Improving emotion recognition is associated with subsequent mental health and well-being in children with severe behavioural problems

Impaired emotion recognition is a transdiagnostic risk factor for a range of psychiatric disorders. It has been argued that improving emotion recognition may lead to improvements in behaviour and mental health, but supportive evidence is limited. We assessed emotion recognition and mental health following a brief and targeted computerised emotion recognition training in children referred into an intervention program because of severe family adversity and behavioural problems (n = 62; aged 7–10). While all children continued to receive their usual interventions, only children impaired in emotion recognition (n = 40) received the emotion training. Teachers blind to whether or not children had received the training rated children’s mental health problems before and 6 months after the training. Participants who received the emotion training significantly improved their recognition of negative and neutral facial expressions. Although both groups showed improved behaviour at follow-up, the reduction in behavioural problems was only significant in children who received the emotion training. Post-training emotion recognition scores predicted mental health problems 6 months later independently of initial emotion recognition ability and severity of behavioural problems. The results are consistent with the view that targeting emotion recognition can improve longer term functioning in individuals with disruptive behaviour, although further research using fully randomised designs is needed before causal conclusions can be drawn with confidence. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1007/s00787-020-01652-y) contains supplementary material, which is available to authorized users.