Notes

Chapter 10: Processes of Perception and Analysis

Section 13: Higher Forms of Perception and Analysis


Biological [forms of] perception

Animals can process data not only from visual or auditory sources (as discussed on pages 577 and 585), but also from mechanical, thermal, chemical and other sources. Usually special receptors for each type of data convert it into electrical impulses in nerve cells. Mechanical and thermal data are often mapped onto an array of nerve cells in the brain, from which features are extracted similar to those in visual perception. Taste involves data from solids and liquids; smell data from gases. The human tongue has millions of taste buds scattered on its surface, each with many tens of nerve cells. Rather little is currently known about how taste data is processed, and it is not even clear whether the traditional notion that there are just four or so primary tastes is correct. The human nose has several tens of millions of receptors, apparently broken into a few hundred distinct types. Each of these types probably has proteins that form pockets with definite shapes, making it respond to molecules whose shapes fit into these pockets. People typically distinguish a few thousand odors, presumably by comparing responses of different receptor types. (Foods usually contain tens of distinct odors; manufactured scents hundreds.) There is evidence that at the first level of processing in the brain all receptors of a given type excite nerve cells that lie in the same spatial region. But just how different regions are laid out is not clear, and may well differ between individuals. Polymers whose lengths differ by more than one or two repeating units often seem to smell different, and it is conceivable that elaborate general features of shapes of molecules can be perceived. But more likely there is no way to build up sophisticated taste or smell data—and no analog of any properties such as repetition or nesting.


From Stephen Wolfram: A New Kind of Science [citation]