Temporal Structure and Interpretability in the Classification of Nonspeech Acoustic Patterns.

Abstract

Two experiments investigated the role of syntactic (sequential structure) and semantic factors in the classification of complex environmental sound patterns. The results of the first experiment were consistent with our earlier findings in revealing that sequentially structured, interpretable patterns are classified more accurately than unstructured patterns, an explicitly provided semantic context enhances initial classification performance with interpretable patterns, however, no semantically-related enhancement results with unstructured, uninterpretable patterns. Experiment 2 examined classification of sequentially-structured, but minimally-interpretable patterns. The results showed that sequential structure alone can lead to optimal classification performance, and providing explicit semantic information impaired performance with these patterns. In both experiments listeners appeared to learn something about the composition rules used to produce the target patterns rather than simple paired-associate responses. Syntactic and semantic factors play a role in the classification of complex nonspeech patterns. (Author)

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 15, 1981
Accession Number
ADA110521

Entities

People

  • James A. Ballas
  • James H. Howard Jr.

Organizations

  • The Catholic University of America

Tags

Communities of Interest

  • Biomedical
  • Human Systems

DTIC Thesaurus Topics

  • Air Force
  • Applied Psychology
  • Biological Sciences
  • Biomedical Research
  • Classification
  • Control Systems
  • Engineering
  • Human Factors Engineering
  • Imaging Techniques
  • Military Research
  • Naval Training
  • Navy
  • Perception
  • Psychology
  • Students
  • Systems Engineering
  • Training

Fields of Study

  • Computer science

Readers

  • Brain and Cognitive Science; Experimental Psychology; Cognitive Neuroscience
  • Computational Linguistics
  • Neural Network Machine Learning.