Dr. Hirshfield’s research explores the use of non-invasive brain measurement to passively classify users’ social, cognitive, and affective states in order to enhance usability testing and adaptive system design. She works primarily with functional near-infrared spectroscopy (fNIRS), a relatively new non-invasive brain imaging device that is safe, portable, robust to noise, which can be implemented wirelessly; making it ideal for research in human-computer interaction. The high density fNIRS equipment in Hirshfield’s lab provides rich spatio-temporal data that is well suited as input into deep neural networks and other advanced machine learning algorithms. A primary tenet of Hirshfield’s machine learning research involves building and labeling large cross-participant, cross-task fNIRS training datasets in order to build robust and generalizable models that can avoid overfitting and succeed in ecologically valid environments outside the lab.
# Research interests