an expansion of his PEN model. His neuroticism offered a strong questionnaire-based dimensional construct that could converge with the lexical domain of Emotional Stability.
What Big Five Neuroticism Did Not Simply Inherit
The modern domain also differs from Eysenck’s original framework in important ways.
It did not inherit one fixed item set. The EPI, EPQ, NEO inventories, BFI/BFI-2, and IPIP-based scales use different items and can emphasize different lower-level content.
It did not inherit one fixed facet system. Chapman and colleagues showed that the EPI’s lower-order structure differs meaningfully from the later NEO facet architecture. The BFI-2 uses yet another three-facet Negative Emotionality structure.
It did not inherit Eysenck’s entire three-factor taxonomy. Big Five models include Agreeableness, Conscientiousness, and Openness/Open-Mindedness as broad domains alongside Extraversion and Neuroticism/Negative Emotionality.
It did not inherit Eysenck’s specific causal theory as a requirement. A researcher can use Big Five neuroticism descriptively without accepting a particular physiological mechanism. Contemporary genetics and neuroscience investigate correlates and mechanisms at the group level, but no brain measure or genetic profile turns neuroticism into a biological diagnosis or gives a deterministic individual prediction.
Finally, it did not inherit the clinical category of neurosis. Modern neuroticism is a personality trait. Historical continuity in vocabulary should not be mistaken for diagnostic continuity.
Was Neuroticism Added to the Big Five?
There was no single date on which psychologists “added neuroticism” to an already completed Big Five.
The lexical tradition had long contained a broad factor usually labeled Emotional Stability or a similar term. Eysenck had already established Neuroticism as a major questionnaire dimension. During the 1980s, researchers demonstrated stronger correspondence between lexical and questionnaire traditions. By the early 1990s, influential Five-Factor Model syntheses increasingly used Neuroticism as the standard label for the domain.
So the better historical description is alignment. A preexisting neuroticism tradition and a preexisting emotional-stability factor increasingly became understood as measuring closely related broad personality structure. The name Neuroticism became especially prominent through the NEO and Five-Factor Model literature, while Emotional Stability remained common in lexical, organizational, and applied Big Five measures.
This is also why sources can disagree superficially while describing the same broad dimension. One may report higher Neuroticism; another may report lower Emotional Stability. Interpretation requires knowing the direction of the scale.
Did Eysenck Invent the Big Five?
No. Eysenck was one of the most important architects of modern dimensional personality psychology, and his Neuroticism and Extraversion factors became central points of comparison for later models. But the Big Five arose from a wider network of lexical and psychometric research involving Allport and Odbert, Cattell, Fiske, Tupes and Christal, Norman, Digman, Goldberg, Costa, McCrae, John, and many others.
The historical relationship is better described as convergence between traditions than as direct descent. Eysenck’s model emphasized a small number of broad, theoretically interpreted dimensions. Lexical Big Five work emphasized recurrent structure in trait language. The NEO/Five-Factor Model tradition helped connect lexical factors with questionnaire scales and elaborate them into a hierarchical domain-and-facet system.
This distinction prevents a common timeline error: Eysenck’s PEN model did not become the Big Five merely by adding Agreeableness and Conscientiousness. The models overlap, but their historical origins, theoretical commitments, and factor hierarchies differ.
How Measurement Changed the Construct Without Making It a Different Trait
Personality constructs are not independent of their measures. Every instrument samples behavior, feelings, and self-descriptions through a particular set of items. As instruments changed, the observable content of neuroticism changed too.
The EPI and EPQ measure Eysenckian Neuroticism within a specific theoretical system. Their raw scores and historical norms belong to those instruments.
The NEO-PI-R and later NEO instruments measure Five-Factor Model Neuroticism with a richer facet representation. A NEO-FFI score is a shorter domain-level measure and should not be treated as if it contains the same facet information as the full NEO inventory.
The BFI-2 uses the label Negative Emotionality and organizes that domain into Anxiety, Depression, and Emotional Volatility. Its scoring system is its own. A BFI-2 result should be reported as BFI-2 Negative Emotionality rather than silently converted into a supposed NEO or Eysenck score.
The International Personality Item Pool is different again. IPIP is a public-domain item pool and framework from which multiple scales can be constructed; it is not one single universal neuroticism test. A result described only as “IPIP neuroticism” is incomplete unless the specific scale and scoring method are identified. See the International Personality Item Pool for the measure family.
This measurement specificity is especially important when reading old studies. A historical EPI score cannot be compared directly with a modern BFI-2 raw score. Even when both are standardized, the meaning of a standard score depends on the reference sample and the content of the underlying scale.
From Clinical Samples to General-Population Personality Science
Another major historical change concerns who was studied. Eysenck’s 1947 work was deeply connected to clinical populations and wartime service patients. Later questionnaire development expanded to broader samples. Lexical research often used peer ratings and self-ratings in nonclinical groups. By the 1980s and 1990s, researchers were testing the five-factor structure across instruments, observers, ages, languages, and cultures.
This expansion changed the status of neuroticism. It became less tied to a specific clinical population and more clearly understood as a population-wide trait dimension. Everyone can be located somewhere along the domain, and most variation is ordinary personality variation rather than evidence of disorder.
McCrae and John’s 1992 review emphasized that both natural-language adjective research and theoretically based questionnaires supported the Five-Factor Model across observers and cultures. That does not mean every culture produces an identical factor structure under every method. It means the evidence base had moved far beyond the early clinical samples from which Eysenck developed his first formulations.
The shift also strengthened the distinction between trait measurement and psychiatric assessment. A personality inventory is designed to describe relatively broad individual differences. A clinical diagnosis requires disorder-specific criteria, symptom patterns, duration, distress or impairment where relevant, differential assessment, and clinical context.
Neuroticism and Mental Health: Historical Connection, Modern Boundary
Neuroticism has always had a close historical relationship with psychological distress. Eysenck’s early work emerged partly from attempts to understand differences among patients described in the clinical language of his time. That history can tempt modern readers to treat neuroticism as a mild psychiatric disorder. Contemporary evidence supports a different interpretation.
Neuroticism is associated with a wide range of mental-health outcomes at the population level. Lahey’s review characterized it as a major public-health-relevant trait, and Ormel and colleagues reviewed its complex relationships with common mental disorders. Ormel et al. emphasized that neuroticism can function as a broad vulnerability or risk-related construct while also sharing content with some forms of psychopathology.
Association is not diagnosis. Higher neuroticism may be statistically associated with anxiety, depression, comorbidity, or symptom severity, but a personality score does not establish an anxiety disorder, depressive disorder, OCD, PTSD, ADHD, autism, a personality disorder, or any other diagnosis. The trait is transdiagnostic in its associations precisely because it is broader than any one disorder.
Modern structural work also shows continuity between normal and abnormal personality without turning normal traits into diagnoses. Markon, Krueger, and Watson found that normal and abnormal personality can be organized within an integrative hierarchical framework; see their 2005 study. Dimensional continuity means that traits can relate to psychopathology across a range. It does not mean a trait score is itself a clinical classification.
What the History Does and Does Not Establish
The historical record establishes that neuroticism is a durable broad personality construct that appears under related labels across major trait traditions. It also establishes that measurement has changed substantially. That combination makes continuity and caution equally important.
Historical continuity does not prove that every questionnaire called a neuroticism scale measures exactly the same content. Construct names can remain stable while items, response formats, facet coverage, and norms evolve.
Historical association with psychiatric populations does not make neuroticism a diagnosis. The modern trait can be studied in clinical and nonclinical populations, and higher levels can correlate with symptoms without being reducible to those symptoms.
Historical biological theories do not establish a single biological cause. Heritability estimates, polygenic associations, physiological correlates, and brain-imaging findings concern population-level variation and statistical relationships. They should not be converted into deterministic explanations of one person’s score.
Historical factor-analytic convergence also does not mean that all researchers agree on one uniquely correct hierarchy. The Big Five is a powerful broad taxonomy, while alternative models can organize personality at different levels or with different dimensions. The history of neuroticism is therefore a history of empirical convergence within an evolving measurement system, not the discovery of one immutable questionnaire.
Why the History Still Matters for Research and Everyday Interpretation
The history of neuroticism explains several problems that otherwise look like contradictions.
If one paper uses Neuroticism and another uses Emotional Stability, the authors may be measuring opposite directions of the same broad domain.
If one study uses EPI Neuroticism and another uses NEO-PI-R Neuroticism, the labels may match while the item content and facet coverage differ.
If a BFI-2 paper reports Negative Emotionality, it is operating in closely corresponding Big Five territory but using the BFI-2’s own terminology and three-facet structure.
If an older article discusses “neurotics” or “neurosis,” it may be using a historical clinical vocabulary that cannot be mapped directly onto modern diagnoses or modern Big Five scores.
If someone receives a “high” neuroticism score online, the number has no universal interpretation unless the instrument, scoring method, reference sample, and population are known.
History therefore improves measurement literacy. It encourages researchers and readers to name the actual instrument, respect score direction, separate broad domains from facets, and avoid treating a personality trait as a disease.
What Neuroticism Means Today
Today, neuroticism is best understood as a broad, dimensional personality construct describing relatively stable differences in negative emotionality and susceptibility to distress. It is one of the core domains in the Five-Factor Model and a close counterpart of Emotional Stability in reverse-scored Big Five systems. Some modern measures, especially the BFI-2, use the label Negative Emotionality.
The historical path to this definition had multiple branches. Eysenck established a powerful dimensional and questionnaire tradition. Lexical researchers independently identified Emotional Stability as one of a recurring set of broad factors. Costa and McCrae connected questionnaire and lexical structures and developed the NEO framework. Goldberg, Digman, John, and others helped consolidate the five-factor taxonomy. Later instruments refined the domain and its facets without making all measures equivalent.
The strongest continuity across this history is structural rather than numerical. Researchers repeatedly found a broad axis involving vulnerability to negative emotion versus emotional stability. The precise words, items, facets, and theories around that axis changed. That is why modern neuroticism is historically connected to Eysenck while also being larger than Eysenck’s original model.
For a current construct-focused explanation rather than the historical development, see Neuroticism as a Personality Trait: How Negative Emotionality Works.
Frequently Asked Questions
Who introduced neuroticism as a modern personality dimension?
Hans J. Eysenck was the central figure in establishing neuroticism as a broad psychometric personality dimension. His 1947 Dimensions of Personality placed Neuroticism alongside Introversion–Extraversion and developed a factor-analytic research program around the two dimensions. The word and older clinical ideas predated this work, but Eysenck’s contribution was crucial to the modern measurable trait.
Did Eysenck invent the Big Five?
No. Eysenck developed a two-factor model and later the three-factor PEN model. The Big Five emerged from a separate but eventually converging lexical and psychometric research tradition involving Allport and Odbert, Cattell, Fiske, Tupes and Christal, Norman, Digman, Goldberg, Costa, McCrae, John, and others.
When did neuroticism become one of the Big Five?
There is no single addition date. Lexical five-factor studies had long identified a broad Emotional Stability factor. During the 1980s, questionnaire Neuroticism and lexical five-factor structures were increasingly linked empirically. By the early 1990s, influential Five-Factor Model syntheses routinely named Neuroticism as one of the five broad domains.
Is Eysenck Neuroticism the same as Big Five Neuroticism?
They overlap strongly at the broad-domain level, but they are not identical measures. Eysenck’s EPI and EPQ have their own item content and theoretical history. NEO, BFI-2, and other Big Five measures sample the domain differently. Raw scores, norms, and facet structures should not be transferred between instruments.
Why do some Big Five tests use Emotional Stability?
Emotional Stability has deep roots in the lexical five-factor tradition. It generally represents the opposite pole of neuroticism. A scale scored toward Emotional Stability therefore usually runs in the reverse direction from a scale scored toward Neuroticism, but exact score transformations depend on the specific instrument.
Why does the BFI-2 say Negative Emotionality instead of Neuroticism?
The BFI-2 uses Negative Emotionality as its domain label and measures three facets: Anxiety, Depression, and Emotional Volatility. It covers closely corresponding Big Five territory but has its own items and scoring system.
Is neuroticism the same as neurosis?
No. Neurosis is a historical clinical concept whose meaning changed across psychiatric and psychoanalytic traditions. Neuroticism is a dimensional personality trait. A neuroticism score does not establish a historical or modern psychiatric diagnosis.
Is high neuroticism a mental disorder?
No. Higher neuroticism is a personality-trait description. It is associated statistically with greater risk for several forms of psychological distress and psychopathology, but the trait itself is not a psychiatric diagnosis.
Can I compare an old EPI score with a NEO or BFI-2 score?
Not directly. The instruments differ in item content, response format, scoring, facet structure, and norms. Comparisons require an explicit psychometric linking or calibration study; raw-score conversion by intuition is not valid.