Contributions to Engineering Models of Human-Computer Interaction. Volume 1.
Abstract
This dissertation presents two engineering models of behavior at the human-computer interface; a model of immediate behavior stimulus-response compatibility and a model of transcription typing. Formulated within the architecture of the Model Human Processor of Card, Moran and Newell, these models are able to make zero-parameter, quantitative predictions of human response time in their respective domains. They are also completely integrated, making good predictions about performance on a dual reaction-time/typing task. Parameters of the models are set using response time data from an abbreviation recall experiment. These parameters are then used to make predictions about response time in another abbreviation recall experiment, three classic stimulus-response experiments, and over 29 experiments that reflect robust phenomena associated with transcription typing. These models are the first to make successful predictions across domain boundaries, both within tasks exhibiting stimulus-response compatibility and outside that paradigm to transcription typing. Keywords: Cognitive models; GOMS(Goals Operations Methods Selection). (KR)
Document Details
- Document Type
- Technical Report
- Publication Date
- May 06, 1988
- Accession Number
- ADA195590
Entities
People
- Bonnie E. John
Organizations
- Carnegie Mellon University