Rotating objects cue spatial attention
via the perception of frictive surface contact
- . Rotating objects cue spatial attention via the perception of frictive surface contact. Cognition, 242, 105655. PDF
We report a new attentional cueing effect, which shows how attention models the physical force of friction. Most objects we see are in frictive contact with a ‘floor’, such that clockwise rotation causes rightward movement and counterclockwise rotation leftward movement. Is this regularity encoded in spatial orienting responses? In Experiment 1, seeing a clockwise-rotating ‘wheel’ produced faster responses to subsequent targets appearing on the right vs. left (and vice versa for counterclockwise rotation). Thus, when viewing a lone rotating wheel, we orient attention toward where we predict it will move next, assuming frictive floor contact. But what happens if the rotating wheel is seen touching another visible surface? In Experiment 2, rotational cueing was stronger for wheels touching a visible floor, was abolished for wheels near but not touching another surface, and reversed for wheels touching a ceiling. We conclude that the visual system makes an assumption of frictive floor contact, and rapidly analyzes visual cues to frictive contact with other surfaces, in order to orient attention toward where objects are likely to move next.
Demonstrations
Experiment 1 — Rotating objects cue spatial attention
When we see an object, it is usually sitting on a surface beneath it, and it is usually in frictive contact with that surface. If these physical regularities are wired into the operation of visual attention, then viewing an isolated rotating stimulus might automatically cue spatial attention in the direction the object would move if in frictive contact with a 'floor'. To test this, we asked subjects to complete a speeded letter identification task. We predicted that a preceding clockwise-rotating wheel might produce faster responses to targets appearing on the right (vs. left), and that a counterclockwise-rotating wheel might produce faster responses to targets appearing on the left (vs. right).
Spatial Cueing by a Rotating Object
Click/tap the bars to view example displaysAs you can see, Clockwise rotation produced faster response times to targets appearing on the right vs. left, and Counterclockwise rotation produced faster response times to targets on the left vs. right. Thus, a lone stationary rotating object orients spatial attention toward where it would normally move assuming contact with a frictive floor.
Experiment 2 — Reversal of the cueing effect by contact with a 'ceiling'
In Experiment 1, the rotating wheel was always presented alone, without any other surfaces visible. It nevertheless produced a strong spatial orienting effect, in the direction the wheel would move if it were in frictive contact with a floor beneath it. One possibility is that rotating objects always orient spatial attention in a way which assumes frictive contact with a floor. (After all, this assumption is consistent with the vast majority of objects that we see.) But another possibility is that, when we see a rotating object make visible contact with another surface, that this might modulate how the rotating object orients attention, in a way which is consistent with how the wheel *should* move, given frictive contact with that surface. In Experiment 2, we investigated how the rotation cueing effect (computed as [RT for letters on the side of the wheel incongruent with terrestrial-rolling] - [RT for letters on the side of the wheel congruent with terrestrial-rolling] - [RT for letters on the side of the wheel congruent with terrestrial-rolling]) is influenced by seeing the wheel touch another visible surface (a horizontal line). If the visual system assumes that contact between visible surfaces is frictive, then a wheel touching a visible surface might produce a stronger cueing effect than a wheel slightly separated from that surface. Moreover, the cueing effect might reverse for a rotating wheel seen touching a 'ceiling' (imagine painting your ceiling with a paint roller; here, clockwise rotation is associated with leftward movement).
Effect of Visible Surface Contact on Rotation Cueing
Click/tap the bars to view example displaysAs you can see, in the Floor-Touching condition, subjects again responded faster to letters located congruently with the direction the wheel would move if in frictive contact with a floor vs. located incongruently. In contrast, in the Floor-Not Touching condition, the cueing effect was abolished, with equally fast RTs for letters located congruently with the direction the wheel would move if in frictive contact with a floor vs. located incongruently. Strikingly, on Ceiling-Touching trials, the cueing pattern reversed! Now subjects responded faster to targets that were incongruent with terrestrial rolling vs. congruent. On Ceiling-Not Touching trials, there was no reliable cueing effect. From these results, we conclude that, when other surfaces are visible rotating objects orient attention in a way that integrates information about whether objects appear to be in contact with those surfaces.