Crowdsourcing experiments online allows for low-cost data gathering with large participant pools; however, collecting data online does not give researchers access to certain metrics. For example, physiological measures such as heart rate (HR) can provide high-resolution data about the physical, emotional, and mental state of the participant. We investigate and characterize the feasibility of gathering HR from videos of online participants engaged in single user and social tasks. We show that room lighting, head motion, and network bandwidth influence measurement quality, but that instructing participants in good practices substantially improves measurement quality. Our work takes a step towards online physiological data collection.
|Title of host publication||CHI '16 : Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems|
|Publisher||Association for Computing Machinery, Inc|
|Number of pages||6|
|Publication status||Published - 2016|