Empathy training for counselling novices: A randomized controlled trial using machine learning and natural language processing.

Abstract

Objective: Empathy is a critical skill for effective counselling, yet novice counsellors often struggle to develop it. Traditional training methods may not sufficiently address the complexities of empathic development. This study aims to develop and assess an empathy training programme using machine learning (ML) and natural language processing (NLP) to enhance empathy in novice counsellors.

Methods: We developed the 'Empathy Feedback' WeChat mini-programme to provide real-time feedback through counselling exercises. In a randomized controlled trial (RCT), 143 participants were randomly assigned to one of four groups: control, feedback, practice or feedback-practice. Over 5 days, participants completed daily empathy exercises through the mini-programme. Empathy levels, as measured by the Helpful Responses Questionnaire (HRQ), and self-efficacy were assessed immediately post-intervention and at a 2-week follow-up.

Results: Significant improvements in empathy and HRQ scores were observed in the experimental groups compared to the control group (p < .05), but no changes were found in self-efficacy scores.

Conclusions: The AI-driven empathy training programme effectively enhanced empathic skills among novice counsellors, demonstrating the potential of integrating machine learning and NLP into counsellor education. The lack of change in self-efficacy suggests that additional strategies may be needed to boost counsellors' confidence in applying these skills in practice.