The Role Of Spatial Configuration In Multiple Identity Tracking

People often need to simultaneously track and identify multiple moving objects. In this type of tasks observers must be able to correctly bind object identities to continuously changing spatiotemporal locations. For example, when watching a soccer game, observers are required to be aware of the positions of each player in any given moment. The perception of this binding relationship provides the basis for humans to interact with the outside world. Previous studies have shown that approximately3or4identity-location bindings can be kept in VWM (visual working memory). However, it’s yet unclear that how these multiple bindings are represented in VWM.In the present study, we deployed the multiple objects tracking method to investigate the representation of multiple identity-location bindings. In Experiment1, in order to examine whether identity-location bindings are based on the spatial configuration, we manipulated the spatial configurations of all targets (i.e. whether the configuration collapse was possible or not) during tracking. The result showed that, the change of the spatial configuration significantly impacted the performance of the identifying task, suggesting that multiple id entity-location bindings are based on spatial configuration. In Experiment2, in order to further explore the mechanism of multiple identity tracking, we used probe-dot detection technique to investigate the division of attention among all targets at the time the configuration collapses. The result revealed that the critical target that violated the coherence of the spatial configuration got more attention than other targets did, indicating that the change of configuration affects the division of attention among different targets, and the critical target holds a higher priority. In Experiment3, the spatial configuration composed by all objects was manipulated, aiming to further investigate the constitution of the configuration. That is, whether the configuration was formed by all objects (including targets and distracters) or instead it consisted of targets only? The results demonstrated that whether the distracter violated the configuration or not did not impact the performance of the identify task, implying that distracters can not affect the representation of the spatial configuration.In short, the main findings of this study are as follows:(1) Multiple identity-location bindings are based on spatial configuration;(2) The maintenance of the configuration is depended upon top-down attention projection, and the critical target that violates the coherence of the configuration are projected with more attention.(3) The configuration consists of targets only, whereas distracters have no impact on the representation of the bindings.