The contribution of motion detectors during multiple-object tracking

Poster Presentation: Sunday, May 18, 2025, 2:45 – 6:45 pm, Pavilion
Session: Motion: Models, neural mechanisms

Maryam Rezaei1, Remy Allard1; 1Universite de Montreal

Motion perception, an essential skill for human beings, relies on two motion processing systems: a low-level system mediated by early direction-selective neurons known as motion detectors, and a high-level system that attentively tracks the position of objects. The current study aimed to investigate the contribution of motion detectors during multiple-object tracking through two experimental manipulations: reverse-phi and stroboscopic motion. By reversing the contrast polarity at each object displacement, reverse-phi inverses the direction response of motion detectors. By introducing a temporal gap between each displacement (i.e., object disappears briefly before reappearing at a new position), stroboscopic motion can weaken the contribution of motion detectors. Five young participants were asked to track and identify four balls among eight identical ones bouncing around within a virtual three-dimensional cube. The maximum speed threshold at which the 4 target balls were successfully tracked were determined under fourteen conditions: 2 contrast polarities (i.e., reversed or not) X 7 temporal gaps (i.e., 0, 17, 33, 50, 67, 83, or 100 msec). A significant interaction between contrast polarity and temporal gap on maximum speed threshold was found (F(6, 24)=9.194, p<.001). Post hoc analyses revealed that with temporal gaps of 0 and 17 msec, the maximum speed threshold was significantly lower when the contrast polarity of the balls was reversed compared to when it was not (p<.05). With longer temporal gaps, however, no significant difference was observed between the two contrast polarity conditions. The results suggest that motion detectors considerably contribute to multiple-object tracking when the temporal gaps between object displacements are short (<33 msec). But for longer temporal gaps, no considerable contribution of motion detectors was observed suggesting that multiple-object tracking relied mainly on the higher-level, attention-based motion processing system.

Acknowledgements: This work was supported by grants from the Natural Sciences and Engineering Research Council of Canada (NSERC) and the Fond du Recherche du Québec-Nature et Technologie (FRQNT) to RA.