Fig. 4: Widespread gaze-dependent brain activity during movie viewing and recall.
From: Neural and behavioral reinstatement jointly reflect retrieval of narrative events
All figures show voxel-wise general linear model results estimated for gaze predictors modeling eye-movement amplitude (i.e., vector length between gaze positions measured or decoded for subsequent functional volumes). Statistical maps show group-level results of non-parametric, one-tailed, one-sample t-tests performed against zero overlaid on FreeSurfer’s FSaverage surface. Results are shown at liberal t-thresholds to show the spatial distribution of effects underlying the pattern in Fig. 5. A Results obtained for movie viewing using camera-based eye-tracking data (n = 13). Results obtained for movie viewing (B) and recall (C) using MR-based eye-tracking data (n = 16) decoded using DeepMReye49. Gaze-position changes correlate with brain activity in both tasks.
