One of the more significant problems people with autism face is their trouble in social situations. Sometimes it can be particularly difficult for those with ASD to follow the tone and other cues of a conversation and interact in an appropriate and comfortable manner. This can lead to social anxiety or struggles in general communication, leading to research efforts such as those being conducted at the Massachusetts’ Institute of Technology’s Computer Science and Artificial Intelligence Laboratory (CSAIL) and its Institute of Medical Engineering and Science (IMES).
Graduate student Tuka Alhanai and Ph.D. student Mohammad Ghassemi are working to refine a wearable social coach that can detect the emotional tone of a conversation. By picking up body signals such as heart rate, skin temperature, and blood flow alongside variations in speech patterns, the program (housed in a Samsung Simband) can run the stimuli through algorithms that identify the difference between happy and sad stories participants told. Furthermore, it can identify small, 5-second segments of conversations as “positive”, “negative”, or “neutral”.
While there are technological and privacy concerns to advance before this program makes it to mainstream commercial use, the team runs it on the Samsung Simband to demonstrate its ability to run on consumer products – the Samsung Strategy and Innovation Center also provided funding for the research. In the future, the team is planning to collect data on a larger scale and with another commercially available product such as the Apple Watch.
Read more about this MIT research through the button below, and share your college experiences and stories through our CAN Do College Project!