Seeing and hearing what has not been said; A multimodal client behavior classifier in Motivational Interviewing with interpretable fusion

TL;DR

This paper introduces a multimodal classifier for analyzing client utterances in Motivational Interviewing, integrating text, audio, and visual cues to improve classification accuracy and interpretability.

Contribution

It presents a novel multimodal classification approach for MI conversations, with an emphasis on interpretability and identifying key modalities influencing decisions.

Findings

High accuracy in classifying MI client utterances

Identification of most influential modalities in decision-making

Insights into multimodal interactions during therapy sessions

Abstract

Motivational Interviewing (MI) is an approach to therapy that emphasizes collaboration and encourages behavioral change. To evaluate the quality of an MI conversation, client utterances can be classified using the MISC code as either change talk, sustain talk, or follow/neutral talk. The proportion of change talk in a MI conversation is positively correlated with therapy outcomes, making accurate classification of client utterances essential. In this paper, we present a classifier that accurately distinguishes between the three MISC classes (change talk, sustain talk, and follow/neutral talk) leveraging multimodal features such as text, prosody, facial expressivity, and body expressivity. To train our model, we perform annotations on the publicly available AnnoMI dataset to collect multimodal information, including text, audio, facial expressivity, and body expressivity. Furthermore, we identify the most important modalities in the decision-making process, providing valuable insights into the interplay of different modalities during a MI conversation.