There is increasing evidence that the brain relies on a set of canonical neural computations, repeating them across brain regions and modalities to apply similar operations to different problems. A promising candidate for such a computation is normalization, in which the responses of neurons are divided by a common factor that typically includes the summed activity of a pool of neurons. Normalization was developed to explain responses in the primary visual cortex and is now thought to operate throughout the visual system, and in many other sensory modalities and brain regions. Normalization may underlie operations such as the representation of odours, the modulatory effects of visual attention, the encoding of value and the integration of multisensory information. Its presence in such a diversity of neural systems in multiple species, from invertebrates to mammals, suggests that it serves as a canonical neural computation.

/pdf/normalization-as-a-canonical-neural-computation-3uvgzn1unc.pdf

Normalization as a canonical neural computation.

How Inhibition Shapes Cortical Activity

http://ilab.usc.edu/publications/doc/Peters_etal05vr.pdf

Components of bottom-up gaze allocation in natural images

We can claim that we know what the visual system does once we can predict neural responses to arbitrary stimuli, including those seen in nature. In the early visual system, models based on one or more linear receptive fields hold promise to achieve this goal as long as the models include nonlinear mechanisms that control responsiveness, based on stimulus context and history, and take into account the nonlinearity of spike generation. These linear and nonlinear mechanisms might be the only essential determinants of the response, or alternatively, there may be additional fundamental determinants yet to be identified. Research is progressing with the goals of defining a single "standard model" for each stage of the visual pathway and testing the predictive power of these models on the responses to movies of natural scenes. These predictive models represent, at a given stage of the visual pathway, a compact description of visual computation. They would be an invaluable guide for understanding the underlying biophysical and anatomical mechanisms and relating neural responses to visual perception.

https://www.jneurosci.org/content/jneuro/25/46/10577.full.pdf

Do we know what the early visual system does

https://www.cns.nyu.edu/~david/courses/sm12/Readings/Olsen-Neuron2010.pdf

Divisive Normalization in Olfactory Population Codes

https://www.ini.uzh.ch/%7ekiper/my_pdfs/freeman.pdf

Suppression without Inhibition in Visual Cortex

The responses of neurons in primary visual cortex (V1) are suppressed by stimuli presented in the region surrounding the receptive field. There is debate as to whether this surround suppression is due to intracortical inhibition, is inherited from lateral geniculate nucleus (LGN), or is due to a combination of these factors. The mechanisms involved in surround suppression may differ from those involved in suppression within the receptive field, which is called cross-orientation suppression. To compare surround suppression to cross-orientation suppression, and to help elucidate its underlying mechanisms, we studied its temporal properties in anesthetized and paralyzed cats. We first measured the temporal resolution of suppression. While cat LGN neurons respond vigorously to drift rates up to 30 Hz, most cat V1 neurons stop responding above 10-15 Hz. If suppression originated in cortical responses, therefore, it should disappear above such drift rates. In a majority of cells, surround suppression decreased substantially when surround drift rate was above approximately 15 Hz, but some neurons demonstrated suppression with surround drift rates as high as 21 Hz. We then measured the susceptibility of suppression to contrast adaptation. Contrast adaptation reduces responses of cortical neurons much more than those of LGN neurons. If suppression originated in cortical responses, therefore, it should be reduced by adaptation. Consistent with this hypothesis, we found that prolonged exposure to the surround stimulus decreased the strength of surround suppression. The results of both experiments differ markedly from those previously obtained in a study of cross-orientation suppression, whose temporal properties were found to resemble those of LGN neurons. Our results provide further evidence that these two forms of suppression are due to different mechanisms. Surround suppression can be explained by a mixture of thalamic and cortical influences. It could also arise entirely from intracortical inhibition, but only if inhibitory neurons respond to somewhat higher drift rates than most cortical cells.

Tobe C.B. Freeman

Papers

Suppression without Inhibition in Visual Cortex

Temporal properties of surround suppression in cat primary visual cortex.