Facial Emotion Recognition Trainings for Children and Adolescents: A Transdiagnostic Systematic Review and Meta-Analysis

Abstract

Facial emotion recognition is a crucial aspect of social interaction and follows a typical developmental trajectory. However, youth with neurodevelopmental and psychological disorders often exhibit difficulties in this domain, making it a potential transdiagnostic intervention target. Extending a meta-analysis focused on autistic samples, this study systematically reviewed and compared the efficacy of facial emotion recognition training programmes for autistic and nonautistic children and adolescents between 4 and 18 years old. On 16 August 2024, PsychInfo, PubMed and Web of Science were searched for studies that (1) implemented a training to teach a clinical sample of youth how to accurately read and label facial emotions, (2) had a premeasure and postmeasure of facial emotion recognition for a training group and a control group and (3) were peer-reviewed and available in English. Twenty-nine studies met the inclusion criteria. Training characteristics, such as session intensity, facial stimuli, format and setting, were extracted and synthesized. A random-effects meta-analysis, sensitivity analysis and publication bias estimate were conducted. Although training characteristics varied across studies, meta-analytic results indicated that both autistic (SMD = 0.681, 95% CI [0.513, 0.848]) and nonautistic (SMD = 0.773, 95% CI [0.001, 1.535]) clinical groups demonstrated improvements in facial emotion recognition following training. Yet, few studies tested the long-term maintenance of these effects. These findings suggest that facial emotion recognition training can be efficacious across clinical populations, though they highlight the need for future research to examine how training characteristics affect training efficacy, long-term maintenance of improvements and social and clinical outcomes.