Language Models Can Generate Human-Like Self-Reports of Emotion

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

3 Citations (Scopus)
213 Downloads (Pure)


Computational interaction and user modeling is presently limited in the domain of emotions. We investigate a potential new approach to computational modeling of emotional response behavior, by using modern neural language models to generate synthetic self-report data, and evaluating the human-likeness of the results. More specifically, we generate responses to the PANAS questionnaire with four different variants of the recent GPT-3 model. Based on both data visualizations and multiple quantitative metrics, the human-likeness of the responses increases with model size, with the largest Davinci model variant generating the most human-like data.

Original languageEnglish
Title of host publication27th International Conference on Intelligent User Interfaces, IUI 2022 Companion
Number of pages4
ISBN (Electronic)978-1-4503-9145-0
Publication statusPublished - 22 Mar 2022
MoE publication typeA4 Conference publication
EventInternational Conference on Intelligent User Interfaces - Virtual, Online, Finland
Duration: 22 Mar 202225 Mar 2022
Conference number: 27

Publication series

NameInternational Conference on Intelligent User Interfaces


ConferenceInternational Conference on Intelligent User Interfaces
Abbreviated titleIUI
CityVirtual, Online


  • affect
  • emotion
  • GPT-3
  • Language models


Dive into the research topics of 'Language Models Can Generate Human-Like Self-Reports of Emotion'. Together they form a unique fingerprint.

Cite this