Abstract
Many experiments in human navigation have relied on measures of blind walking, where participants view targets and then walk to them without vision. Even without practice participants can carry out such tasks with high precision. This suggests that participants must possess an accurate calibration between the viewed distance of a target and the kinesthetic, proprioceptive and/orvestibular stimulation associated with walking to the target. We set out to explore the nature of this calibration using deception. Partipants were shown targets at a range of distances from 6 to 16 metres on an outdoor walkway. They were informed that they were to practice blind walking to visual targets and that they would receive an auditory signal when they had reached the target location. The participants were trained to walk to locations that were equal to, or 10% or 20% closer or further away than the viewed target location. In debriefing, very few (7.4%) subjects reported that they had noticed the mismatch between walked and viewed distances. Each subject received 21 training trials, following which they were presented with visual targets at 8, 12 and 16 m, in random order and they were asked to walk to the targets without vision. Our findings showed that even a brief period of training with a misleading calibration between visual and locomotor targets produced systematic and robust biases in blind walking in the predicted directions. These findings suggest a remarkable plasticity in the calibration between visual and locomotor space in the blind walking task and they underline the importance of practice effects in this commonly used task.
Research supported by a grant to CGE from the Natural Sciences and Engineering Research Council of Canada