Comparison of the effects of damage to the perirhinal and parahippocampal cortex on transverse patterning and location memory in rhesus macaques

Abstract

Monkeys with damage to the parahippocampal (TH/TF) or perirhinal (PRh) cortex were tested on two sets of the transverse patterning (TP) problem (A+/B-, B+/C-, C+/A- and D+/E-, E+/F-, F+/D-) and delayed nonmatching-to-location paradigm (DNML), with delays ranging from 10 to 600 s. Damage to either area impaired acquisition and performance of TP but not of linear discriminations (e.g., A>B>C>X). Damage to areas TH/TF impaired performance of the DNML at all delays but only affected memory for objects at the longest delay, as measured by a delayed nonmatching-to-sample task (DNMS) (Nemanic et al., 2004). Damage to the PRh impaired performance of the DNMS but not of the DNML. The results present a dissociation in object and place memory for these two cortical regions and suggest a role for each in the cortical circuitry supporting configural/relational memory.

Figure 1.

Intended lesion and surface reconstructions of actual damage for group TH/TF. For detailed reconstruction on coronal sections and histological sections, see Nemanic et al. (2004), their Figures 7 and 8 (reproduced as supplemental material, available at www.jneurosci.org). Numerals on the right indicate the approximate distance in millimeters from the interaural plane. amt, Anterior middle temporal sulcus; ERh, entorhinal cortex; pmt, posterior middle temporal sulcus; rh, rhinal sulcus; ot, occipitotemporal sulcus; TE, TEO, TH, and TF, cytoarchitectonic fields as described by von Bonin and Bailey (1974).

Figure 2.

Intended lesion and surface reconstructions of actual damage for group PRh. For detailed reconstruction on coronal sections and histology of a representative case, see Nemanic et al. (2004), their Figures 4-6 (reproduced as supplemental material, available at www.jneurosci.org). sts, Superior temporal sulcus. All other abbreviations are as in Figure 1.

Figure 3.

Acquisition of transverse patterning phases 1-3 and transfer. Trials to criterion of three subjects in group N who received phase training (filled bars), group PRh (gray bars), and group TH/TF (open bars) are shown.

Figure 4.

Performance of transverse patterning averaged across the last five training sessions. Group N includes all seven subjects. A, Percentage of correct performance on each problem of transverse patterning, averaged across sets 1 and 2 (i.e., training problem 1 = averaged performance across the first 2 discriminations of each problem set). B, Left, Percentage of correct performance plotted by problem rank (i.e., the best, intermediate, and worst performance levels averaged across the 2 sets). Right, Performance as a function of problem rank on the transfer set of problems for groups TH/TF and PRh. Conventions are as in Figure 3.

Figure 5.

Performance on DNMS [data are from Nemanic et al. (2004)] and DNML, collapsed across the four delays. Filled squares, Group N; gray triangles, group TH/TF; open circles, group PRh.