Epilepsia, 49(4):710–713, 2008 doi: 10.1111/j.1528-1167.2007.01483.x BRIEF COMMUNICATION Temporal lobe epilepsy does not impair visual perception * Arthur C. Grant, * Kiely M. Donnelly, †Charlie Chubb, * William B. Barr, * Ruben Kuzniecky, and * Orrin Devinsky * Department of Neurology, New York University, New York, New York, U.S.A. and †Department of Social Sciences, University of California Irvine, Irvine, California, U.S.A. SUMMARY Temporal lobe epilepsy (TLE) can impair interic- tal cognitive function. In the perceptual domain, previous psychophysical studies demonstrated spe- cific deficits in auditory and tactile perception in patients with TLE. This study compared perfor- mance of 25 TLE subjects and 27 controls on two low-level, visual tasks: luminance discrimina- tion and frequency discrimination. Both tasks were performed under a relatively easy and a relatively difficult condition, by adjusting the stimulus dura- tion. TLE subjects performed as well as controls on both tasks at both stimulus durations. These results imply that interictal occipital lobe func- tion, as reflected in performance on low-level visual tasks, is not impaired in TLE, consistent with func- tional imaging data. Furthermore, since TLE sub- jects performed normally while taking therapeutic doses of multiple AEDs, the data suggest that these AEDs do not impair visual perception. KEY WORDS: Temporal lobe epilepsy, Visual per- ception, Psychophysics, Cognitive function, AED. Epilepsy may affect cognition, and specific cognitive deficits are associated with different epilepsy syndromes. Patients with temporal lobe epilepsy (TLE) often suffer from impaired memory and language function (Ogden- Epker & Cullum, 2001). Cognitive deficits related to ex- tratemporal structures, particularly frontal lobe, may also occur (Helmstaedter & Kockelmann, 2006). Consistent with these behavioral data, a variety of functional imaging studies have demonstrated that mesial TLE is frequently associated with widespread metabolic disturbances, in- cluding both bilateral temporal and ipsilateral extratempo- ral regions (Henry & Votaw, 2004; Mueller et al., 2004). These and other data, such as a true cure rate of only about 25% after temporal lobectomy (Schmidt et al., 2004), suggest that TLE may be accurately viewed as a network disease. We are investigating one broad cognitive domain— perception—in patients with TLE to better define and un- derstand the extent of cognitive dysfunction associated with this disease. Using low-level and quantitative psy- chophysical tasks, we and others have demonstrated spe- cific deficits in the tactile and auditory domains (Ehrle Accepted November 15, 2007; Online Early publication December 31, 2007. Address correspondence to Arthur C. Grant, 403 East 34th St., 4th Floor, New York, NY 10016, U.S.A. E-mail: arthur.grant@med.nyu.edu Blackwell Publishing, Inc. C  2008 International League Against Epilepsy et al., 2001; Grant et al., 2006). Here we report studies of visual perception in TLE. METHODS Subjects Subjects consisted of 29 patients with medically uncon- trolled TLE undergoing epilepsy surgery evaluation at the NYU epilepsy center. TLE subjects were tested in an out- patient setting while taking their usual AEDs, and when- ever possible AED levels were measured on the day of testing. Thirty neurologically normal control subjects of similar age and education were recruited from the com- munity through flyers and by word of mouth. Exclusion criteria included corrected visual acuity worse than 20/40 in either eye, clinically significant bilateral hearing loss, calluses at the tip of the index fingers, history of peripheral neuropathy or upper extremity nerve injury, or inability to understand English. The study was approved by the NYU Institutional Review Board. Subjects provided written in- formed consent and were paid for their participation. Visual psychophysics Subjects performed two low-level visual tasks— luminance discrimination and spatial frequency discrimination—both based on classic studies (Campbell et al., 1970; Whittle, 1986). Both tasks were performed at two stimulus durations chosen to represent relatively easy 710