Parallel And Serial Theories Of Timing See Scalar Timing Theory

PARALLEL DISTRIBUTED PROCESS ING MODEL. This model of knowledge- and information-processing represents units and items of information as patterns of connections of differing strengths between locations within a network model (i.e., any model consisting of a collection of units, each joined to one or more other units that it may inhibit or excite). In this model, also generally known as a distributed network model, information processing assumes the form of the parallel processing of collections of activated links. The parallel distributed processing model (PDP) is tied closely to one of the definitions of connectionism - the design of "intelligent" systems in the area of artificial intelligence where neural networks consist of patterns of activation over collections of units and where the patterns are adaptive in function as they are capable of "learning" from previous experience. The notion of PDP - in which separate subprocesses are distributed within a single entity - is distinguished from that of "distributed cognition" which refers to information processing that is shared among several separate entities. An example of "distributed cognition" may be seen in the cooperative principle - a concept described by the English philosopher Herbert Paul Grice (1913-1988) wherein individuals normally attempt to cooperate when communicating by following conversational rules of truth/quality, information/ quantity, relevance/explanation, and clarity/ manner, and according to which conversation between humans proceeds, typically, on the assumption that the rules are being followed implicitly by the communicating individuals. The back-propagation algorithm/model is a special PDP method whereby the adjustment of the output of a multi-leveled neural network is used to produce a desired state for a given input by first checking the input and computing the required output for that input, and then comparing the present output with the required output and adjusting the "connection weights" (via the "Delta rule" or the "Hebbian synapse rule" - that is, the use of a "teacher unit" having a predetermined level of activation that governs the activation levels of other units to achieve a desired target level) in order to decrease the discrepancy between the required output and the present output, and then repeating these steps of adjustment for the next level down in the system and for each lower level of the system, in turn, down to the lowest level (this "causes" the system to "learn" to produce the required output). The term anneal (literally, "to burn" or "set on fire") refers to the use of random shocks to the system to alter the states of units in a network of connected units until they are all responding consistently to signals received from one another and no further adjustments lead to further improvements in the network. In general, the PDP models are associationist networks that are an improvement over previous paradigms, such as Markov chaining - a theory that central motor mechanisms involve a chain reaction in which each movement depends on a preceding movement which sends a feedback signal - that were once the strongest modeling available. However, some researchers assert that calling PDP models "neural networks" does not necessarily make them into models of brain function; thus, PDP models are suggestive of cognitive structures for mental functions, but are not necessarily substitutes for human study at the neuronal or cognitive levels. See also FUZZY SET THEORY; INFORMATION/INFORMATION-PROCESSING THEORY; MEANING, THEORIES/ASSESSMENT OF; NEURAL NETWORK MODELS OF INFORMATION PROCESSING. REFERENCES

Grice, H. P. (1957). Meaning. The Philosophical Review, 64, 377-388. McClelland, J. L., & Rumelhart, D. E. (1986).

Parallel distributed processing: Explorations in the microstructure of cognition. Cambridge, MA: M.I.T. Press.

Morris, R. G. M. (Ed.) (1989). Parallel distributed processing: Implications for psychology and neurobiology. New York: Oxford University Press. Parks, R. W., Long, D. L., Levine, D. S., & Crockett, D. J. (1991). Parallel distributed processing and neural networks: Origins, methodology, and cognitive functions. International Journal of Neuroscience, 60, 195214.

Davis, S. (Ed.) (1992). Connectionism: Theory and practice. London: Oxford University Press.

Brain Blaster

Brain Blaster

Have you ever been envious of people who seem to have no end of clever ideas, who are able to think quickly in any situation, or who seem to have flawless memories? Could it be that they're just born smarter or quicker than the rest of us? Or are there some secrets that they might know that we don't?

Get My Free Ebook

Post a comment