Dynamics of feature categorization

Daniel Martí, John Rinzel

Research output: Contribution to journalArticle

Abstract

In visual and auditory scenes, we are able to identify shared features among sensory objects and group them according to their similarity. This grouping is preattentive and fast and is thought of as an elementary form of categorization by which objects sharing similar features are clustered in some abstract perceptual space. It is unclear what neuronal mechanisms underlie this fast categorization. Here we propose a neuromechanistic model of fast feature categorization based on the framework of continuous attractor networks. The mechanism for category formation does not rely on learning and is based on biologically plausible assumptions, for example, the existence of populations of neurons tuned to feature values, feature-specific interactions, and subthreshold-evoked responses upon the presentation of single objects. When the network is presented with a sequence of stimuli characterized by some feature, the network sums the evoked responses and provides a running estimate of the distribution of features in the input stream. If the distribution of features is structured into different components or peaks (i.e., is multimodal), recurrent excitation amplifies the response of activated neurons, and categories are singled out as emerging localized patterns of elevated neuronal activity (bumps), centered at the centroid of each cluster. The emergence of bump states through sequential, subthreshold activation and the dependence on input statistics is a novel application of attractor networks. We show that the extraction and representation of multiple categories are facilitated by the rich attractor structure of the network, which can sustain multiple stable activity patterns for a robust range of connectivity parameters compatible with cortical physiology.

Original languageEnglish (US)
Pages (from-to)1-45
Number of pages45
JournalNeural Computation
Volume25
Issue number1
DOIs
StatePublished - Jan 2013

Fingerprint

Neurons
Learning
Population
Attractor
Neuron

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Arts and Humanities (miscellaneous)

Cite this

Dynamics of feature categorization. / Martí, Daniel; Rinzel, John.

In: Neural Computation, Vol. 25, No. 1, 01.2013, p. 1-45.

Research output: Contribution to journalArticle

Martí, Daniel ; Rinzel, John. / Dynamics of feature categorization. In: Neural Computation. 2013 ; Vol. 25, No. 1. pp. 1-45.
@article{0d1428d3f26349ba9c771a3810b41e5f,
title = "Dynamics of feature categorization",
abstract = "In visual and auditory scenes, we are able to identify shared features among sensory objects and group them according to their similarity. This grouping is preattentive and fast and is thought of as an elementary form of categorization by which objects sharing similar features are clustered in some abstract perceptual space. It is unclear what neuronal mechanisms underlie this fast categorization. Here we propose a neuromechanistic model of fast feature categorization based on the framework of continuous attractor networks. The mechanism for category formation does not rely on learning and is based on biologically plausible assumptions, for example, the existence of populations of neurons tuned to feature values, feature-specific interactions, and subthreshold-evoked responses upon the presentation of single objects. When the network is presented with a sequence of stimuli characterized by some feature, the network sums the evoked responses and provides a running estimate of the distribution of features in the input stream. If the distribution of features is structured into different components or peaks (i.e., is multimodal), recurrent excitation amplifies the response of activated neurons, and categories are singled out as emerging localized patterns of elevated neuronal activity (bumps), centered at the centroid of each cluster. The emergence of bump states through sequential, subthreshold activation and the dependence on input statistics is a novel application of attractor networks. We show that the extraction and representation of multiple categories are facilitated by the rich attractor structure of the network, which can sustain multiple stable activity patterns for a robust range of connectivity parameters compatible with cortical physiology.",
author = "Daniel Mart{\'i} and John Rinzel",
year = "2013",
month = "1",
doi = "10.1162/NECO_a_00383",
language = "English (US)",
volume = "25",
pages = "1--45",
journal = "Neural Computation",
issn = "0899-7667",
publisher = "MIT Press Journals",
number = "1",

}

TY - JOUR

T1 - Dynamics of feature categorization

AU - Martí, Daniel

AU - Rinzel, John

PY - 2013/1

Y1 - 2013/1

N2 - In visual and auditory scenes, we are able to identify shared features among sensory objects and group them according to their similarity. This grouping is preattentive and fast and is thought of as an elementary form of categorization by which objects sharing similar features are clustered in some abstract perceptual space. It is unclear what neuronal mechanisms underlie this fast categorization. Here we propose a neuromechanistic model of fast feature categorization based on the framework of continuous attractor networks. The mechanism for category formation does not rely on learning and is based on biologically plausible assumptions, for example, the existence of populations of neurons tuned to feature values, feature-specific interactions, and subthreshold-evoked responses upon the presentation of single objects. When the network is presented with a sequence of stimuli characterized by some feature, the network sums the evoked responses and provides a running estimate of the distribution of features in the input stream. If the distribution of features is structured into different components or peaks (i.e., is multimodal), recurrent excitation amplifies the response of activated neurons, and categories are singled out as emerging localized patterns of elevated neuronal activity (bumps), centered at the centroid of each cluster. The emergence of bump states through sequential, subthreshold activation and the dependence on input statistics is a novel application of attractor networks. We show that the extraction and representation of multiple categories are facilitated by the rich attractor structure of the network, which can sustain multiple stable activity patterns for a robust range of connectivity parameters compatible with cortical physiology.

AB - In visual and auditory scenes, we are able to identify shared features among sensory objects and group them according to their similarity. This grouping is preattentive and fast and is thought of as an elementary form of categorization by which objects sharing similar features are clustered in some abstract perceptual space. It is unclear what neuronal mechanisms underlie this fast categorization. Here we propose a neuromechanistic model of fast feature categorization based on the framework of continuous attractor networks. The mechanism for category formation does not rely on learning and is based on biologically plausible assumptions, for example, the existence of populations of neurons tuned to feature values, feature-specific interactions, and subthreshold-evoked responses upon the presentation of single objects. When the network is presented with a sequence of stimuli characterized by some feature, the network sums the evoked responses and provides a running estimate of the distribution of features in the input stream. If the distribution of features is structured into different components or peaks (i.e., is multimodal), recurrent excitation amplifies the response of activated neurons, and categories are singled out as emerging localized patterns of elevated neuronal activity (bumps), centered at the centroid of each cluster. The emergence of bump states through sequential, subthreshold activation and the dependence on input statistics is a novel application of attractor networks. We show that the extraction and representation of multiple categories are facilitated by the rich attractor structure of the network, which can sustain multiple stable activity patterns for a robust range of connectivity parameters compatible with cortical physiology.

UR - http://www.scopus.com/inward/record.url?scp=84878176898&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84878176898&partnerID=8YFLogxK

U2 - 10.1162/NECO_a_00383

DO - 10.1162/NECO_a_00383

M3 - Article

VL - 25

SP - 1

EP - 45

JO - Neural Computation

JF - Neural Computation

SN - 0899-7667

IS - 1

ER -