Enhancing emotional communication between autistic and non-autistic individuals through assistive Information Technology

Abstract: Recognising people’s emotions is a promising research area in human-computer interaction as emotional communication plays a crucial role in humans’ lives. One of the main reasons for ineffective emotional communication is a deficit in understanding emotional signals such as facial expressions and body posture. There is a bidirectional challenge between autistic and non-autistic individuals since they display their emotional signals differently. This thesis discovers differences in emotional signals, in particular facial expressions, body posture, and physiological signals. Based on the interviews and questionnaires conducted in this thesis, the need to design an aid tool to assist autistic and non-autistic participants during their emotional communication is identified. Therefore, Emognition, a smartwatch, and its mobile application is designed to blur these differences and enhance the emotional communication between them. Furthermore, Emognition’s user evaluation indicates that the smartwatch could successfully detect nonautistic participants’ sadness and happiness. Also, they found the mobile application useful and aesthetically motivating to interact with. Even though we could not evaluate how well the Emognition recognises autistic participants’ sadness and happiness, it is promising to measure their emotions successfully by accurate sensors and, more importantly, by finding their autonomic response patterns to different emotions and enhance emotional communication between autistic and nonautistic people by Emognition.