Spatial Scene Recognition Memory in Epilepsy Surgery

This study investigates the anatomical and physiological basis of spatial scene recognition memory in patients with temporal lobe epilepsy and temporal lobe lesions. Standard neuropsychological tests are insensitive to important memory deficits experienced by patients, particularly in spatial/scene memory, recollective experience, and familiarity processing. Using a validated virtual tour paradigm, the study examines how familiarity-based recognition and recall of spatial scenes relate to specific brain structures. In Aim I, a large cohort of patients with varied temporal lobe lesions at Emory University undergoes the virtual tour task with voxel-based lesion-symptom mapping to localize necessary brain regions. In Aim II, scalp event-related potentials and eye tracking in healthy participants at UC Davis characterize the temporal dynamics and lateralization of scene recognition. In Aim III, intracranial EEG recordings (including local field potentials and single-unit activity) in epilepsy surgery patients at UC Davis determine the precise network dynamics underlying spatial scene familiarity and recall. The long-term goal is to improve the prediction and prevention of cognitive morbidity from epilepsy surgery by providing a more complete model of spatial recognition memory circuits.

Recognition memory can be divided into familiarity (a sense that something has been encountered before) and recollection (identification or elaborative recall). Patients with temporal lobe epilepsy often report subjective memory difficulties that are not captured by conventional neuropsychological tests, which lack assessments of true episodic and scene memory. This study uses a virtual tour paradigm that objectively separates familiarity-based recognition from recall for spatial scenes. The central hypothesis is that spatial scene recognition memory critically involves the convergence of dorsal and ventral visual streams in the inferior parietal lobule and parahippocampal gyrus, with familiarity-based recognition involving neocortical structures and recall involving the parahippocampal gyrus, entorhinal cortices, and hippocampus. The study employs three complementary approaches: (1) lesion-symptom mapping in ~310 surgical patients and 150 controls, (2) scalp ERP and eye-tracking in 80 healthy participants, and (3) intracranial electrophysiology (LFP and single-unit recordings) in ~80 patients undergoing stereoelectroencephalography (SEEG) for clinical seizure localization. Research electrodes (FDA-approved Dixi micro-macro or Behnke-Fried with tetrode components) are placed at clinically determined locations to additionally capture single-neuron activity. This study was classified as a Basic Experimental Studies in Humans (BESH) mechanism; it does not evaluate a health-related clinical outcome but uses clinical populations and FDA-approved research electrodes to study basic neuroscience questions about memory.