Abstracts

Rationale: Our research with epilepsy surgical (ES) patients demonstrates that language dominant anterior temporal lobe (ATL) dysfunction is associated with category-related naming problems while nondominant ATL dysfunction is associated with category-related deficits in recognition/familiarity judgments. Object categories disrupted by ATL dysfunction tend to be visually complex (e.g., landmarks/famous faces vs. man-made items). Category-related deficits are often present preoperatively in TL seizure patients, and often worsen dramatically after surgery. We used a multiple choice (MC) paradigm to elucidate the nature of these deficits. Methods: Naming and recognition of famous faces and landmarks was assessed in a sample of 16 pre- or post-surgical TL patients (half with language dominant seizure onset) using Powerpoint slides. Category-related deficits were present in all our patients in the expected direction (i.e., recognition deficits in nondominant TL patients, naming deficits in dominant TL patients). We also employed a MC paradigm requiring patients to pick one of 4 orally presented names for the objects, and analyzing group performance across two conditions (i.e., items recognized but not named & items rated as unfamiliar). For a subset of post-surgical patients, we also re-administered 30 items initially missed by them but identified with the MC paradigm, as we suspected dominant TL patients could only hold onto the names for a few seconds based on our clinical experience. Results: Dominant and nondominant TL patients could identify with MC more than 90% of the famous faces and landmarks they recognized but were unable to name. However, three dominant TL patients chosen to provide the name of the stimuli a second time performed significantly worse than their nondominant counterparts (p<.01), who performed perfectly on this task (35% vs. 100% accuracy). For the majority of items, dominant TL patients could not come up with the correct name even after selecting it only minutes earlier. When we provided the dominant TL patients with the MC paradigm on items they rated as unfamiliar, they exhibited a chance level performance (i.e., 25%). However, the nondominant TL patients showed significant improvement on these initially unfamiliar items when provided with MC options (i.e., 55%), which again reflects a significant difference between groups (p<.01), and indicates that their recognition deficits are not absolute. Conclusions: Results suggest that category-related deficits observed in TL surgical patients reflect disconnection syndromes, as modifications in task parameters led to improved performance in both naming and recognition. These deficits appear to reflect problems with access to information rather than a degradation of the same. Such a disconnection would not be surprising in individuals undergoing surgical resection, as several white matter pathways and structural regions are disrupted in these cases. Presurgical appearance of deficits may relate to alterations in white matter integrity resulting from seizure activity and associated physiological processes.