TY - RPRT TI - Visual Homing is possible without Landmarks: A Path Integration Study in Virtual Reality AU - Riecke, B. E. AU - van Veen, H. A. H. C AU - Bülthoff, H. H AB - The literature often suggests that proprioceptive and especially vestibular cues are required for navigation and spatial orientation tasks involving rotations of the observer. To test this notion, we conducted a set of experiments in virtual reality where only visual cues were provided. Subjects had to execute turns, reproduce distances or perform triangle completion tasks: After following two prescribed segments of a triangle, subjects had to return directly to the unmarked starting point. Subjects were seated in the center of a half-cylindrical 180$^\circ$ projection screen and controlled the visually simulated ego-motion with mouse buttons. Most experiments were performed in a simulated 3D field of blobs providing a convincing feeling of self-motion (vection) but no landmarks, thus restricting navigation strategies to path integration based on optic flow. Other experimental conditions included salient landmarks or landmarks that were only temporarily available. Optic flow information alone proved to be sufficient for untrained subjects to perform turns and reproduce distances with negligible systematic errors, irrespective of movement velocity. Path integration by optic flow was sufficient for homing by triangle completion, but homing distances were biased towards mean responses. Additional landmarks that were only temporarily available did not improve homing performance. However, navigation by stable, reliable landmarks led to almost perfect homing performance. Mental spatial ability test scores correlated positively with homing performance especially for the more complex triangle completion tasks, suggesting that mental spatial abilities might be a determining factor for navigation performance. Compared to similar experiments using virtual environments (Péruch et al., 1997, Bud, 2000) or blind locomotion (Loomis et al., 1993), we did not find the typically observed distance undershoot and strong regression towards mean turn responses. Using a virtual reality setup with a half-cylindrical 180$^\circ$ projection screen allowed us to demonstrate that visual path integration without any vestibular or kinesthetic cues is sufficient for elementary navigation tasks like rotations, translations, and homing via triangle completion. CN - 0008 DA - 2000/09// PY - 2000 PB - Max Planck Institute for Biological Cybernetics, Tübingen, Germany. [Avaliable: ftp://ftp.kyb.mpg.de/pub/mpi-memos/pdf/TR-082.pdf] SN - 82 KW - MPI-TR KW - Riecke KW - Riecke_2000 KW - TR KW - bernhard-riecke KW - iSpaceWeb ER -