This paper presents a novel interface to support video coding of social attention in the assessment of children with autism spectrum disorder. Video-based evaluations of social attention during therapeutic activities allow observers to find target behaviors while handling the ambiguity of attention. Despite the recent advances in computer vision-based gaze estimation methods, fully automatic recognition of social attention under diverse environments is still challenging. The goal of this work is to investigate an approach that uses automatic video analysis in a supportive manner for guiding human judgment. The proposed interface displays visualization of gaze estimation results on videos and provides GUI support to allow users to facilitate agreement between observers by defining social attention labels on the video timeline. Through user studies and expert reviews, we show how the interface helps observers perform video coding of social attention and how human judgment compensates for technical limitations of the automatic gaze analysis.

Atypical gaze behavior in response to a face has been well documented in individuals with autism spectrum disorders (ASDs). Children with ASD appear to differ from typically developing (TD) children in gaze behavior for spoken and dynamic face stimuli but not for nonspeaking, static face stimuli. Furthermore, children with ASD and TD children show a difference in their gaze behavior for certain expressions. However, few studies have examined the relationship between autism severity and gaze behavior toward certain facial expressions. The present study replicated and extended previous studies by examining gaze behavior towards pictures of facial expressions. We presented ASD and TD children with pictures of surprised, happy, neutral, angry, and sad facial expressions. Autism severity was assessed using the Childhood Autism Rating Scale (CARS). The results showed that there was no group difference in gaze behavior when looking at pictures of facial expressions. Conversely, the children with ASD who had more severe autistic symptomatology had a tendency to gaze at angry facial expressions for a shorter duration in comparison to other facial expressions. These findings suggest that autism severity should be considered when examining atypical responses to certain facial expressions.

This paper explores how human perceptions, actions, and interactions can be changed through an embodied and active experience of being a smaller person in a real-world environment, which we call an egocentric smaller person experience. We developed a wearable visual translator that provides the perspective of a smaller person by shifting the wearer's eyesight level down to their waist using a head-mounted display and a stereo camera module, while allowing for field of view control through head movements. In this study, we investigated how the developed device can modify the wearer's body representation and experiences based on a field study conducted at a nursing school and museums, and through lab studies. It was observed that the participants changed their perceptions, actions, and interactions because they are considered to have perceived themselves as being smaller. Using this device, designers and teachers can understand the perspectives of other people in an existing environment.

The present study used functional near-infrared spectroscopy (fNIRS) to measure 5-to 6-month-old infants’ hemodynamic response in the prefrontal cortex (PFC) to visual stimuli differing in saliency and social value. Nineteen Japanese 5-to 6-month-old infants watched video clips of Peek-a-Boo (social signal) performed by an anime character (AC) or a human, and hand movements without social signal performed by an AC. PFC activity of infants was measured by 22-channel fNIRS, while behaviors including looking time were recorded simultaneously. NIRS data showed that infants’ hemodynamic responses in the PFC generally decreased due to these stimuli, and the decrease was most prominent in the frontopolar (FP), covering medial PFC (MPFC), when infants were viewing Peek-a-Boo performed by an AC. Moreover, the decrease was more pronounced in the dorsolateral PFC (DLPFC) when infants were viewing Peek-a-Boo performed by an AC than by a human. Accordingly, behavioral data revealed significantly longer looking times when Peek-a-Boo was performed by an AC than by a human. No significant difference between Peek-a-Boo and non-Peek-a-Boo conditions was observed in either measure. These findings indicate that infants at this age may prefer stimuli with more salient features, which may be more effective in attracting their attentions. In conjunction with our previous findings on responses to self-name calling in infants of similar age, we hypothesize that the dynamic function of the MPFC and its vicinity (as part of default mode network: enhanced by self-focused stimuli, attenuated by externally focused stimuli), which is consistently observed in adults, may have already emerged in 5-to 6-month-old infants.

Interacting with toys and other people is fundamental for developing social communication skills. However, children with autism spectrum disorder (ASD) are characterized by having a significant impairment in social interaction, and often, this leads to a deficit in play skills. For this reason, methods of teaching play skills to young children with ASD have been well documented. Although previous studies have examined a variety of instructional strategies for teaching skills, few studies have evaluated the potential of using robotic devices. The purpose of the present study is to examine the effects of feedback made by colored lights and vibration via paired robotic devices, COLOLO, in social play with children with ASD. COLOLO is a system of paired spherical devices covered with soft fabric. All participants in this study were recruited as volunteers through the Department of Psychology at Keio University. The pilot study included three participants diagnosed with PDDs (5- to 6-year-old boys), and compared the conditions with and without automatic feedback from the devices (colored lights and vibration). The results indicated that the participants in the condition that included feedback from the devices had an increased number of manipulations of the toy and increased number of times looking at the therapist’s toy. In the experiment, a systematic replication of the pilot study was performed with three other participants diagnosed with PDDs (3- to 6-year-old boys), using an A-B-A-B design. Again, the results showed that in the condition with colored lights and vibration, the children increased their manipulations of the toy as well as looking at the therapist’s toy. These findings are discussed in terms of the potential of paired robotic devices as a method to facilitate social play for children with ASD.