December 2010
Volume 10, Issue 15
OSA Fall Vision Meeting Abstract  |   December 2010
Normalization of contrast responses in the visual cortex of humans and mice as seen in the frequency-tagged EEG
Author Affiliations
  • Anthony M. Norcia
    Department of Psychology, Stanford University, Stanford, CA, USA
Journal of Vision December 2010, Vol.10, 30. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Anthony M. Norcia; Normalization of contrast responses in the visual cortex of humans and mice as seen in the frequency-tagged EEG. Journal of Vision 2010;10(15):30. doi:

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

There is now considerable evidence that the brain uses a relatively small set of basic computations across sensory systems. One prominent computation is divisive gain control or “contrast normalization”. I will describe experimental data and computational modeling studies based on EEG recordings of the frequency-tagged Visual Evoked Potential. In a frequency-tagged EEG experiment, the non-linear operation of divisive gain control mechanisms can be observed in great detail. In a typical experiment, the contrast response function is measured for a “test” stimulus that is modulated at a specific temporal frequency. This response function is then remeasured in the presence of a second stimulus of fixed contrast --- the “mask” --- which is modulated at a different temporal frequency that is not an integer multiple of the test frequency. By making the test and mask frequencies distinctive, the population responses to the test and mask stimuli can be measured separately, even though they are both simultaneously present. The addition of the masker shifts the contrast response function of the test stimulus rightward, consistent with a divisive contrast gain control. At the same time, the masker response decreases as the test contrast increases (mutual normalization). Finally, due to the non-linear nature of the normalization process, additional response components are observed at frequencies that are low-order sums and differences of the two tag frequencies. These responses are maximal when the test and mask are equal in contrast. Computational modeling of these frequency-domain responses using common formulations of the normalization process captures each of these effects. Together, the modeling and empirical results demonstrate in a clear fashion that divisive gain control mechanisms provide a means to not only adjust contrast gain/sensitivity, but also to implement a winner-take-all mechanism. Contrast gain control mechanisms are present but immature in young infants and are measurable in EEG recordings from the dura of alert mice.


This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.