The "tunnel task":: a method to examine cognitive strategies for spatial orientation

We report on the theoretical development and empirical testing of a new spatial cognition task, during which participants complete a tour through a virtual tunnel. Upon arrival they are asked to specify a ,,homing" vector by pointing an arrow towards the starting point of their virtual journey. Systematic experimental variations include number, length and sharpness of turns. Solving the task requires the development of a spatial representation, as alternative strategic cues (e.g., visual landmarks) are not available. Performance measures discriminate between general training effects and specific task variables. As an example, "side errors" indicate a loss of orientation and "position hits,, reflect the accuracy of the spatial representation. Participants use either an egocentric or allocentric frame of reference. The task does not imply the use of egocentric vs. allocentric spatial co-ordinates, yet it allows for a reliable a-priori identification of individually preferred strategies. Our data indicate that the spatial representations formed by both groups are functionally equivalent. However, the groups differ with respect to specific patterns of orientation errors.