Awareness of gaze behaviour is limited: Insights from a novel tracking paradigm

When asked where they have previously looked, people rarely report their visual behaviour correctly. However, previous tasks have probed participants’ awareness of their gaze behaviour through explicit measures, like recognition, and it remains unclear whether poor awareness is also seen in implicit measures. We investigated this with a novel tracking paradigm in which participants first completed a visual search task while their eye movements were recorded. Next, participants completed a tracking task where they followed a moving red dot on the screen with their eyes. In Experiment 1, the dot replayed either their own previously recorded gaze position or that of another participant. We measured tracking performance by cross-correlating the previously recorded gaze position with the tracked positions. Participants were not significantly faster or more accurate in tracking their own eye movements compared to another participant’s. Experiment 2 tested this with a more extreme manipulation, in which participants tracked either unaltered or temporally reversed sequences of their own eye movements, which resulted in higher accuracy in the forward compared to the reversed condition. Finally, Experiment 3 examined whether performance would decrease when participants tracked a participant with very dissimilar gaze behaviour from their own. Tracking performance was similar when tracking their own scanpath compared to a dissimilar participant, but participants’ tracking latencies were shorter when tracking a participant with consistent scanning behaviour. Overall, our results suggest that though awareness of one’s eye movements is generally poor, extreme manipulations, like temporally reversing one’s eye movements, can influence tracking performance.