---
title: "Automatic Recognition of Personality in Conversation — paper by Mairesse and Walker (2006)"
description: "The paper asks whether a speaker's Big Five personality traits can be predicted automatically from everyday… 22 references, cited by 85 papers here."
authors: ["François Mairesse", "Marilyn Walker"]
year: 2006
venue: "Proceedings of the Human Language Technology Conference of the NAACL, Companion Volume: Short Papers"
source: "https://aclanthology.org/N06-2022/"
licence: "CC BY-NC-SA 3.0 (ACL Anthology, before 2016)"
licence_url: "https://creativecommons.org/licenses/by-nc-sa/3.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/mairesse-2006-automatic-recognition-personality-conversation/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/mairesse-2006-automatic-recognition-personality-conversation/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2006](https://md.research.elamur.ai/free/en/papers/2006/) / Automatic Recognition of Personality in Conversation

# Automatic Recognition of Personality in Conversation

**Authors:** François Mairesse and [Marilyn Walker](https://md.research.elamur.ai/free/en/authors/marilyn-walker/) · **Year:** 2006 · **Venue:** [Proceedings of the Human Language Technology Conference of the NAACL, Companion Volume: Short Papers](https://md.research.elamur.ai/free/en/venues/naacl/) ([NAACL 2006](https://md.research.elamur.ai/free/en/venues/naacl/2006/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY-NC-SA 3.0 (ACL Anthology, before 2016)](https://creativecommons.org/licenses/by-nc-sa/3.0/)

**Links:** [Publisher's page](https://aclanthology.org/N06-2022/) · [PDF](https://aclanthology.org/N06-2022.pdf) · [OpenAlex](https://openalex.org/W2079506198)

**Citations:** 79 citations (OpenAlex, 2026-10-14) · [85](https://md.research.elamur.ai/free/en/papers/mairesse-2006-automatic-recognition-personality-conversation/cited-by/) in our library, of the 141,840 papers whose bibliographies we read

## Summary
The paper asks whether a speaker's Big Five personality traits can be predicted automatically from everyday conversation, and whether such predictions hold for people the model has never seen. The authors drew on conversation snippets from 96 participants, each of whom carried an Electronically Activated Recorder (EAR) over a two-day period. Each person's transcribed speech was scored by 5–7 outside observers using the Big Five Inventory. From each person's transcripts and audio, the authors extracted 117 features in four groups: LIWC word categories, MRC psycholinguistic norms, utterance-type ratios and Praat prosodic measures. Because personality judgments are relative, they treated the task as pairwise ranking and trained one RankBoost model per trait. Evaluation used 10-fold cross-validation against a random-order baseline. Models for extraversion, agreeableness, conscientiousness and intellect beat the baseline significantly. Emotional stability was the hardest trait to model. The rule-based models reproduced known language–personality links and also surfaced new prosodic and lexical cues. This matters for adaptive dialogue systems, meeting analysis and matching applications.

## Key points
- **Data:** The corpus, collected by Mehl et al., contains 97,468 words in 15,269 utterances from 96 participants.
  - Only each participant's own speech was transcribed, and recording was random, to protect privacy.
  - Observer ratings were highly reliable (r = 0.84).
- **Features (117 total):**
  - 88 LIWC features.
  - 14 MRC features.
  - 4 utterance-type features: command, prompt, question and assertion. These were tagged with parse-tree heuristics that were 88% accurate on 100 sampled utterances.
  - 11 prosodic features from Praat: pitch, intensity, voiced time and speech rate.
- **Ranking error with all features** (random baseline = 0.5):
  - Agreeableness 0.31
  - Conscientiousness 0.33
  - Extraversion 0.35
  - Intellect 0.38
  - Emotional stability 0.40 (not significant)
- **Results by feature group:**
  - Prosody alone gave the best extraversion result (0.26), better than the full feature set.
  - MRC features worked best for emotional stability (0.39).
  - Utterance-type features alone never beat the baseline.
  - Models trained on self-reported personality never significantly outperformed the baseline.
- **Example learned cues:**
  - Extraverts speak faster, with higher pitch, and produce more speech. Introverts show more pitch variation.
  - Swearing pushes agreeableness and conscientiousness scores down.
  - Talking about one's job signals conscientiousness.

## Interesting ideas
- **Frame subjective trait scores as pairwise ranking rather than regression or classification.** Personality is inherently relative to other people, so this framing fits the target. It can be reused for other relative judgments, such as text formality, readability or annotator-scored quality.
- **Use a ranking learner that outputs threshold rules with weights.** RankBoost does this, so the trained model can double as an analysis tool. The rules expose cues for both ends of a scale, including cues a simple correlation study would miss.
- **Train and score each feature family separately as well as together.** This ablation showed one cheap group (prosody) outperforming the full set for one trait. It is a useful check before committing to expensive feature pipelines, such as those that depend on ASR.
- **Hold out entire speakers, not just utterances.** Testing on unseen individuals measures real generalization and avoids rewarding a model for memorizing people. This applies to any speaker- or author-level profiling task.
- **Compare observer ratings with self-reports as targets.** Here, observer-perceived personality was learnable while self-reported personality was not. This suggests that the choice of label source can decide whether a subjective task is feasible at all.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [University of Sheffield](https://md.research.elamur.ai/free/en/universities/university-of-sheffield/) — University, Sheffield, United Kingdom · 76 works · 20,808 citations

📚 Shelves (7): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 9,086 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 8,889 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 9,296 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 9,206 works · [💭 Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/) · 798 works · [💭 Experimental and Cognitive Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/experimental-and-cognitive-psychology/) · 463 works · [💭 Social Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/social-psychology/) · 129 works

## Cite

Cite as: François Mairesse and Marilyn Walker. 2006. Automatic Recognition of Personality in Conversation. Proceedings of the Human Language Technology Conference of the NAACL, Companion Volume: Short Papers.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{mairesse2006automatic,
  title = {{Automatic Recognition of Personality in Conversation}},
  author = {François Mairesse and Marilyn Walker},
  booktitle = {Proceedings of the Human Language Technology Conference of the NAACL, Companion Volume: Short Papers},
  year = {2006},
  doi = {10.3115/1614049.1614071},
  url = {https://doi.org/10.3115/1614049.1614071}
}
```

- [← previous Parsing Arabic Dialects](https://md.research.elamur.ai/free/en/papers/chiang-2006-parsing-arabic-dialects/)
- [next → Talbanken05: A Swedish Treebank with Phrase Structure and Dependency Annotation](https://md.research.elamur.ai/free/en/papers/nivre-2006-talbanken05-swedish-treebank-phrase-structure/)
- [↑ back to papers of 2006](https://md.research.elamur.ai/free/en/papers/2006/)

## References

22 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. E. André, M. Klesen, P. Gebhard, S. Allen, and T. Rist. 1999. Integrating models of personality and emotions into lifelike characters. In Proc. of the International Workshop on Affect in Interactions, p. 136–149. [find](https://scholar.google.com/scholar?q=E%2E+Andre%CC%81%2C+M%2E+Klesen%2C+P%2E+Gebhard%2C+S%2E+Allen%2C+and+T%2E+Rist%2E+1999%2E+Integrating+models+of+personality+and+emotions+into+lifelike+characters%2E+In+Proc%2E+of+the+International+Workshop+on+Affect+in+Interactions)
2. P. Boersma. 2001. Praat, a system for doing phonetics by computer. Glot International, 5(9/10):341–345. [find](https://scholar.google.com/scholar?q=P%2E+Boersma%2E+2001%2E+Praat%2C+a+system+for+doing+phonetics+by+computer%2E+Glot+International%2C+5%289%2F10%29%3A341%E2%80%93345%2E)
3. M. Coltheart. 1981. The MRC psycholinguistic database. Quarterly J. of Experimental Psychology, 33A:497–505. [find](https://scholar.google.com/scholar?q=M%2E+Coltheart%2E+1981%2E+The+MRC+psycholinguistic+database%2E+Quarterly+J%2E+of+Experimental+Psychology%2C+33A%3A497%E2%80%93505%2E)
4. B. Donnellan, R. D. Conger, and C. M. Bryant. 2004. The Big Five and enduring marriages. J. of Research in Personality, 38:481–504. [find](https://scholar.google.com/scholar?q=B%2E+Donnellan%2C+R%2E+D%2E+Conger%2C+and+C%2E+M%2E+Bryant%2E+2004%2E+The+Big+Five+and+enduring+marriages%2E+J%2E+of+Research+in+Personality%2C+38%3A481%E2%80%93504%2E)
5. Y. Freund, R. Iyer, R. E. Schapire, and Y. Singer. 1998. An efficient boosting algorithm for combining preferences. In Proc. of the 15th ICML, p. 170–178. [find](https://scholar.google.com/scholar?q=Y%2E+Freund%2C+R%2E+Iyer%2C+R%2E+E%2E+Schapire%2C+and+Y%2E+Singer%2E+1998%2E+An+efficient+boosting+algorithm+for+combining+preferences%2E+In+Proc%2E+of+the+15th+ICML%2C+p%2E+170%E2%80%93178%2E)
6. D. Funder and C. Sneed. 1993. Behavioral manifestations of personality: An ecological approach to judgmental accuracy. J. of Personality and Social Psychology, 64(3):479–490. [find](https://scholar.google.com/scholar?q=D%2E+Funder+and+C%2E+Sneed%2E+1993%2E+Behavioral+manifestations+of+personality%3A+An+ecological+approach+to+judgmental+accuracy%2E+J%2E+of+Personality+and+Social+Psychology%2C+64%283%29%3A479%E2%80%93490%2E)
7. A. Furnham, 1990. Handbook of Language and Social Psychology, chapter Language and Personality. Winley. [find](https://scholar.google.com/scholar?q=A%2E+Furnham%2C+1990%2E+Handbook+of+Language+and+Social+Psychology%2C+chapter+Language+and+Personality%2E+Winley%2E)
8. A. J. Gill and J. Oberlander. 2002. Taking care of the linguistic features of extraversion. In Proc. of the 24th Annual Conference of the Cognitive Science Society, p. 363–368. [find](https://scholar.google.com/scholar?q=A%2E+J%2E+Gill+and+J%2E+Oberlander%2E+2002%2E+Taking+care+of+the+linguistic+features+of+extraversion%2E+In+Proc%2E+of+the+24th+Annual+Conference+of+the+Cognitive+Science+Society%2C+p%2E+363%E2%80%93368%2E)
9. Agreeableness with all Nphon ≥ 2.66 Tentat ≥ 2.83 Colon ≥ 0.03 Posemo ≥ 2.67 Voiced ≥ 584 Relig ≥ 0.43 Insight ≥ 2.09 Prompt ≥ 0.06 Comma ≥ 4.60 Money ≥ 0.38 Fam ≥ 601.61 Swear ≥ 0.41 Anger ≥ 0.92 Time ≥ 3.71 Negate ≥ 3.52 Fillers ≥ 0.54 Time ≥ 3.69 Swear ≥ 0.61 Swear ≥ 0.45 WPS ≥ 6.13 Conscientious… [find](https://scholar.google.com/scholar?q=Agreeableness+with+all+Nphon+%E2%89%A5+2%2E66+Tentat+%E2%89%A5+2%2E83+Colon+%E2%89%A5+0%2E03+Posemo+%E2%89%A5+2%2E67+Voiced+%E2%89%A5+584+Relig+%E2%89%A5+0%2E43+Insight+%E2%89%A5+2%2E09+Prompt+%E2%89%A5+0%2E06+Comma+%E2%89%A5+4%2E60+Money+%E2%89%A5+0%2E38+Fam+%E2%89%A5+601%2E61+Swear+%E2%89%A5+0%2E41+Anger+%E2%89%A5+0%2E92+Tim)
10. F. Heylighen and J.-M. Dewaele. 2002. Variation in the contextuality of language: an empirical measure. Context in Context, Special issue of Foundations of Science, 7:293–340. [find](https://scholar.google.com/scholar?q=F%2E+Heylighen+and+J%2E%2DM%2E+Dewaele%2E+2002%2E+Variation+in+the+contextuality+of+language%3A+an+empirical+measure%2E+Context+in+Context%2C+Special+issue+of+Foundations+of+Science%2C+7%3A293%E2%80%93340%2E)
11. R. Hogan, G. J. Curphy, and J. Hogan. 1994. What we know about leadership: Effectiveness and personality. American Psychologist, 49(6):493–504. [find](https://scholar.google.com/scholar?q=R%2E+Hogan%2C+G%2E+J%2E+Curphy%2C+and+J%2E+Hogan%2E+1994%2E+What+we+know+about+leadership%3A+Effectiveness+and+personality%2E+American+Psychologist%2C+49%286%29%3A493%E2%80%93504%2E)
12. O. P. John and S. Srivastava. 1999. The Big Five trait taxonomy: History, measurement, and theoretical perspectives. In L. A. Pervin and O. P. John, editors, Handbook of personality theory and research. New York: Guilford Press. [find](https://scholar.google.com/scholar?q=O%2E+P%2E+John+and+S%2E+Srivastava%2E+1999%2E+The+Big+Five+trait+taxonomy%3A+History%2C+measurement%2C+and+theoretical+perspectives%2E+In+L%2E+A%2E+Pervin+and+O%2E+P%2E+John%2C+editors%2C+Handbook+of+personality+theory+and+researc)
13. J. Liscombe, J. Venditti, and J. Hirschberg. 2003. Classifying subject ratings of emotional speech using acoustic features. In Proc. of Eurospeech - Interspeech 2003, p. 725–728. [find](https://scholar.google.com/scholar?q=J%2E+Liscombe%2C+J%2E+Venditti%2C+and+J%2E+Hirschberg%2E+2003%2E+Classifying+subject+ratings+of+emotional+speech+using+acoustic+features%2E+In+Proc%2E+of+Eurospeech+%2D+Interspeech+2003%2C+p%2E+725%E2%80%93728%2E)
14. M. R. Mehl, S. D. Gosling, and J. W. Pennebaker. In press. Personality in its natural habitat: Manifestations and implicit folk theories of personality in daily life. J. of Personality and Social Psychology. [find](https://scholar.google.com/scholar?q=M%2E+R%2E+Mehl%2C+S%2E+D%2E+Gosling%2C+and+J%2E+W%2E+Pennebaker%2E+In+press%2E+Personality+in+its+natural+habitat%3A+Manifestations+and+implicit+folk+theories+of+personality+in+daily+life%2E+J%2E+of+Personality+and+Social+Psyc)
15. W. T. Norman. 1963. Toward an adequate taxonomy of personality attributes: Replicated factor structure in peer nomination personality rating. J. of Abnormal and Social Psychology, 66:574–583. [find](https://scholar.google.com/scholar?q=W%2E+T%2E+Norman%2E+1963%2E+Toward+an+adequate+taxonomy+of+personality+attributes%3A+Replicated+factor+structure+in+peer+nomination+personality+rating%2E+J%2E+of+Abnormal+and+Social+Psychology%2C+66%3A574%E2%80%93583%2E)
16. S. Nunn. 2005. Preventing the next terrorist attack: The theory and practice of homeland security information systems. J. of Homeland Security and Emergency Management, 2(3). [find](https://scholar.google.com/scholar?q=S%2E+Nunn%2E+2005%2E+Preventing+the+next+terrorist+attack%3A+The+theory+and+practice+of+homeland+security+information+systems%2E+J%2E+of+Homeland+Security+and+Emergency+Management%2C+2%283%29%2E)
17. J. W. Pennebaker and L. A. King. 1999. Linguistic styles: Language use as an individual difference. J. of Personality and Social Psychology, 77:1296–1312. [find](https://scholar.google.com/scholar?q=J%2E+W%2E+Pennebaker+and+L%2E+A%2E+King%2E+1999%2E+Linguistic+styles%3A+Language+use+as+an+individual+difference%2E+J%2E+of+Personality+and+Social+Psychology%2C+77%3A1296%E2%80%931312%2E)
18. J. W. Pennebaker, L. E. Francis, and R. J. Booth, 2001. LIWC: Linguistic Inquiry and Word Count. [find](https://scholar.google.com/scholar?q=J%2E+W%2E+Pennebaker%2C+L%2E+E%2E+Francis%2C+and+R%2E+J%2E+Booth%2C+2001%2E+LIWC%3A+Linguistic+Inquiry+and+Word+Count%2E)
19. B. Reeves and C. Nass. 1996. The Media Equation. University of Chicago Press. [find](https://scholar.google.com/scholar?q=B%2E+Reeves+and+C%2E+Nass%2E+1996%2E+The+Media+Equation%2E+University+of+Chicago+Press%2E)
20. K. R. Scherer. 1979. Personality markers in speech. In K. R. Scherer and H. Giles, editors, Social markers in speech, p. 147–209. Cambridge University Press. [find](https://scholar.google.com/scholar?q=K%2E+R%2E+Scherer%2E+1979%2E+Personality+markers+in+speech%2E+In+K%2E+R%2E+Scherer+and+H%2E+Giles%2C+editors%2C+Social+markers+in+speech%2C+p%2E+147%E2%80%93209%2E+Cambridge+University+Press%2E)
21. S. Tucker and S. Whittaker. 2004. Accessing multimodal meeting data: Systems, problems and possibilities. Lecture Notes in Computer Science, 3361:1–11. [find](https://scholar.google.com/scholar?q=S%2E+Tucker+and+S%2E+Whittaker%2E+2004%2E+Accessing+multimodal+meeting+data%3A+Systems%2C+problems+and+possibilities%2E+Lecture+Notes+in+Computer+Science%2C+3361%3A1%E2%80%9311%2E)
22. M. Walker and S. Whittaker. 1990. Mixed initiative in dialogue: an investigation into discourse segmentation. In Proc. of the 28th Annual Meeting of the ACL, p. 70–78. [find](https://scholar.google.com/scholar?q=M%2E+Walker+and+S%2E+Whittaker%2E+1990%2E+Mixed+initiative+in+dialogue%3A+an+investigation+into+discourse+segmentation%2E+In+Proc%2E+of+the+28th+Annual+Meeting+of+the+ACL%2C+p%2E+70%E2%80%9378%2E)

*Updated 24 October 2026*