The rat retrosplenial cortex is required when visual cues are used flexibly to determine location

Abstract

The present study examined the consequences of retrosplenial cortex lesions in rats on two novel spatial tasks. In the first experiment, rats discriminated opposing room views from the same general location, along with their opposing directions of travel ('Perspective' task). Rats were trained with food rewards using a go/no-go design. Extensive retrosplenial cortex lesions involving both the granular and dysgranular areas impaired acquisition of this discrimination, which relied on distal visual cues. The same rats were then trained on a non-spatial go/no-go discrimination between different digging media. No lesion effect was apparent. In the final experiment, rats discriminated between two locations within a room ('Location' task) such that direction of travel at each location would be of less help in solving the problem. Both extensive retrosplenial lesions and selective dysgranular retrosplenial lesions impaired this Location task. These results highlight the importance of the retrosplenial cortex (areas 29 and 30), including the dysgranular cortex (area 30), for the effective use of distal visual cues to solve spatial problems. The findings, which help to explain the bias away from visual allocentric solutions that is shown by rats with retrosplenial cortex lesions when performing spatial tasks, also support the notion that the region assists the integration of different categories of visuospatial information.

Keywords: Cingulate cortex; Navigation; Orientation; Perspective; Spatial memory.

Fig. 1

Perspective task (Experiment 1). A schematic of the two trial types in the Perspective task, which was used to test viewpoint discrimination. Rats were rewarded (tick) when the cup was in one direction relative to the animal's starting position (upper panel), but not rewarded (cross) if the rat had to travel in the opposite direction to reach the cup (lower panel). Visual cues (various dashed lines) were attached to the walls to help them to be distinguished. (The apparatus, room and test arena are not to scale).

Fig. 2

Location task (Experiment 3). In the Location task animals were rewarded for digging when the arena was in one corner of the room, but not if the arena was in the opposite corner. The animal could not use room direction per se to solve the task, as both directions were equally rewarded and unrewarded. Various visual cues (dashed lines) were attached to the walls to allow them to be distinguished. (The apparatus, room and test arena are not to scale).

Fig. 3

(A) A series of coronal sections showing the cases with the largest and smallest lesions included in the combined dysgranular and granular retrosplenial lesion group (RScomb). Light grey represents the largest lesion, and dark grey the smallest. The numbers correspond to the distance behind bregma in mm . (B) Coronal NeuN sections showing the retrosplenial cortex (both hemispheres) in a sham surgery control rat (top), and a representative rat from the combined dysgranular and granular (RScomb) lesion group. The dashed lines show the limits of the retrosplenial cortex along with its granular and dysgranular sub-regions. The scale bar is 500 μm long. Abbreviations: cb, cingulum bundle; dys, dysgranular retrosplenial cortex; gran, granular retrosplenial cortex.

Fig. 4

(A) A series of coronal sections showing the cases with the largest and smallest lesions included in the dysgranular retrosplenial lesion group (RSdysg). Light grey represents the largest lesion, and dark grey the smallest. The numbers correspond to the distance behind bregma in mm . (B) Coronal NeuN sections showing the retrosplenial cortex (both hemispheres) in a sham surgery control rat (top) and a representative rat from the dysgranular (RSdysg) lesion group. The dashed lines show the limits of the retrosplenial cortex along with its granular and dysgranular sub-regions. The scale bar is 500 μm long. Abbreviations: cb, cingulum bundle; dys, dysgranular retrosplenial cortex; gran, granular retrosplenial cortex.

Fig. 5

Experiment 1—Perspective task (Cohort 1). Left: Graph showing acquisition of the Perspective discrimination task by Sham1 and combined granular and dysgranular retrosplenial (RScomb) lesion animals. Rats discriminated between two spatial views, each with their own associated direction of travel. Performance is shown as the cumulative difference in dig latencies between incorrect (‘no-go’) and correct (‘go’) trials. Error bars show the standard error of the mean. Right: Histograms showing the performance of the Sham1 group and RScomb groups on the probe trials administered following acquisition of the perspective discrimination task. ‘Original’ refers to performance on the final day of training. When visual cues were occluded (‘Curtain’) both groups’ performance fell. Neither group performed above chance in the dark with the curtain drawn (‘Dark’). Error bars show the standard error of the mean. Asterisk (*) shows group difference p < 0.05.

Fig. 6

Experiment 2—Digging cup discrimination (Cohort 1, left, and Cohort 2, right). The histograms show the mean cumulative latency difference scores between incorrect (‘no-go’) and correct (‘go’) trials for the two sessions of testing ± standard error of the mean. Neither dysgranular retrosplenial (RSdysg) nor combined granular and dysgranular retrosplenial (RScomb) lesions affected performance on this non-spatial go/no-go discrimination. Rats began to acquire the discrimination within the first session, leaving overall performance above chance.

Fig. 7

Experiment 3—Location task (Cohort 1). Graph showing acquisition of the Location task, where animals were rewarded for digging if the arena was in one corner of the room, but not in the opposite corner. Performance was measured as the cumulative difference between digging latencies on incorrect (‘no-go’) and correct (‘go’) trials. A higher difference score indicates superior discrimination. Although both groups performed significantly above chance level by the end of training, animals with combined lesions of the granular and dysgranular retrosplenial cortex (RScomb) showed retarded acquisition of this perspective discrimination task. Neither group was able to perform the same discrimination successfully in the dark. Error bars show the standard error of the mean.

Fig. 8

Experiment 3—Location task (Cohort 2). Graph showing acquisition of Location task, where rats were rewarded for digging in one corner of the room but not in the opposite corner. Performance is measured as the cumulative digging latency difference between incorrect (‘no-go’) and correct (‘go’) trials. A higher difference score indicates superior discrimination. Although both groups performed significantly above chance levels by the end of training, dysgranular retrosplenial lesioned animals (RSdysg) acquired this location discrimination task more slowly than the Sham2 animals. Error bars show the standard error of the mean.