The Space-time Continuum: Temporal Segmentation Effects on Spatial Memory

Spatial navigation requires learning environments larger than what can be perceived at one time, requiring the integration of distinct episodes into one representation. Modern theories postulate hierarchical models of space, in which environments are mentally segmented, allowing more precise navigation within subregions and less precise strategies across them. But how does space become segmented into subregions? Segmentation, a component of many domains in human cognition, is the process by which a continuous stream of input is divided into chunks. In spatial navigation, segmentation often leverages natural divisions, like rivers or highways. Here, we hypothesize that temporal segmentation, a powerful organizer of information thought to support event cognition (Zacks & Swallow, 2007), could play a role in segmenting space as well. To test this hypothesis, in a carefully controlled virtual environment experiment, participants learned landmarks around a square virtual environment in two temporally separated sessions designed to elicit temporal divisions during learning while controlling other factors. We then queried participants’ spatial (and non-spatial) memories for evidence of temporal segmentation across multiple behavioral measures. We found a gradient of temporal segmentation effects: measures more likely to depend on time, like free recall and distance judgments, were more affected by temporal segmentation, while measures more likely to depend on space (pointing judgments and maps) were less affected. Results support the notion that temporal segmentation is task-dependent, with variable effects on spatial memory. We interpret our findings in the context of theories of hippocampal encoding of space and time.