Academic Journal
View in EDS

PsychAdapter: adapting LLMs to reflect traits, personality, and mental health.

Bibliographic Details
Title: PsychAdapter: adapting LLMs to reflect traits, personality, and mental health.
Authors: Vu, Huy, Nguyen, Huy Anh, Ganesan, Adithya V., Juhng, Swanie, Kjell, Oscar N. E., Sedoc, Joao, Kern, Margaret L., Boyd, Ryan L., Ungar, Lyle, Schwartz, H. Andrew, Eichstaedt, Johannes C.
Source: npj Artificial Intelligence; 3/2/2026, Vol. 2 Issue 1, p1-14, 14p
Subject Terms: Artificial intelligence, Personality, Mental health, Program generators (Computer programs), Language models, Psychological adaptation, Natural language processing, Transformer models, Five-factor model of personality
Abstract: AI language generators are now ubiquitous but typically produce generic text that fails to reflect individual differences. Here, we introduce PsychAdapter, a lightweight LLM architectural modification that uses empirically derived links between language and personality, demographic, and mental health traits to generate trait-reflective text, regardless of prompt. PsychAdapter was applied to GPT-2, Gemma-2B, and LLaMA-3, and expert raters confirmed that the generated text matched the specified traits: it produced Big Five personality traits with 87.3% and depression and life satisfaction with 96.7% accuracy. PsychAdapter is a novel method for embedding psychological behavioral patterns into language models by conditioning every transformer layer, without relying on prompting. Beyond personality-conditioned generation, this approach has potential uses for simulated patients reflecting psychopathology and translation tailored to reading or educational level. It also enables generation of characteristic sentences for studying the language of traits, expanding the language processing toolkit for psychology. [ABSTRACT FROM AUTHOR]
Copyright of npj Artificial Intelligence is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Complementary Index
FullText Text:
  Availability: 0
Header DbId: edb
DbLabel: Complementary Index
An: 192010979
RelevancyScore: 1061
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 1060.7568359375
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: PsychAdapter: adapting LLMs to reflect traits, personality, and mental health.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Vu%2C+Huy%22">Vu, Huy</searchLink><br /><searchLink fieldCode="AR" term="%22Nguyen%2C+Huy+Anh%22">Nguyen, Huy Anh</searchLink><br /><searchLink fieldCode="AR" term="%22Ganesan%2C+Adithya+V%2E%22">Ganesan, Adithya V.</searchLink><br /><searchLink fieldCode="AR" term="%22Juhng%2C+Swanie%22">Juhng, Swanie</searchLink><br /><searchLink fieldCode="AR" term="%22Kjell%2C+Oscar+N%2E+E%2E%22">Kjell, Oscar N. E.</searchLink><br /><searchLink fieldCode="AR" term="%22Sedoc%2C+Joao%22">Sedoc, Joao</searchLink><br /><searchLink fieldCode="AR" term="%22Kern%2C+Margaret+L%2E%22">Kern, Margaret L.</searchLink><br /><searchLink fieldCode="AR" term="%22Boyd%2C+Ryan+L%2E%22">Boyd, Ryan L.</searchLink><br /><searchLink fieldCode="AR" term="%22Ungar%2C+Lyle%22">Ungar, Lyle</searchLink><br /><searchLink fieldCode="AR" term="%22Schwartz%2C+H%2E+Andrew%22">Schwartz, H. Andrew</searchLink><br /><searchLink fieldCode="AR" term="%22Eichstaedt%2C+Johannes+C%2E%22">Eichstaedt, Johannes C.</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: npj Artificial Intelligence; 3/2/2026, Vol. 2 Issue 1, p1-14, 14p
– Name: Subject
  Label: Subject Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Artificial+intelligence%22">Artificial intelligence</searchLink><br /><searchLink fieldCode="DE" term="%22Personality%22">Personality</searchLink><br /><searchLink fieldCode="DE" term="%22Mental+health%22">Mental health</searchLink><br /><searchLink fieldCode="DE" term="%22Program+generators+%28Computer+programs%29%22">Program generators (Computer programs)</searchLink><br /><searchLink fieldCode="DE" term="%22Language+models%22">Language models</searchLink><br /><searchLink fieldCode="DE" term="%22Psychological+adaptation%22">Psychological adaptation</searchLink><br /><searchLink fieldCode="DE" term="%22Natural+language+processing%22">Natural language processing</searchLink><br /><searchLink fieldCode="DE" term="%22Transformer+models%22">Transformer models</searchLink><br /><searchLink fieldCode="DE" term="%22Five-factor+model+of+personality%22">Five-factor model of personality</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: AI language generators are now ubiquitous but typically produce generic text that fails to reflect individual differences. Here, we introduce PsychAdapter, a lightweight LLM architectural modification that uses empirically derived links between language and personality, demographic, and mental health traits to generate trait-reflective text, regardless of prompt. PsychAdapter was applied to GPT-2, Gemma-2B, and LLaMA-3, and expert raters confirmed that the generated text matched the specified traits: it produced Big Five personality traits with 87.3% and depression and life satisfaction with 96.7% accuracy. PsychAdapter is a novel method for embedding psychological behavioral patterns into language models by conditioning every transformer layer, without relying on prompting. Beyond personality-conditioned generation, this approach has potential uses for simulated patients reflecting psychopathology and translation tailored to reading or educational level. It also enables generation of characteristic sentences for studying the language of traits, expanding the language processing toolkit for psychology. [ABSTRACT FROM AUTHOR]
– Name: Abstract
  Label:
  Group: Ab
  Data: <i>Copyright of npj Artificial Intelligence is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edb&AN=192010979
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1038/s44387-026-00071-9
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 14
        StartPage: 1
    Subjects:
      – SubjectFull: Artificial intelligence
        Type: general
      – SubjectFull: Personality
        Type: general
      – SubjectFull: Mental health
        Type: general
      – SubjectFull: Program generators (Computer programs)
        Type: general
      – SubjectFull: Language models
        Type: general
      – SubjectFull: Psychological adaptation
        Type: general
      – SubjectFull: Natural language processing
        Type: general
      – SubjectFull: Transformer models
        Type: general
      – SubjectFull: Five-factor model of personality
        Type: general
    Titles:
      – TitleFull: PsychAdapter: adapting LLMs to reflect traits, personality, and mental health.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Vu, Huy
      – PersonEntity:
          Name:
            NameFull: Nguyen, Huy Anh
      – PersonEntity:
          Name:
            NameFull: Ganesan, Adithya V.
      – PersonEntity:
          Name:
            NameFull: Juhng, Swanie
      – PersonEntity:
          Name:
            NameFull: Kjell, Oscar N. E.
      – PersonEntity:
          Name:
            NameFull: Sedoc, Joao
      – PersonEntity:
          Name:
            NameFull: Kern, Margaret L.
      – PersonEntity:
          Name:
            NameFull: Boyd, Ryan L.
      – PersonEntity:
          Name:
            NameFull: Ungar, Lyle
      – PersonEntity:
          Name:
            NameFull: Schwartz, H. Andrew
      – PersonEntity:
          Name:
            NameFull: Eichstaedt, Johannes C.
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 02
              M: 03
              Text: 3/2/2026
              Type: published
              Y: 2026
          Numbering:
            – Type: volume
              Value: 2
            – Type: issue
              Value: 1
          Titles:
            – TitleFull: npj Artificial Intelligence
              Type: main
ResultId 1