Tool Module: "Grandmother Cells", or Synchronous Discharges of Neurons?

The ability to recognize faces is especially adaptive from an evolutionary standpoint. But just because some cells respond specifically to faces as a stimulus does not necessarily mean that the brain contains group of neurons that process information about faces exclusively. Quite possibly, these neurons may also contribute to the processing of other kinds of stimuli as well.

Nevertheless, it is true that a convergence of visual information can be observed all along the visual pathways. From the retina to the LGN and the primary visual cortex and on to the extrastriate areas, the receptive fields of the cells grow increasingly complex as they integrate increasingly complex information, up to and including images of faces. This hierarchical processing structure thus does raise the possibility that there may be a small number of as yet unidentified neurons that may be activated by the perception of specific, or even unique, objects. The standard way of raising this question is to ask whether the brain may contain "grandmother cells"—a very small number of cells that have a receptive field so specific that they would respond to only one precise stimulus: the face of the person's grandmother.

Despite the attractiveness of this hypothesis, there are a lot of data that argue against it. Researchers have made recordings of cell activity in a great many different areas of the monkey brain without ever discovering any cells specific to the thousands of individuals and objects that individuals are likely to encounter in the course of their lives. The principles on which the nervous system operates seem to rely more on redundancy and distributed processing of stimuli. Also, the same cells that seem to be selective for a given characteristic of a stimulus often respond more weakly to some of its other properties. It would be far too risky for the nervous system to rely too much on selectivity. If it did, a well placed blow to the head could damage this small number of "grandmother cells", and you might never recognize dear old Granny again.

Our growing understanding of the visual system indicates that it is actually organized into functional modules that are relatively independent of one another. In addition, the higher information-processing areas return information to the lower ones, so that information travels in both directions between the modules, and not just upward.

One possible role for these "re-entrant circuits" may be to make the higher areas' visual maps match the more precise ones in the lower areas. Because the visual fields of the higher areas are generally larger, they are not so precise and detailed as the ones found in areas V1 and V2. The colour that area V4 attributes to an area of space located on a table might thus be more specifically attributed to a book lying on that table, for example.

In the absence of grandmother cells, visual perception might be based on the parallel processing that takes place in the visual pathways, from the M and P-IB channels and the blobs to the various modules of the extrastriate areas. Since the brain appears to engage in a certain division of labour to analyze the various visual attributes of an object, perception may arise from the association of the various cortical areas that respond to the object in question. But then how would the brain combine these various attributes of the object to produce the unified image that we perceive? On the basis of the numerous horizontal connections in layer III of the cortex that link neurons from various modules with one another, some scientists believe that the interactions among modules that are activated by the same object might be what makes our perception of this object possible.

Other researchers, such as Charles Gray and Wolf Singer, think that the key to the problem of how the various parallel processing modules link up with one another, known as the "binding problem," may lie in the synchronous discharge of populations of neurons in very different areas of the brain. Indeed, in 1989, Gray and Singer discovered that some cells in several different modules of the visual cortex appear to discharge their action potentials simultaneously when they are activated by parts of the same object.

Thus your mental image of your grandmother may reside not in a few specialized neurons, but rather in a large population of neurons distributed across various cortical areas. The functional co-operation of neurons that discharge synchronously might thus be the "glue" that takes all the distinct elements analyzed separately by the brain and binds them into a coherent whole.


Close this window.