Perirhinal cortex is necessary for acquiring, but not for retrieving object-place paired association

Abstract

Remembering events frequently involves associating objects and their associated locations in space, and it has been implicated that the areas associated with the hippocampus are important in this function. The current study examined the role of the perirhinal cortex in retrieving familiar object-place paired associates, as well as in acquiring novel ones. Rats were required to visit one of two locations of a radial-arm maze and choose one of the objects (from a pair of different toy objects) exclusively associated with a given arm. Excitotoxic lesions of the perirhinal cortex initially impaired the normal retrieval of object-place paired-associative memories that had been learned presurgically, but the animals relearned gradually to the level of controls. In contrast, when required to associate a novel pair of objects with the same locations of the maze, the same lesioned rats were severely impaired with minimal learning, if any, taking place throughout an extensive testing period. However, the lesioned rats were normal in discriminating two different objects presented in a fixed arm in the maze. The results suggest that the perirhinal cortex is indispensable to forming discrete representations for object-place paired associates. Its role, however, may be compensated for by other structures when familiar object-place paired associative memories need to be retrieved.

Figure 1.

Illustration of the radial arm maze and behavioral paradigms. (A) Phase 1: Two objects (Spider-Man and LEGO block) were presented on arms 3 and 5 in gray color. Only one of the objects was rewarded in arm 3 (Spider-Man) and arm 5 (LEGO block) irrespective of its locations in the choice platform. Possible configuration of objects and appropriate choices are provided for both arms. In each trial, only one arm was open in the maze and objects were available in that open arm. (B) Phase 2: For acquisition of novel object–place paired associations, a pair of new objects (Barney and Girl) was presented on arms 3 and 5. Possible locations of the objects are shown as in A. Each object was rewarded only in a particular arm (Barney in arm 3 and Girl in arm 5) irrespective of its location in the choice platform. (C) Phase 3: Illustration of the task using only one arm (arm 4) in the maze. Two new objects (Mr. Potatohead and Cylinder) were used and the Mr. Potatohead choice was rewarded regardless of its location in the choice platform.

Figure 2.

Histological verifications of perirhinal cortical lesions. (A) Representative sections from the PR-LES and PR-CTRL groups. The upper panel shows Nissl-stained sections and the lower panel shows the adjacent sections stained for myelin. (B) Serial, Nissl-stained sections presented from anterior to posterior directions (from −2.8 to −7.3 mm from bregma).

Figure 3.

Effects of perirhinal cortical lesions on behavioral performance. (A) Effects of the perirhinal cortical lesions on previously acquired paired-associative memory. Mean ± SEM. (B) Acquisition curves for 10 d after introducing the new objects. Mean ± SEM.

Figure 4.

Response biases during behavioral testing. (A) Response bias index (0 = no response bias; 1 = complete response bias) calculated across 8 d of testing for previously acquired paired associates. (B) Response bias index during 10 d of testing for the acquisition of novel object–place paired associates. Mean ± SEM.

Figure 5.

Choice latency. (A) Average latency to choosing an object across 8 d of testing. (B) Average latency during 10 d of acquisition of novel object–place paired associates. Mean ± SEM.

Figure 6.

Acquisition of simple object-discrimination task. Learning curves across 3 d. Mean ± SEM.