Facial emotion training as an intervention in autism spectrum disorder: A meta-analysis of randomized controlled trials

Individuals with autism spectrum disorder often have difficulty interpreting and regulating their own emotions, understanding the emotions expressed by others, and labeling emotions based on viewing the faces of others. Emotion reading training has been fronted as an intervention in autism spectrum disorders. Various paradigms have such as pictures of eyes, faces, animated objects, social scenes, and more have been adopted to help autism spectrum disorder patients to decode complex mental states and emotions. Consequently, several computer-based training paradigms have been developed to help improve these patients’ facial emotion recognition ability. These programs use an interactive instructional approach where a computer presents instructional materials and monitors specific abilities such as reading and problem-solving. These computer programs generally work by drawing attention to facial features. Also, computer and video-based techniques are specifically designed to create an engaging environment for training autism spectrum disorder patients. Therefore, facial emotion training programs are non-invasive and affordable.

Generalization extent may be an issue with these emotion training methods with inconsistent findings having been reported. Generalization is the ability to transfer a skill or behavior learned in one context or situation to another related or similar one. This appears to be problematic for autism spectrum disorder patients because they learn a skill or routine only in one location or use only one material. Therefore, it’s critical to understand whether improvements from targeted interventions using facial emotion training can generalize to other related or different tasks.

an An essential aspect of the current study was to uncover the robustness of generalization effects after facial emotion training.

Given this background, Qianqian Zhang, Renjing Wu, Siyu Zhu, Jiao Le Yuanshu Chen, Chunmei Lan, Dr. Shuxia Yao, Dr. Weihua Zhao, and led by Professor Keith Kendrick all at the University of Electronic Science and Technology of China conducted standard meta-analytic methods to investigate the effects of facial emotion training including generalization and maintenance. They aimed to provide summary information on existing facial emotion training intervention studies for children and adolescents with autism spectrum disorders and examine the outcomes of these studies to discern whether they provide positive benefits for autism spectrum disorders. Their research work is published in the journal Autism Research.

The research team examined the effects of facial emotion training in studies of 595 participants with autism spectrum disorders. They found that the interventions significantly improved emotion recognition for patients under training compared with controls. Their findings also revealed a medium effect size in post-training improvement, suggesting that facial emotion training could be an effective intervention for improving emotion recognition in autism spectrum disorder patients.

However, the researchers couldn’t find any evidence for training improvement in general social skills, although only a handful of studies have investigated this. On the other hand, they did find evidence of some generalization of training effects in similar and different contexts. However, further studies are needed to establish the robustness of these generalizations.

Computer-based interventions provide realistic multi-sensory interactions that motivate participants. They also provide consistent and repetitive approaches which children and parents can use at their pace, thereby creating intrinsic motivation and progress. Therefore, computer-based training programs could play an important role in helping autism spectrum disorder children to improve their facial emotion recognition skills. Further research is needed to determine whether this form of training can be effective in improving social skills more generally.