Delle Monache Sergio, Lacquaniti Francesco, Bosco Gianfranco
Neuroscience Section, Department of Systems Medicine, University of Rome "Tor Vergata", Via Montpellier, 1, 00133, Rome, Italy.
Exp Brain Res. 2015 Feb;233(2):359-74. doi: 10.1007/s00221-014-4120-9. Epub 2014 Oct 14.
Manual interceptions are known to depend critically on integration of visual feedback information and experience-based predictions of the interceptive event. Within this framework, coupling between gaze and limb movements might also contribute to the interceptive outcome, since eye movements afford acquisition of high-resolution visual information. We investigated this issue by analyzing subjects' head-fixed oculomotor behavior during manual interceptions. Subjects moved a mouse cursor to intercept computer-generated ballistic trajectories either congruent with Earth's gravity or perturbed with weightlessness (0 g) or hypergravity (2 g) effects. In separate sessions, trajectories were either fully visible or occluded before interception to enforce visual prediction. Subjects' oculomotor behavior was classified in terms of amounts of time they gazed at different visual targets and of overall number of saccades. Then, by way of multivariate analyses, we assessed the following: (1) whether eye movement patterns depended on targets' laws of motion and occlusions; and (2) whether interceptive performance was related to the oculomotor behavior. First, we found that eye movement patterns depended significantly on targets' laws of motion and occlusion, suggesting predictive mechanisms. Second, subjects coupled differently oculomotor and interceptive behavior depending on whether targets were visible or occluded. With visible targets, subjects made smaller interceptive errors if they gazed longer at the mouse cursor. Instead, with occluded targets, they achieved better performance by increasing the target's tracking accuracy and by avoiding gaze shifts near interception, suggesting that precise ocular tracking provided better trajectory predictions for the interceptive response.
已知手动拦截严重依赖于视觉反馈信息的整合以及基于经验对拦截事件的预测。在此框架内，注视与肢体动作之间的耦合也可能有助于拦截结果，因为眼球运动有助于获取高分辨率视觉信息。我们通过分析受试者在手动拦截过程中头部固定的眼动行为来研究这个问题。受试者移动鼠标光标以拦截计算机生成的弹道轨迹，这些轨迹要么符合地球引力，要么受到失重（0g）或超重（2g）效应的干扰。在不同的实验环节中，轨迹在拦截前要么完全可见，要么被遮挡，以强化视觉预测。受试者的眼动行为根据他们注视不同视觉目标的时间量和扫视总数进行分类。然后，通过多变量分析，我们评估了以下两点：（1）眼球运动模式是否取决于目标的运动规律和遮挡情况；（2）拦截表现是否与眼动行为相关。首先，我们发现眼球运动模式显著取决于目标的运动规律和遮挡情况，这表明存在预测机制。其次，根据目标是可见还是被遮挡，受试者将眼动行为与拦截行为进行了不同的耦合。对于可见目标，如果受试者更长时间地注视鼠标光标，他们的拦截误差会更小。相反，对于被遮挡的目标，他们通过提高目标的跟踪精度并避免在拦截附近进行注视转移来取得更好的表现，这表明精确的眼部跟踪为拦截反应提供了更好的轨迹预测。
