Search PubMed⌕ Search
Home/Archive/PubMed

PubMed · 7596429

Pattern recognition computation using action potential timing for stimulus representation.

J J Hopfield
View original source ↗Export metadata ↓

Abstract

A computational model is described in which the sizes of variables are represented by the explicit times at which action potentials occur, rather than by the more usual 'firing rate' of neurons. The comparison of patterns over sets of analogue variables is done by a network using different delays for different information paths. This mode of computation explains how one scheme of neuroarchitecture can be used for very different sensory modalities and seemingly different computations. The oscillations and anatomy of the mammalian olfactory systems have a simple interpretation in terms of this representation, and relate to processing in the auditory system. Single-electrode recording would not detect such neural computing. Recognition 'units' in this style respond more like radial basis function units than elementary sigmoid units.

Explore related subjects

Action PotentialsAnimalsAuditory PerceptionModels, NeurologicalNeuronsOscillometryPerceptionRatsSmellTime FactorsJournal ArticleResearch Support, Non-U.S. Gov'tResearch Support, U.S. Gov't, Non-P.H.S.

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

J J Hopfield. 1995-07-06. Pattern recognition computation using action potential timing for stimulus representation.. https://doi.org/10.1038/376033a0

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed1995

Blockade of action potential activity alters initial arborization of thalamic axons within cortical layer 4.

In the formation of connections during the development of the nervous system, it is generally accepted that there is an early phase not requiring neural activity and a later activity-dependent phase. The initial processes of axonal pathfinding and target selection are not thought to require neural activity, whereas the later fine-tuning of connections into their final adult patterns does. We report an apparent exception to this rule in which action potential activity seems to be required very early in development for thalamic axons to form appropriate patterns of terminal arborizations with their ultimate target neurons in layer 4 of the cerebral cortex. Blockade of sodium action potentials during the 2-week fetal period when visual thalamic axons initially grow into the primary visual cortex in cats prevents the normally occurring branching of lateral geniculate nucleus axons within layer 4. This observation implies a role for action-potential activity in cerebral cortical development far earlier than previously suspected, weeks before eye-opening and the onset of the well-known process of activity-dependent reorganization of axonal terminal arbors that leads to the formation of ocular dominance columns.

Action Potentials↗
PubMed1995

Concerted signaling by retinal ganglion cells.

To analyze the rules that govern communication between eye and brain, visual responses were recorded from an intact salamander retina. Parallel observation of many retinal ganglion cells with a microelectrode array showed that nearby neurons often fired synchronously, with spike delays of less than 10 milliseconds. The frequency of such synchronous spikes exceeded the correlation expected from a shared visual stimulus up to 20-fold. Synchronous firing persisted under a variety of visual stimuli and accounted for the majority of action potentials recorded. Analysis of receptive fields showed that concerted spikes encoded information not carried by individual cells; they may represent symbols in a multineuronal code for vision.

Action Potentials↗