The concept ‘cortical point image’ can be traced to the work of Hubel and Wiesel (1974) where they obsessed over the relationship between the amount of V1 tissue dedicated to processing visual information within the region of a receptive field. In later years, this concept evolved into determining how much tissue in V1 is employed to process all the visual attributes such as orientation, spatial frequency, color, motion, and depth at one point of visual space (Ji et al. 2015; Tehovnik et al. 2020). This space of course is not a point for cells in the visual system have receptive fields that allow cells to receive information beyond the size of a cell body as defined by a dendritic arbor. A good way to appreciate the significance of the cortical point image is to imagine what the information processing capability of neurons in V1 would be like if receptive fields were circumscribed by the extent of a cell body rather than by the extent of a dendritic arbor. Let’s take the mouse. In the mouse, V1 has receptive fields that are approximately 13 degrees of visual angle in diameter (Tehovnik et al. 2020). If we were to eliminate all the dendrites, which define the sensory field of reception of a neuron, we would be left with a group of cells in V1 that could only be activated if a specific percept—a unique orientation, spatial frequency, color, motion, and depth—were presented to the cell body field whose size would be a small fraction of a conventional receptive field since the 13 degree area of a normal receptive field would be composed of a population of cell bodies, each cell body encoding a specific percept (Peters 1994). In this situation, perception would be fragmented by the parallel processing capability of each cell body. What this indicates is that once we have the capability of activating a single cell body in V1 for the production of a percept (Tehovnik et al. 2007, 2009), this percept will be incomplete since normal vision depends on dendrites and by bypassing the dendrites through prosthetics the percept created will be parallel and fragmented. This calls for an understanding of how dendrites and the cells within a cortical point image share information to create a non-fragmented unitary percept, a major concern of Hubel and Wiesel (1974). This problem is central to all sensory systems and will therefore be a limiting factor if we are intent on using central nervous system prosthetics with single-cell resolution to restore sensation to receptor-damaged humans.

Reference

Hubel DH, Wiesel TN (1974) Anatomical demonstration of orientation columns in macaque monkeys. J Comp Neurol 177:361-388.

Ji W, Gamanut R, Bista P, D’Souza RD, Wang Q, Burkhalter A (2015) Modularity in the organization of mouse primary visual cortex. Neuron doi:10.1016/j.neuron. 2015.07.004.

Peters A (1994) The organization of the primary visual cortex in the macaque. In: Peters A, Jones EG, eds., Cerebral Cortex. Plenum Press, New York, pp. 1-35.

Tehovnik EJ, Slocum WM (2007) Phosphene induction by microstimulation of macaque V1. Brain Res Res 53:337-343.

Tehovnik EJ, Slocum WM, Smirnakis SM, Tolias AS (2009) Microstimulation of visual cortex to restore vision. Prog Brain Res 175:347-375.

Tehovnik EJ, Froudarakis E, Scala F, Smirnakis SM, Tolias AT (2020) Visuomotor control in mice and primates. In preparation.