# Large Language Model (LLM) and Human Performance in Child Investigative Interviewing Question Formulation Tasks

## TL;DR

This study compares how well large language models and humans perform in creating questions during child interviews, finding strengths and weaknesses in each.

## Contribution

The study introduces a novel comparison of LLMs and humans in dynamic and static child interview question formulation tasks.

## Key findings

- LLMs used fewer recommended and more non-recommended questions in dynamic interviews compared to professionals.
- GPT-4 outperformed psychologists in static tasks by using more invitations and fewer option-posing questions.
- LLMs excelled in controlled environments but struggled with adaptive dialogue in dynamic settings.

## Abstract

We compared the performance of large language models (LLMs) and humans with various levels of expertise in child investigative interviewing on tasks related to question formulation. Two tasks were employed: a static Interview Excerpt Task where participants (60 psychologists, 60 naive participants, GPT‐4, and Llama‐2) formulated follow‐up questions to 100 interview excerpts, and a dynamic Avatar Interviewing Task where participants (32 professionals, 32 students, and GPT‐4) conducted 10‐min interviews with AI‐driven child avatars. In the dynamic task, LLMs used fewer recommended questions (M = 8.69 vs. 18.75) and more non‐recommended questions (M = 17.69 vs. 6.81) than professionals. Conversely, in the static task, GPT‐4 outperformed psychologists, using more invitations (67.8% vs. 5.4%) and fewer option‐posing questions (3.7% vs. 31.4%). While LLMs demonstrated strong question formulation skills in controlled environments, they struggled with adaptive dialogs.