Relating Attention to Visual Mechanisms

Abstract

The effect of attention on visual perception has been a subject of great controversy. Much research on attention over the last thirty years has been framed by the debate on this question between proponents of early and late selection. Late selection theorists have argued that all perceptual encoding, including recognition, and certain semantic analyses are accomplished in parallel, while early selection theorists have countered that only simple 'physical' analyses can be conducted in parallel. More sophisticated analyses of shape that support object recognition and memory access are conducted by limited capacity systems. Although the terms 'perceptual' and 'semantic' include a wide variety of processes, there has been an unfortunate tendency to extrapolate conclusions to the entire collection of processes based on results from a few. For example, researchers have tacitly assumed that if any evidence for semantic processing of unattended material is found, then perceptual operations must be parallel. This inference, however, only applies to the perceptual operations relevant to the recognition of the experimental stimuli, which are generally upright block alphanumeric characters. Recognition of these stimuli does not require the resolution of a vast number of problems the visual system must solve: the analysis of motion, three-dimensional space, constancies of various sorts, etc.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Feb 28, 1989
Accession Number
ADA206452

Entities

People

  • Gordon L. Shulman

Organizations

  • University of Washington

Tags

Communities of Interest

  • Human Systems

DTIC Thesaurus Topics

  • Coding
  • Computer Vision
  • Detection
  • Identification
  • Image Recognition
  • Military Research
  • Motor Skills
  • Notation
  • Object Recognition
  • Parallel Computing
  • Parallel Processing
  • Pattern Recognition
  • Perception
  • Psychology
  • Reaction Time
  • Recognition
  • Visual Perception

Readers

  • Computer Vision.
  • Educational Psychology

Technology Areas

  • AI & ML
  • Space