Neural Representation and Neural Computation

Churchland, Patricia Smith; Sejnowski, Terrence J.

doi:10.2307/2214198

Cited by 83 publications

(52 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Topographically specialized maps in the brain are known to be formed, across all species, by specific sensory inputs of each modalityvisual, somesthetic, or auditory [e.g., Obermayer et al 1991]. Speech sounds and their ultimate mapped representations are thought to be no different in this respect [Churchland and Sejnowski, 1989]. In both neuroethological and connectionist models, it is well recognized that components of input signals that possess statistical regularities are more easily and efficiently processed than unrelated signal inputs ].…”

Section: Discussionmentioning

confidence: 99%

The Effect of Speaking Style on a Locus Equation Characterization of Stop Place of Articulation

1998

View full text Add to dashboard Cite

Locus equations were employed to assess the phonetic stability and distinctiveness of stop place categories in reduced speech. Twenty-two speakers produced stop consonant + vowel utterances in citation and spontaneous speech. Coarticulatory increases in hypoarticulated speech were documented only for /dV/ and [gV] productions in front vowel contexts. Coarticulatory extents for /bV/ and [gV] in back vowel contexts remained stable across style changes. Discriminant analyses showed equivalent levels of correct classification across speaking styles. CV reduction was quantified by use of Euclidean distances separating stop place categories. Despite sensitivity of locus equation parameters to articulatory differences encountered in informal speech, stop place categories still maintained a clear separability when plotted in a higher-order slope × y-intercept acoustic space.

show abstract

Section: Discussionmentioning

confidence: 99%

The Effect of Speaking Style on a Locus Equation Characterization of Stop Place of Articulation

1998

View full text Add to dashboard Cite

show abstract

“…By and large they are defined by means of their spatial (i.e., anatomical) and temporal features. These entities form a hierarchy according to size and complexity (Churchland & Sejnowski, 1989). The peak of the anatomical hierarchy starts with the all-encompassing category of the whole organism, which is then differentiated into subsequently smaller and more specific entities-the brain, the hemispheres, areas of the cerebral cortex, nuclei of the brainwithin these structures-networks (e.g., columns within the cortex), neurons, synapses-and within these subsynaptic entities-ion channels or single transmitter molecules.…”

Section: Concepts Levels and Hierarchiesmentioning

confidence: 99%

“…Measures can comprise the whole lifespan (developmental changes of the nervous system), days or hours (synaptic pruning), seconds (metabolic changes), or milliseconds (action potentials). As first mentioned by Churchland & Sejnowski, 1989) each neuroscientific measurement operation and conceptual entity can be located somewhere in a two-dimensional grid with two axes-one is the anatomical size of the measured entity, extending from nanometer to meter, and the other the temporal extension of the measures, extending from microseconds to years. All these entities are easily defined by an objective measurement operation.…”

Section: Concepts Levels and Hierarchiesmentioning

confidence: 99%

Some unsettled problems in behavioral neuroscience research

Rösler

2012

Psychological Research

View full text Add to dashboard Cite

The goal of behavioral neuroscience is to map psychological concepts onto physiological and anatomical concepts and vice versa. The present paper reflects on some of the hidden obstacles that have to be overcome in order to find unique psychophysiological relationships. These are, among others: (1) the different status of concepts which are defined in the two domains (ontological subjectivity in psychology and ontological objectivity in physiology); (2) the distinct hierarchical levels to which concepts from the two domains may belong; (3) ambiguity of concepts, because-due to limited measurement resolution or definitional shortcomings-they sometimes do not cover unique states or processes; (4) ignored context dependencies. Moreover, it is argued that due to the gigantic number of states and state changes, which are possible in a nervous system, it seems unlikely that neuroscience can provide exact causal explanations and predictions of behavior. Rather, as in statistical thermodynamics the transition from the microlevel of explanations to the macrolevel is only possible with probabilistic uncertainty.

show abstract

“…I will focus on feedforward, classificatory neural networks (Rumelhart et al, 1986), for their relative ease of access together with their non-trivial computational potential (Sejnowski & Rosenberg, 1987;Churchland & Sejnowski, 1990). There are two layers of analysis I will implement.…”

Section: Introductionmentioning

confidence: 99%

Polytopes as vehicles of informational content in feedforward neural networks

Azhar

2016

Philosophical Psychology

View full text Add to dashboard Cite

Localizing content in neural networks provides a bridge to understanding the way in which the brain stores and processes information. In this paper, I propose the existence of polytopes in the state space of the hidden layer of feedforward neural networks as vehicles of content. I analyze these geometrical structures from an information-theoretic point of view, invoking mutual information to help define the content stored within them. I establish how this proposal addresses the problem of misclassification, and provide a novel solution to the disjunction problem, which hinges on the precise nature of the causal-informational framework for content advocated herein.

show abstract

Neural Representation and Neural Computation

Cited by 83 publications

References 36 publications

The Effect of Speaking Style on a Locus Equation Characterization of Stop Place of Articulation

The Effect of Speaking Style on a Locus Equation Characterization of Stop Place of Articulation

Some unsettled problems in behavioral neuroscience research

Polytopes as vehicles of informational content in feedforward neural networks

Contact Info

Product

Resources

About