Related Experiment Video
Updated: Jul 30, 2026

The Double-H Maze: A Robust Behavioral Test for Learning and Memory in Rodents
Published on: July 8, 2015
Retrospectively and prospectively modulated hippocampal place responses are differentially distributed along a common
Julien Catanese1, Alessandro Viggiano2, Erika Cerasti2
1College de France, Center for Interdisciplinary Research in Biology, Paris F-75005, France, and CNRS, UMR 7241, Paris F-75005, France catanese.julien@gmail.com sidney.wiener@college-de-france.fr.
Abstract:
Hippocampal place responses can be prospectively or retrospectively modulated by the animal's future or prior trajectory. Two main hypotheses explain this. The "multiple-map hypothesis" switches between different maps for different trajectories (rate remapping). In contrast, in the "buffer hypothesis," the hippocampus encodes an ongoing representation that includes the recent past and/or the impending future choice. This study examines the distribution of prospective and retrospective responses distributed along a common path in a continuous T-maze (providing all four combinations of provenance and destination) during a visual discrimination task. The multiple-map hypothesis predicts either uniform distributions or concerted shifts about a task-decision relevant point, whereas the buffer hypothesis predicts a time-limited overexpression around choice points (with retrospective responses after the central arm entry point and prospective responses nearer its exit). Here bilateral recordings in the dorsal CA1 region of the rat hippocampus show that retrospective responses were twice as prevalent as prospective responses. Furthermore, retrospective and prospective modulations have distinct spatial distributions, with retrospective primarily in the first two-thirds of the central arm and prospective restricted to the last third. To test for possible trial-by-trial remapping in relation to the two-thirds transition point, data from the first and second halves of the sessions were compared. Backward drift of path-modulated activity was significant only for retrospective, but not prospective, fields. Thus, these data are more consistent with the buffer hypothesis. Retrospective and prospective modulation would then participate in a single hippocampal representation of spatial and behavioral context.

