Vestibular and vestibulo-proprioceptive perception of motion in the horizontal plane in blindfolded man--III. Route inference.

The inference of direction and length of a subsequent route during performance of a triangle completion task was studied in blindfolded human subjects. Subjects were transported and walked with guidance along two sides of left- and right-oriented isosceles triangles. Subjects had to walk without assistance along the inferential bases of triangles back to a starting point. The influence of two variables in the triangular trajectory--the side length (2, 3 and 4 m) and the angle between sides (30-150 degrees--on route inference accuracy was investigated. Changes in the length of the movement trajectory with retained configuration lead to an alteration in the linear but not the angular estimation of the inferred route. Changes in the configuration of equidistant movement trajectories result in an alteration in both the linear and angular estimations of the inferred route. Estimations of direction and length of inferred routes following passive transportation and those following walking along triangular sides showed similar degrees of accuracy. When the inference of a route was regarded as a geometrical sum of subjective angular and linear estimations of displacements, trajectories of backward paths could be predicted. The results obtained show that the vestibular cue can only provide a gross orientation when moving along a complex trajectory in the horizontal plane; the proprioceptive cue does not improve accuracy adequately.