Abstract:

Path integration refers to a type of navigation in which navigators integrate information regarding self-motion in order to update the spatial relationship between themselves and the starting point of their journey. Human path integration has been often assessed via a path completion task in which the participants travel along several segments and then attempt to directly return to the origin of an outbound path. Previous research has revealed the influence of non-sensory factors on human path integration, such as memory, previous experience, target knowledge, and route decision-making. It remains unclear, however, how individuals’ expectations regarding the outbound paths influence their return-to-origin behavior.

In the present study, we used head-mounted display virtual reality to present hallway mazes and provided different instructions to 3 groups of participants in order to manipulate their expectations concerning the correct homing distance (i.e., the Euclidean distance between the starting and ending points of an outbound path) before they performed the path completion task. Specifically, the 3 groups were informed that the number of segments of each outbound path was positively correlated, negatively correlated, or uncorrelated with the correct homing distance. In actuality, we used an orthogonal design to make the correct homing distance uncorrelated to the number of segments.

The results revealed that the participants exhibited less accurate return-to-origin responses when their expectations concerning the correct homing distance were violated, compared to when these expectations were confirmed by the actual experiences. When the participants expected the correct homing distance to be positively correlated with the number of segments, they showed even greater errors for more complex outbound paths than those who expected a negative correlation. That being said, it should also be noted that all participants’ return-to-origin responses were subject to the influence of the number of segments and correct homing distance, even though only one group of participants was instructed to have valid expectations.

Taken together, these results demonstrate the influence of expectation on human path integration, although only having valid expectations cannot eliminate the influence of path properties on path integration. Moreover, individuals may rely more on expectation for more complex outbound paths. These findings therefore highlight the important roles that non-perceptual factors play in human path integration, and demonstrate that human path integration is an adaptive and strategic process.

Key words: expectation, path integration, spatial navigation, virtual reality