You can track but not detect: How attention alters change-detection rates in MOT

Introduction The ability to monitor a number of moving objects is often tested with the Multiple Object Tracking (MOT) task. For this task it is assumed that attention is object-based rather than space-based (Cavanagh & Alvarez, 2005), meaning that attention is linked to the targets that need to be tracked. Recently, it could be shown that peripheral vision, and thereby presumably covert attention, is used for target monitoring and target-change detection (Vater, Kredel, & Hossner, 2016a), and that target-change detections are affected by perceptual characteristics of the change (Vater, Kredel, & Hossner, 2016b). Since attention seems to be linked to targets, we investigated in the current study how attention affects detection rates, while higher detection rates were expected for target than for non-target changes. Methods 12 participants were tested in a MOT task that requires to monitor 4 out of 10 moving squares for 6 s. As secondary change-detection task, participants had to press a button as soon as one of the 10 squares stopped moving or changed its form (from square to diamond). Beside the change type and change location, object-change eccentricities were introduced as a third independent variable. Changes occurred either near ( 15 °) from the centroid (center of mass of the four targets). Participants’ gaze behavior was measured with an integrated mobile eye-tracking system (EyeSeeCam, 220 Hz) to calculate the percentage of peripheral detections (i.e., gaze was not on the change-object between change-onset and button press). Change-detection rates as well as peripheral detections were analyzed with a change location (target vs. distractor) x eccentricity (near vs. far) x change type (stop vs. form) ANOVA with repeated measures on all factors. Results Main effects for change location, eccentricity and change type were observed (all ps .70). Target changes were more often detected than distractor changes, stop changes were better detected than form changes and near changes were better detected than far changes. A three-way interaction was also revealed (p .10, ηp2 < .27) and was overall on average at 90.9 %. Discussion/Conclusion The current results highlight the influence of attention on detection rates of object changes. The current results extent those obtained by Vater et al. (2016), because decreased detection rates at far eccentricities decrease not only as a function of change type but also as a function of attention allocated to change-objects. Since distractor changes were missed more often, attention seems to be indeed linked to objects rather than space (Cavanagh & Alvarez, 2005). References Cavanagh, P., & Alvarez, G. A. (2005). Tracking multiple targets with multifocal attention. Trends in Cognitive Sciences, 9, 349-354. Vater, C., Kredel, R., & Hossner, E. J. (2016a). Detecting single-target changes in multiple object tracking: The case of peripheral vision. Attention, Perception, & Psychophysics, 78, 1004-1019. Vater, C., Kredel, R., & Hossner, E. J. (2016b). Detecting target changes in multiple object tracking with peripheral vision: More pronounced eccentricity effects for form than for motion changes. Manuscript submitted for publication