Comparing the performance of a large language model and naive human interviewers in interviewing children about a witnessed mock-event
The findings show that the LLM was competent in formulating questions that adhere to best practice guidelines while human interviewers asked more questions following up on the child responses in trying to find out what the children had witnessed. The results indicate LLMs could possibly be used to support child investigative interviewers. However, substantial further investigation is warranted to ascertain the utility of LLMs in more realistic investigative interview settings.
Authors
- Ophelia Zhang (ORCID: https://orcid.org/0000-0002-0258-3786)
- Julia Korkman (ORCID: https://orcid.org/0000-0002-8734-9453)
- Pekka Santtila (ORCID: https://orcid.org/0000-0002-0459-1309)
- David I. Shapiro‐Ilan (ORCID: https://orcid.org/0000-0002-5666-4848)
- Shumpei Haginoya (ORCID: https://orcid.org/0000-0001-6909-674X)
- Liisa Järvilehto
- Yanling Sun (ORCID: https://orcid.org/0009-0005-9616-9875)
- Haohai Pang
Institutions
- Meiji Gakuin University (JP)
- Åbo Akademi University (FI)
- New York University Shanghai (CN)
- Helsinki University Hospital (FI)
- Maastricht University (NL)
- The European Institute for Crime Prevention and Control, affiliated with the United Nations (FI)
- New York University (US)
- East China Normal University (CN)
Publication Details
- Journal
- PLoS ONE
- Published
- 2025-02-28
- DOI
- https://doi.org/10.1371/journal.pone.0316317
- Citations
- 5
- Primary Topic
- Memory Processes and Influences
- Type
- article
- Field-Weighted Citation Impact
- 7.16
Funders
- New York University Shanghai