Adaptive Human-Computer Interfaces Using Expert Profiles

Abstract

Adaptive human-computer systems accommodate a wide variety of users learning to interact with computers because they adjust to different skill levels and provide novices with appropriate levels of expertise needed to perform certain tasks. This effort was directed toward developing improved models of experts based on goal-based models and toward assaying and isolating individual differences of inexperienced users in order to adapt the software interface to these individual differences. Results show that inexperienced users and that slower users were more variable in their search times than experienced users and that slower users selected more inefficient search commands. Two performance-based and one cognitive-based command selection aid improved search performance and strategies of slower, inexperienced users. Since spatial and verbal ability were found to correlate of slower, positively with search strategies, inexperienced users learned to select fewer and more efficient commands when provided with spatial augmentation (graphic presentation). Keywords: Educational psychology, Cognitive psychology, Human-computer interaction.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 01, 1990
Accession Number
ADA226906

Entities

People

  • Brian C. Hayes
  • Jay Elkerton
  • Kim J. Vicente
  • Robert C. Williges

Organizations

  • Virginia Tech

Tags

Communities of Interest

  • Autonomy
  • Human Systems

DTIC Thesaurus Topics

  • Applied Psychology
  • Cognition
  • Cognitive Systems Engineering
  • Combat Support
  • Computers
  • Educational Psychology
  • Human Factors Engineering
  • Human Systems Integration
  • Human-Computer Interaction
  • Human-Computer Interfaces
  • Human-Machine Interaction
  • Information Processing
  • Information Retrieval
  • Operating Systems
  • Operations Research
  • Psychology
  • Task Performance And Analysis

Readers

  • Team-Based Human-Centered Cognitive Task Decision Making and Information Performance.