Abstract
The multifocal visual evoked potential (mfVEP) is largely generated in V1 [1–4]. To better understand the neural substrate of the mfVEP, response-contrast functions were compared to predictions from a model based upon single cell recordings from monkey V1 [5,6]. Monocular mfVEPs were obtained from three normal subjects with a pattern reversing dartboard pattern (VERIS, EDI), The display contained 16 sectors each with 64 checks and both the sectors and the checks were scaled approximately for cortical magnification. The conditions included six contrast levels and three viewing distances (check sizes varied by a factor of 8). The records from 3 channels of recording were analyzed with custom software [4]. Two measures of response amplitude, the RMS and the first principle component [3], gave similar results. Even though the check size varied by a factor of 8, the overall response-contrast functions (RvsC) were similar in shape, saturating by 35 to 50%. These functions were well fitted up to 40 to 50% contrast by the theoretical population curve for V1 neurons developed by Albrecht & Geisler (see [6]); there was a systematic deviation for higher contrasts. However, when the results for the central 10° were analyzed separately, the larger checks saturated at higher contrasts than did the relatively smaller checks. These results deviate from the predictions of the model. However, in general the results are consistent with a model of V1 neuron population, which predicts no change in the RvsC function with spatial frequency of display. The systematic deviation from the theoretical RvsC function for higher contrasts and for central vision may be due to a weakness in the model and/or a mixture of components with different waveforms in the mfVEP response [1]. 1. Baseler & Sutter, (1997) 2. Slotnick et al (1999) 3. Zhang & Hood (2004) 4. Hood & Greenstein (2003) 5. Albrecht & Hamilton (1982); 6. Heeger et al (2000).