Monocular distance Perception of reaching and grasping

Indiana University Bloomington, Bloomington IN

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): We study space perception and reaching. How might dynamic monocular and binocular vision yield accurate and stable perception of the distance, size and shape of objects to guide reaching effectively? Distortions are found consistently in passive judgment studies. In our studies, participants moved their head to generate optic flow while they viewed a spherical target. Then, they reached to touch the front, left, right or back of the sphere. We found distortions and instability. We tested whether calibration by touching a target would correct distortions. Distance was corrected, but shape was not. Using a Virtual Reality (VR) lab, we now investigate the use of visual and haptic feedback to calibrate distance, size and shape perception. We first compare and confirm in the virtual environment the results obtained in actual environments and we establish the functional effects of accommodation/vergence and IPD variations in VR. We use VR to measure and then to eliminate the use of disparity matching and thus isolate binocular distance and shape perception with concurrent vision of a target during a reach. Participants will reach to visible targets with an invisible hand. We investigate whether distance and shape distortions reflect a single continuous transformation of reach space or are independent. We investigate whether visual feedback from a near object generalizes to a far object both without and with a visible intervening support surface. We test the calibration of newly discovered eye height scaled information about distance along a flat surface. We test how quickly errors appear after calibration is removed. Using distorted haptic feedback, we test the strength of the coupling between haptic feedback and visual information about distance, while varying the strength of visual information. Finally, we test whether calibration is limited to the limb used to obtain feedback or generalizes to another limb. Can two limbs simultaneously be calibrated in different ways by distorted haptic feedback? Next, we investigate 3 ways that accurate shape perception might be achieved. First, we test whether objects with planar surfaces might allow observers to use contour invariants to achieve correct shape. Second, we test whether a visible support surface yields accurate shape perception because it introduces wide angle structure and large internal depth structure. Third, we test whether feedback might be effective for slant perception as a form of shape perception that allows feedback about shape to be provided in a single reach.