Path integration absent in scent-tracking fimbria-fornix rats: evidence for hippocampal involvement in "sense of direction" and "sense of distance" using self-movement cues

Abstract

Allothetic and idiothetic navigation strategies use very different cue constellations and computational processes. Allothetic navigation requires the use of the relationships between relatively stable external (visual, olfactory, auditory) cues, whereas idiothetic navigation requires the integration of cues generated by self-movement and/or efferent copy of movement commands. The flexibility with which animals can switch between these strategies and the neural structures that support these strategies are not well understood. By capitalizing on the proclivity of foraging rats to carry large food pellets back to a refuge for eating, the present study examined the contribution of the hippocampus to the use of allothetic versus idiothetic navigation strategies. Control rats and fimbria-fornix-ablated rats were trained to follow linear, polygonal, and octagonal scent trails that led to a piece of food. The ability of the rats to return to the refuge with the food via the shortest route using allothetic cues (visual cues and/or the odor trail available) or using ideothetic cues (the odor trail removed and the rats blindfolded or tested in infrared light) was examined. Control rats "closed the polygon" by returning directly home in all cue conditions. Fimbria-fornix rats successfully used allothetic cues (closed the polygon using visual cues or tracked back on the string) but were insensitive to the direction and distance of the refuge and were lost when restricted to idiothetic cues. The results support the hypothesis that the hippocampal formation is necessary for navigation requiring the integration of idiothetic cues.

Fig. 1.

A, The foraging table showing the location of the refuge hole (●) and a string in one of its configurations with a food pellet located at its end. B, An example of the string that the rats tracked and food pellets that they carried back to the refuge. C, An example of a rat tracking along the string to the food pellet.

Fig. 2.

Photomicrographs of a control and lesion brain.Top, Cresyl violet sections shows the intact fimbria–fornix (FF) at the septal tip of the hippocampus (left) and an illustration of the absent fimbria–fornix (right). Bottom, Acetylcholinesterase sections through the hippocampus illustrating dense acetylcholinesterase staining in CA1 and dentate gyrus (left) and absence of staining in the FF lesion section (right). The slight cut in the left topof the cresyl violet section is an electrode tract.

Fig. 3.

Illustrations of the homeward trajectories (solid lines) of the control (top) and fimbria–fornix (FF) rats along a scented string in the blindfold condition. The dotted line is a nonscented string.

Fig. 4.

A, The rats traveled one of five linear distances to find a food pellet. The scent string was removed when they reached the food pellet. B, Errors as a function of distance on the return trip (mean and SE) consisted of a visit to an incorrect hole. C, Latencies to return home as a function of distance (mean and SE). FFB, Fimbria–fornix blindfold; FFV, fimbria–fornix vision;CV, control vision; CB, control blindfold.

Fig. 5.

Turning angles of rats after food retrieval as a function of distance traveled to reach the food. Note that the fimbria–fornix rats were inaccurate at all distances when blindfolded.

Fig. 6.

Turning strategies of control and fimbria–fornix rats on retrieving a food pellet when the scent string was removed. The control rat stretches forward to retrieve the food and then recoils backward and pivots to face the refuge. The fimbria–fornix rat walks over to the food and fails to pivot back to the return path.

Fig. 7.

A, Three polygon string tracks that the rats were required to follow. B, Time (mean and SEs on the three problems) taken to travel to the food (Out) and return (Back) in control and fimbria–fornix (FF) rats in mask and blindfold conditions. C, Distance traveled to reach the food by control and fimbria–fornix rats in masked and blindfolded conditions. The top portion of the gray bars(Out) represents the distance to the food along the string. The bottom portion of the gray bars (Back) represents the shortest distance from the food to the home refuge. Note that the performance of the groups is quite similar and approximates the string out and shortest back distance, with the exceptions of the return back performance of the fimbria–fornix rats when blindfolded.

Fig. 8.

A, Homeward trajectories for control and fimbria–fornix (FF) rats under room light and infrared light on one polygon string problem. Returns of fimbria–fornix rats were inaccurate in the dark, because they were largely following the scent string back to the refuge. Return latencies were mean and SEs. Note the significantly longer latencies of the fimbria–fornix rats under infrared light.

Fig. 9.

A, A string led from the refuge hole to an octagonal circuit on which there was no food. Once the rat was on the octagon, the string connecting it to the home cage was removed. B, Travel paths made by control and fimbria–fornix rats. Note that the control rats averaged one circle on the octagon before returning to the home cage, whereas the fimbria–fornix rats made many turns around the octagon.C, Time (mean and SE) spent traveling on the octagon for control and fimbria–fornix rats.

Fig. 10.

Paths and destinations of control and fimbria–fornix rats when the home cage was moved to a new location (shaded refuge). Note that in the mask condition in which the rats could see, most control rats and all fimbria–fornix rats carried the food to the old home location. In the blindfold condition in which the rats could not see, all control rats carried food to the new home location via a direct route, whereas fimbria–fornix rats reached other nearby holes or followed the string back to the new home location.