Described herein are systems and methods for detecting a physiological response based on thermal measurements while accounting for touching the face. In one embodiment, a system includes an inward-facing head-mounted thermal camera (CAM) that takes thermal measurements of a region of interest (THROI) on a user's face, and a sensor that provides measurements (M) indicative of times at which the user touches the region of interest (ROI). The system also includes a computer that detects the physiological response based on THROI and M. Optionally, the computer generates feature values based on THROI and M, and utilizes a model to detect, based on the feature values, the physiological response. Optionally, the model was trained based on samples, each including: (i) feature values generated based on previous THROI taken while M indicated touching the ROI, and (ii) a corresponding label indicative of an extent of the physiological response.