RATS AND COGNITIVE MAPPING

This is a "classic" experiment reported in 1983 by R. G. M. Morris. ("An attempt to dissociate 'spatial-mapping' and 'working memory' theories of hypocampal function", in Neurobiology of the hippocampus, ed. W. Seifert, 405-32, London: Academic Press; described in Explaining Science, A Cognitive Approach, Ronald N. Giere, 1990, University of Chicago Press.)

A rat is placed in a round tank roughly 4 feet in diameter and deep enough that the rat's feet cannot touch bottom with nose above the water. The tank is filled with milk (at a chilly 25 degrees Centigrade) to make the liquid too opaque for surveyance. The tank also con-
tains a small "clear plastic platform" which would raise a rat above most of the chilly fluid.

A rat placed in the tank for the first time usually swims about "in gradually decreasing circles" until it finds the platform and climbs onto it. The rat is then removed from the tank and placed in it later. The rat usually swims "straight" for the platform and clambers to comfort.

Says Giere, "A current theory is that when it first locates the platform the rat constructs a cognitive map [my emphasis], that is, an internal representation of the spatial features of its surroundings [my emphasis]. When placed back into the tank at a different position, it locates itself in its cognitive map and then heads in the direction of the platform as indicated by the [cognitive] map. The rat's brain thus is not literally doing geometrical calculations, any more than does a person reading a [standard] map. The repre-
sentational system, [i.e.] the map, already contains all the neces-
sary information. It is an additional virtue of this theory that it postulates a direct connection between the representational system and the sensorimotor system [my emphasis], that is, between thought and action [my emphasis].

"Recent experiments have even been interpreted as showing that at least part of the mapmaking mechanism [my emphasis] is located in the hypocampus ["The hypocampus as a cognitive map", J. O'Keefe and L. Nagel, Clarendon Press, 1978 -- but see below.] Recordings of the firing rate of single cells in the hypocampus of freely moving rats show that specialized cells (now called place cells) fire differentially as a function of location in the environment. In effect these maps are like lights on a map that indicate 'you are here'. In different environments the same cell may be assigned different locations with no special relationships among maps for different environments. Nevertheless, rats are capable of retaining at least half a dozen different map simultaneously.

"Moreover, rats who have mastered the tank revert to their original circular search pattern after an operation that destroys the hypocampus. Control rats given sham operations retain their ability to find the platform. Apparently, destroying the hippocampus makes it impossible for the rat to use its previously constructed map. ["An attempt to dissociate 'spatial-mapping' and 'working memory' theories of hypocampal function", Neurobiology of the hypocampus, ed. W. Seifert, 405-32, Academic Press, 1983.]

"Finally, some recent theories of sensimotor control suggests that the brain works more on geometrical than on algebraic principles ["Cognitive neurobiology: A computaitonal hypothesis for laminar cortex", P. M. Churchland, Biology and Philsophy 1, 25-32, 1983.] Studies of the cerebellum ["Space-time representation in the brain: The cerebellum as a predictive space-time metric tensor", A. Pellionisz & R. LLinas,Neuroscience 7, 2949-70, 1982] indicate that one of [the cerebellum's] functions is to transform sensory input vectors into motor-controlling output vectors."

This experiment suggests to me (and, apparently, to many others) that ("Cognitive Hypothesis":) the brain forms maps of "reality" to guide our behavior. (Note, however, that certain experiments -- hyperlinked, below, by "Back" -- describe the mapping locale of the brain as, not in the hypocampus, but in "the fissure between the parietal and occipital lobes". Also new experiments on infants invoke comparisons.)