Towards Detecting Need for Empathetic Response in Motivational Interviewing

TL;DR

This paper introduces a novel approach for detecting the need for empathetic responses at each turn in motivational interviewing sessions, utilizing pre-trained language models and a labeller-detector framework to enhance real-time empathy detection.

Contribution

It proposes a new turn-level empathy detection task in motivational interviewing using a labeller-detector framework with pre-trained language models, advancing beyond session-level assessment.

Findings

Proposed a labeller-detector framework for turn-level empathy detection.

Developed strategies for extending the detector with additional inputs and multi-task learning.

Laid groundwork for real-time empathy detection in clinical settings.

Abstract

Empathetic response from the therapist is key to the success of clinical psychotherapy, especially motivational interviewing. Previous work on computational modelling of empathy in motivational interviewing has focused on offline, session-level assessment of therapist empathy, where empathy captures all efforts that the therapist makes to understand the client's perspective and convey that understanding to the client. In this position paper, we propose a novel task of turn-level detection of client need for empathy. Concretely, we propose to leverage pre-trained language models and empathy-related general conversation corpora in a unique labeller-detector framework, where the labeller automatically annotates a motivational interviewing conversation corpus with empathy labels to train the detector that determines the need for therapist empathy. We also lay out our strategies of extending the detector with additional-input and multi-task setups to improve its detection and explainability.