Dreaming of adaptive interface agents

  • Authors:
  • Bill Tomlinson;Eric Baumer;Man Lok Yau;Paul Mac Alpine;Lorenzo Canales;Andrew Correa;Bryant Hornick;Anju Sharma

  • Affiliations:
  • University of California;University of California;University of California;University of California;University of California;University of California;University of California;University of California

  • Venue:
  • CHI '07 Extended Abstracts on Human Factors in Computing Systems
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

This interactive project uses the metaphor of human sleep and dreaming to present a novel paradigm that helps address problems in adaptive user interface design. Two significant problems in adaptive interfaces are: interfaces that adapt when a user does not want them to do so, and interfaces where it is hard to understand how it changed during the process of adaptation. In the project described here, the system only adapts when the user allows it to go to sleep long enough to have a dream. In addition, the dream itself is a visualization of the transformation of the interface, so that a person may see what changes have occurred. This project presents an interim stage of this system, in which an autonomous agent collects knowledge about its environment, falls asleep, has dreams, and reconfigures its internal representation of the world while it dreams. People may alter the agent's environment, may prevent it from sleeping by making noise into a microphone, and may observe the dream process that ensues when it is allowed to fall asleep. By drawing on the universal human experience of sleep and dreaming, this project seeks to make adaptive interfaces more effective and comprehensible.