How fibers subserve computing capabilities: similarities between brains and machines.

Can the principles underlying the design of computers help to explain the cognitive capabilities of the human brain? This chapter shows that these principles can provide insight into the capabilities of the human cerebellum, the internal structure of which bears a remarkable resemblance to the design of a versatile computer. In computers, information processing is accomplished both by the hardware in the system (its circuitry) and by the software (the communication capabilities inherent in its circuitry), which is combination can produce a versatile information-processing system, capable of performing a wide variety of functions, including motor, sensory, cognitive, and linguistic ones. Such versatility of function is achieved by computer hardware in which many modules of similar circuits are organized into parallel processing networks; this structural organization is exemplified in the cerebellum by its longitudinal modules of similar circuits, which are arrayed in parallel zones throughout the structure. On the basis of this known cerebellar "hardware," it is possible to investigate the "software" capabilities inherent in the circuitry of the modules. Each module in the lateral cerebellum seems able to communicate with the cerebral cortex by sending out signals over a segregated bundle of nerve fibers, which is a powerful way of communicating information. We show why this bundling of fibers can enable the cerebellum to communicate with the cerebral cortex (including the prefrontal cortex) at a high level of discourse by using internal languages that are capable of conveying complex information about what to do and when to do it. We propose that such communication activity is reflected in the activation obtained on functional imaging of the cerebro-cerebellar system during the performance by humans of complex motor, sensory, cognitive, linguistic, and affective tasks. Further, we propose a new way of analyzing such cerebro-cerebellar activation, in order to ascertain whether the cerebellar circuitry can (like the circuitry in a versatile computer) perform a wide repertoire of computations on this wide range of information. It seems important to ascertain whether cerebellar circuitry is versatile in its computing capabilities because the demonstration of such versatile capabilities would enable theorists to resolve many of the current controversies about cognitive processing in the mammalian brain.