The Parahippocampal Cortex and its Functional Connection with the Hippocampus are Critical for Nonnavigational Spatial Memory in Macaques

Abstract

The Hamilton Search Task (HST) is a test of nonnavigational spatial memory that is dependent on the hippocampus. The parahippocampal cortex (PHC) is a major route for spatial information to reach the hippocampus, but the extent to which the PHC and hippocampus function independently of one another in the context of nonnavigational spatial memory is unclear. Here, we tested the hypotheses that (1) bilateral pharmacological inactivation of the PHC would impair HST performance, and (2) that functional disconnection of the PHC and hippocampus by contralateral (crossed) inactivation would likewise impair performance. Transient inactivation of the PHC impaired HST performance most robustly with 30 s intertrial delays, but not when color cues were introduced. Functional disconnection of the PHC and hippocampus, but not separate unilateral inactivation of either region, also selectively impaired long-term spatial memory. These findings indicate a critical role for the PHC and its interactions with the hippocampus in nonnavigational spatial memory.

Keywords: hippocampus; intracerebral microinfusion; macaque; parahippocampal cortex; spatial memory.

Figure 1

HST schematic. An array of 8 identical boxes, each with a spring-loaded lid, containing food reinforcers.

Figure 2

MRI and histology verification of parahippocampal and hippocampal infusion sites. (A) An MRI scan of subject LO in vivo with contrast generated by tungsten electrodes placed above the target sites, used to calculate site coordinates for microinfusion. (B) Postmortem MRI scans for subjects AN and RA showing cannula tracks where microinfusions were placed in the PHC and hippocampus. These sites are mapped in Figure 3. (C) A thionin-stained section showing a cannula track into the hippocampal microinfusion site of subject OD. (D) A thionin-stained section showing a cannula track and minor cell loss at the parahippocampal microinfusion site in subject SL.

Figure 3

Confirmed parahippocampal and hippocampal sites for each subject. Based on the postmortem MRI scans and histology, the parahippocampal and hippocampal microinfusion sites for each monkey were mapped onto coronal section schematics of the brain (above). The PHC is shaded in gray in the right hemisphere, with boundaries described in the Methods. Subjects YO, LO, and SL each have one site in the PHC per hemisphere; subjects OD and AN each have one parahippocampal and 2 hippocampal sites per hemisphere; and subject RA has 2 parahippocampal and 2 hippocampal sites per hemisphere, with the anterior and posterior pairs of parahippocampal sites used on separate experimental sessions. The parahippocampal sites for each monkey are also mapped onto a schematic of the ventral surface of the brain (below) with markers scaled to indicate the approximate drug spread for each microinfusion. The PHC is shaded in gray bilaterally. In this schematic, only the anterior parahippocampal sites are shown for subject RA, as the posterior sites for this animal largely overlapped with those indicated for subjects YO, LO, SL, and OD.

Figure 4

HST performance after bilateral inactivation of the PHC. Animals (n = 6) performed the 30 s spatial, 1 s spatial, and 30 s color-cued variations of the task after bilateral microinfusion of KYNA or saline in the PHC. (A) The total number of trials required for the monkey to locate all 8 reinforcer locations, such that a perfect score would be 8 trials and random performance would average to 21 trials, represented by the dashed bar. (B) The repetition index, a measure of the frequency and severity of box opening errors, where chance performance of 51.4 is represented by the dashed bar. For all graphs, each symbol represents the average of an individual monkey’s performance under those conditions, error bars represent standard error to the mean (SEM), and ^**P < 0.01, ^****P < 0.0001.

Figure 5

HST performance after bilateral inactivation of the hippocampus. Animals (n = 6, including 3 published in Forcelli et al. 2014) performed the 30 s spatial, 1 s spatial, and 30 s color-cued variations of the task after bilateral microinfusion of KYNA or saline in the hippocampus. (A) The total number of trials required for the monkey to locate all 8 reinforcer locations, such that a perfect score would be 8 trials and random performance would average to 21 trials, represented by the dashed bar. (B) The repetition index, a measure of the frequency and severity of box opening errors, where chance performance of 51.4 is represented by the dashed bar. For all graphs, each symbol represents the average of an individual monkey’s performance under those conditions, error bars represent SEM, and ^**P < 0.01, ^****P < 0.0001.

Figure 6

HST performance after crossed-inactivation of the contralateral hippocampus and PHC. Animals (n = 3) performed the 30 s spatial, 1 s spatial, and 30 s color-cued variations of the task after simultaneous contralateral microinfusion of KYNA or saline in the hippocampus and PHC. (A) The total number of trials required for the monkey to locate all 8 reinforcer locations, such that a perfect score would be 8 trials and random performance would average to 21 trials, represented by the dashed bar. (B) The repetition index, a measure of the frequency and severity of box opening errors, where chance performance of 51.4 is represented by the dashed bar. For all graphs, each symbol represents the average of an individual monkey’s performance under those conditions, error bars represent SEM, and ^**P < 0.01.

Figure 7

HST performance after unilateral inactivation of the PHC. Animals (n = 3) performed the 30 s spatial, 1 s spatial, and 30 s color-cued variations of the task after unilateral microinfusion of KYNA or bilateral microinfusion of saline in the PHC. (A) The total number of trials required for the monkey to locate all 8 reinforcer locations, such that a perfect score would be 8 trials and random performance would average to 21 trials, represented by the dashed bar. (B) The repetition index, a measure of the frequency and severity of box opening errors, where chance performance of 51.4 is represented by the dashed bar. For all graphs, each symbol represents the average of an individual monkey’s performance under those conditions, error bars represent SEM, and significance was set at P < 0.05.

Figure 8

HST performance after unilateral inactivation of the hippocampus. Animals (n = 3) performed the 30 s spatial, 1 s spatial, and 30 s color-cued variations of the task after unilateral microinfusion of KYNA or bilateral microinfusion of saline in the hippocampus. (A) The total number of trials required for the monkey to locate all 8 reinforcer locations, such that a perfect score would be 8 trials and random performances would average to 21 trials, represented by the dashed bar. (B) The repetition index, a measure of the frequency and severity of box opening errors, where chance performance of 51.4 is represented by the dashed bar. For all graphs, each symbol represents the average of an individual monkey’s performance under those conditions, error bars represent SEM, and significance was set at P < 0.05.