Abstracts

BACKGROUND: There are various types of neuropathology associated with the hippocampus (HPC) in both temporal lobe epilepsy and schizophrenia. Moreover, it has been hypothesized that there are often changes in HPC connectivity or function without concomitant neuroanatomical changes evident from structural imaging.
It is known from a variety of converging data sources that the HPC is critical for spatial memory. One memory task particularly sensitive to HPC function is the Morris water task. In this task, the animal must use the spatial relations in a room to remember and navigate to a hidden goal location within a pool of opaque water. Virtual versions of this task have been used successfully to assess spatial memory in humans and to elicit HPC activation during fMRI in normals (Astur & Constable, 2002).
OBJECTIVE: The goal of this study is to reveal how fMRI can be used during performance of a virtual navigation task to assess the functioning of the HPC in participants with epilepsy or schizophrenia.
METHODS: 10 participants with temporal lobe epilepsy, 10 participants with schizophrenia, and 10 controls performed a virtual Morris water task in a block design fMRI paradigm. Participants were placed in a virtual round pool that had 4 balls floating in the water. In the activation condition, the platform was always in the same location, the balls were identical, and participants had to use the spatial cues to navigate to the ball that floated above the goal platform. In the baseline condition, the platform changed locations from trial to trial, 1 unique ball floated above the platform, and participants navigated to the platform by swimming to the unique ball among 3 identical balls.
RESULTS: Behaviorally, all groups are able to learn this task in a spatial manner and hence should be utilizing HPC resources. All activation maps are the result of a subtraction of the Activation condition (i.e. Spatial memory) minus the Baseline condition (i.e. Cue memory). We find that for the control subjects, there were selective bilateral deactivations in HPC proper. There also were strong left middle frontal gyrus and right striatal activations as well as bilateral insular and posterior cingulate deactivations. In addition, we note bilateral inferior gyrus deactivations anteriorly and positive activations posteriorly. For the patient groups, the left middle frontal gyrus activation appears similar to the control group, but the HPC activation differs both in intensity and location for the two patient groups, and from each other. Subtraction maps as well as ROI analysis are presented to delineate these differences further.
CONCLUSIONS: These results indicate that a spatial navigation task commonly employed with nonhumans can be adapted for use with humans in an fMRI paradigm designed to assess HPC function. We discuss the manner in which these activations differ between patients with epilepsy vs. those with schizophrenia, as well as how these activations may reveal differences that are absent on standard MRI structural analysis.
[Supported by: NIH NS-40497 and NIH 1 F32 MN64290-01]