Figure 1.
Predictions for integration of remembered landmarks and path integration.
We predicted that when path integration and remembered landmarks provide congruent information about the observer's location, then these sources of information are combined. In this case, the estimate of location is more precise and biased towards the more reliable source of information. If remembered landmarks are incongruent with path integration, then only path integration is used to estimate location.
Figure 2.
Images of the hallway with (A) normal viewing, (B) viewed through the blur goggles, and (C) as represented by the tactile map.
Figure 3.
Participants' weighting of path integration in the combined cue task.
Amount of weight (and 95% confidence interval) that participants gave to path integration in the combined cue task. Weights were computed from participants' estimates of their location in the hallway.
Figure 4.
Weight given to path integration by individual participants.
Amount of weight that individual subjects gave to location information obtained by path integration when viewed and walked locations were perceived as congruent versus incongruent.
Figure 5.
Precision of estimates in the combined cue task.
The variability (root mean squared error) of localization estimates collapsed across viewing conditions when: 1) no landmarks, 2) remembered landmarks were judged to be incongruent with path integration information, and 3) remembered landmarks were judged to be congruent with path integration.