Text analysis of 450 autobiographical accounts found that each dimension of psychopathy carried its own linguistic signature — and the researchers are explicit that this is not a diagnostic test.

Psychopathy is popularly understood as a single thing a person either is or is not. Among researchers who study it, that framing has been giving way for some time to a view of psychopathy as several distinct dimensions, present to some degree in everyone, each predicting different outcomes. A study from the University of Mississippi, published in Personality and Individual Differences, tests that view in an unusual place: how people narrate their own lives.

“People tend to think of psychopathy as one thing, but it’s actually multidimensional,” said Ky Bray, a doctoral candidate in psychology at Ole Miss. “What we mean when we say that is everyone exists on this spectrum. Everyone has some degree of meanness, boldness and disinhibition.”

What was measured

The researchers surveyed 450 participants, including college students and people who reported prior incarceration. Each answered open-ended questions about the high points, low points and turning points of their life, producing hundreds of autobiographical accounts. They then completed a questionnaire measuring psychopathic traits along three dimensions: meanness, boldness and disinhibition.

Text-analysis software counted words in categories including positive tone, negative tone, achievement, clout and risk — the standard approach in this area, which treats word frequencies in particular semantic categories as measurable features of a text rather than attempting to interpret meaning.

Three dimensions, three patterns

The three dimensions did not behave alike, which is the study’s central result.

Higher disinhibition — impulsivity, poor behavioural control — went with more negative words, and with fewer references to friends, fewer positive words and fewer achievement words. The picture is of a life narrated as difficult and somewhat solitary.

Higher meanness — callousness, lack of empathy — went the other way: more positive tone, more achievement and reward language, less negative emotional language. A life narrated as going well, in terms of what was gained.

Higher boldness — social dominance, fearlessness, stress immunity — went with fewer anxiety words such as fear, afraid or worry, and fewer references to mental illness.

The contrast between disinhibition and meanness is the finding worth sitting with. Both are components of the same construct, and on the measure of emotional tone they point in opposite directions. Any approach that treated psychopathy as a single quantity would average these two signals and find close to nothing. That is a reasonable argument for the dimensional model, made with data rather than assertion.

The researchers also report that the associations between words and traits depended on which narrative question a participant was answering — high point, low point or turning point. Context shaped the signal, which is a caution against reading any single word count as a fixed trait marker.

The researchers’ own limits

Bray is unusually direct about what the work is not. “The goal here is not to diagnose psychopathy based on language,” he said. “It’s to point to the fact that psychopathy shows up in everyday behaviors, even subtle behaviors like language.”

That disclaimer deserves amplification rather than paraphrase, because this is exactly the kind of result that gets misread. These are group-level statistical associations across 450 people. They do not permit inference about any individual, and nothing here supports using language to screen, assess or label a particular person. The release reports no effect sizes, no correlation coefficients and no indication of how much of the variation in trait scores the word categories account for — which, in text-analysis studies of personality, is typically small.

The sample composition also warrants attention. Combining college students with people reporting prior incarceration produces a wider range of trait scores than either group alone, which is useful for detecting associations. It also means the sample is not representative of any single population, and that group differences in life circumstances could contribute to differences in how people narrate highs, lows and turning points, independently of personality.

Why narrative, and not something simpler

The choice to study life stories rather than, say, speech in a laboratory task reflects a specific theoretical commitment. “What this tells us is the story we tell about our life has meaning,” said Laura Drislane, an assistant professor of psychology. “It helps people make sense of their experiences, develop a worldview and a coherent sense of who they are.”

“What we’re starting to see is that in psychopathy, there are traits that relate to differences in meaning-making and the ways people think about their life,” she said.

Narrative identity is an established area of personality research resting on the premise that how someone organises their life into a story reveals something that a questionnaire does not. Applying it to psychopathy is a reasonable extension, and the finding that the three dimensions produce different narrative signatures is consistent with the premise.

Why the distinctions carry weight

Drislane situates the work in a shift in how the field talks about the subject. “That’s in part why there’s a movement away from calling people ‘psychopaths,’ because we don’t view psychopathy as a category,” she said. “But psychopathic dimensions can predict important outcomes, things like crime, recidivism or intimate partner violence.”

The practical argument follows from that. “What our research, including this paper, suggests is that it’s important to look beyond that label and into the different components or dimensions of psychopathy, because each of those dimensions actually has a unique set of outcomes or things that it can predict,” she said.

Where that leads is prevention rather than classification. “If we want to be able to treat or ideally even prevent psychopathy from developing, we need to understand the mechanisms at play,” Drislane said. “That comes down to understanding the unique contributors for all three of these dimensions.” The release notes that psychopathy can be treated to minimise symptoms — a point that sits awkwardly against the popular assumption that it cannot.

Bray frames the methodological ambition more broadly: “If we want to understand psychopathy and the different traits that comprise it, we need to think creatively about how these traits shape social interaction.”

The release does not identify a first or corresponding author, naming only Bray and Drislane and stating that they conducted the participant questioning; the endnote below therefore carries the paper title alone. No funding sources and no conflict-of-interest statement are listed.

---

Notes

1. “Investigating interpersonal and affective features of psychopathy through language use.” Personality and Individual Differences, 2026. DOI: 10.1016/j.paid.2026.113978.
2. University of Mississippi. “Researchers study how psychopathic traits shape language.” EurekAlert!, 5 October 2026. https://www.eurekalert.org/news-releases/1146564.