The effect of attention on visual ERPs in children with specific learning disability

Society Proceedings/Electroencephalography 31. Two cases of generalized auditory agnosia: electrophysiological distinction and neurobehavioral correlates. - D. Kaufer, M. Nuwer and D.F. Benson (Department of Neurology, School of Medicine, UCLA, Los Angeles, CA) Auditory agnosia refers to impaired sound recognition due to central auditory dysfunction with preserved language abilities. These rare disorders may involve defective recognition of verbal sounds (pure word deafness), environmental sounds (non-verbal auditory agnosia), or both (generalized auditory agnosia). The vast majority of adult cases are due to cerebrovascular insults producing gross structural lesions evident on CT or MRI scans. In the absence of focal structural changes, however, an auditory agnosia may easily be misinterpreted as a primary psychiatric disturbance. Two young adults with complex neuropsychiatric presentations were evaluated at the UCLA Neurobehavior Clinic. Careful clinical assessment and peripheral audiometry revealed each to have a generalized auditory agnosia. MRI brain scans failed to reveal any gross focal abnormalities. Brain-stem auditory evoked potentials and EEG along with P300 cortical evoked potentials implicated a cortical lesion site in one patient and a subcortical lesion site in the other. These 2 cases highlight the utility and limitations of electrophysiological assessment of central auditory disturbances. The interplay between psychoacoustic and neurobehavioral phenomena is explored in terms of their relative contributions to overt clinical manifestation and the functional anatomy of the central auditory system. 32. P300 topography predicts treatment response in attention deficit disorder. - J.M. Sangal, RB. Sangal and B. Persky ’ (Attention Disorders Institute, Troy, MI, and ’ Walden University, Minneapolis, MN) 45 children with attention deficit disorder (ADD) age 6-15 were administered auditory and visual P300 using 31 scalp electrodes. Patients were treated with pemoline, and good and poor responders compared. Poor pemoline responders were treated with imipramine, and good and poor responders compared. ADD did not differ from normals in P300 topography. Good and poor responders to pemoline were clinically identical, as were good and poor responders to imipramine. Poor pemoline responders had smaller right fronto-central auditory amplitudes than good responders. Poor imipramine responders had longer auditory and visual latencies than good responders. The ratio of right fronto-central to parietal auditory amplitude > 0.5 as a test for good pemoline response was highly specific (0.7) and sensitive (0.76). Age-adjusted mean visual latency < 605 msec as a test for good imipramine response was highly sensitive (0.86) and specific (0.8). P300 topography classifies ADD into 3 groups: group 1 with normal P300 topography, and good response to pemoline; group 2 with small right fronto-central auditory P300 amplitudes, poor response to pemoline, and good response to imipramine; and group 3 with small right fronto-central auditory P300 amplitudes, long auditory and visual P300 latencies, and poor response to pemoline and imipramine. 33. Effects of antiepileptic drugs on the P3 potential in healthy subjects. - KJ. Meador, D.W. Loring, M.E. Nichols, D.W. King, B.B. Gallagher, RE. Obenan, E.E. Moore and W.O. Thompson a (Departments of Neurology and a Biostatistics, Medical College of Georgia, Augusta, GA) Comparative cognitive effects of antiepileptic drugs (AEDs) remain controversial. The P3 event-related potential offers a sensitive index of cognitive processing speed. As part of a neurobehavioral study, P3 latency was recorded using the tonal oddball paradigm in 59 healthy volunteers in a double-blind, randomized, incomplete block, crossover design comparing phenobarbital, phenytoin, and valproate. Each subject was treated with 2 of the 3 AEDs for 1 month each and tested when

and clinical Neurophysiology 95 (1995) 15P-41P

21P

anticonvulsant blood levels were within standard therapeutic ranges. Assessments were conducted at baseline, after 1 month on each of 2 AEDs, and a 2 repeat non-drug baselines 3 months after each drug washout. No differences were noted across the 3 non-drug baseline conditions with mean P3 latencies ranging from 344 to 348 msec. Overall, AEDs slowed P3 latency compared to the non-drug conditions (F (3, 58) = 12.36, P = O.OOOl), but the AEDs did not differ. Least squares means ( f SE.) in msec were: phenobarbital = 368 ( + 3), phenytoin = 358 ( +_31, valproate = 364 ( f 3). and average non-drug baselines = 346 ( + 2). The results demonstrate cognitive slowing by all 3 AEDs, but no differential AED effects. This research was supported

by a grant from Parke-Davis.

34. The effect of attention on visual ERPs in children with specific learning disability. - M. Brigell and M.A. Gordon (Department of Neurology, Loyola University Chicago, Maywood, IL)

Twenty-seven children, 10 with mathematical learning disability (MLD), 7 with verbal LD (VLD), and 10 normal controls, participated in a Hillyard-type visual attention ERP study. Children ranged from 8 to 11 years of age. Responses were recorded for 100 msec pre- and 900 msec post-stimulus presentation from 19 electrodes placed at International lo-20 locations. Separate averages were obtained for either target or non-target stimuli presented in either the attended or unattended field. Analysis was performed on group mean data. For controls there was a robust effect of attention on PlOO amplitude at occipital and parietal areas contralateral to the stimulus field. The magnitude of the attention effect was roughly symmetric for stimuli in left and right visual fields. In contrast, VLD children showed an attention effect on PlOO amplitude only for stimuli presented in the left field. Children with MLD showed an abnormal attention effect only to target stimuli presented in the right field. Our results suggest that allocation of cortical resources is abnormal in the left hemispheres of children with specific mathematical or verbal disability.

coherence mapping reveals tongue activa35. Electrocorticograpbic tion areas of the human sensorimotor cortex. - D.L. Sherman, S.S. Nathan “, N.E. Crone ‘, RP. Lesser ‘, B. Gordon ’ and N.V. Thahor (Department of Biomedical Engineering and a The Zanvyl Krieger Mind-Brain Institute and Departments of Neurology and Cognitive Neurology, Johns Hopkins University, Baltimore, MD)

To enhance capabilities aimed at localizing tongue motor functions and specifying their accompanying ECoG (electrocorticographic) patterns, EEG coherence function measurements were obtained from pairs of subdurally implanted electrodes. Coherence measures the degree of linear association and level of local electrical signal correlation between electrodes independent of power spectral content. Tongue protrusion tasks were evaluated to measure effects on the mu, beta and gamma rhythms through 50 Hz. Coherence was calculated during five 500 msec epochs stretching from 0.5 set pre stimulus to 2 set post stimulus. FFT-derived power measurements were averaged across time and frequency band and normalized coherence was calculated. These results were compared with the functional anatomy defined by cortical electrical stimulation mapping. There were significant (P < 0.0005) localized changes in mu coherence between two adjacent electrode pairs in regions where cortical electrical stimulation interfered with tongue motor function. Monotonically increasing and bi-directional (increase, then restorative) trends in mu coherence level across all of the epochs were noted. In addition, these and two adjoining electrodes pairs exhibited significant (P < 0.001) pairwise contrasts between epochs for beta and gamma frequency ranges.