Figure 3
From: Active vision in immersive, 360° real-world environments
Active viewing increases top-down attentional allocation. Estimated marginal effects of salience and meaning maps on predicting overt attention in each condition (active vs. passive). We found that viewing condition (active vs. passive) significantly modulated gaze behavior (salience*meaning*condition: p < 0.001). Specifically, active viewers disproportionately directed their attention to meaningful, over salient, scene regions. Individual points represent random item effects (i.e., individual scenes). Error bars represent prediction intervals (± 1 STE). *** denotes p < 0.001.
