Ecological Interface Design: Progress and challenges

1
Ecological Interface DesignProgress and
challenges

• Kim J. Vicente, University of Toronto

2
Ecology

• 4 metatheoretical commitments in human factors
• Reciprocity of person and environments
• Representative design of experiments and
  evaluations
• Primacy of perception
• Start with analyzing

3
The goal of EID

• improve user performance and overall system
  reliability for anticipated and unanticipated
  events in a complex system
• more cognitive resources, higher cognitive
  progress
• one-level-at-a-time display led to the fastest
  detection times.

4
Two Key concepts

• the abstraction hierarchy (Rasmussen, 1985)
• How and why relationships
• Means-ends links
• Work domain Analysis
• Functional purpose
• The goals purpose of the system
• Abstract function
• Underlying laws principles govern the goals of
  system
• Generalized function
• Explain the processes involved in laws
  principles
• Physical function
• The physical components or equipment associated
  with the processes
• Physical form
• The condition, location, physical appearance of
  components

5
Two Key concepts

• the skills, rules, knowledge framework
  (Rasmussen, 1983)
• Three types of behavior or psychological
  processes present in operator information
  processing
• Skill-based behavior
• Very little or no conscious control
• Rule-based level
• Use of rules and procedures to select a course of
  action in a familiar work situation
• Knowledge-based level
• More advanced level of reasoning

6
Progress

• Does EID Improve Performance?
• EID lead to better performance than contemporary
  interface (Reising Sanderson, 1998)
• Be localized to more complex tasks
• Why Does EID Improve Performance?
• Not attributable merely to differences in visual
  form (Ham Yoon, 2001b)
• Functional information is critical (Ham Yoon,
  2001a)
• Abstraction hierarchy analysis

7
Progress

• How can EID Be Leveraged?
• Model-based training(Hunter, Vincente, Tanabe,
  1996)
• Self-explanation (Howie, Vicente, 1998)
• Videotaped replay
• Verbal protocols
• Wholist cognitive style (Pask Scott, 1972)
• Relational thinking ability

8
Application

• many proof-of-concept tests showed EID can be
  applied to large-scale problems.
• not all of the information will fit on one
  display or monitor
• In aviation, EID was not sure to apply.
• network management, Software engineering,
  Medicine, Command Control, Information retrieval

9
CHALLENGES

• Time and Effort to Conduct the Analysis
• Multimodal character
• Visual form, audio channel
• Sensor noise Failure on performance
• Not only EID But also Integrated Systems Design
• Critical Analysis is needed