Computational modeling of human performance in multiple monitor environments with ACT-R cognitive architecture

Hyungseok Oh, Seongsik Jo, Rohae Myung

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Much research on modeling human performance associated with visual perception is formulated by schematic models based on neural mechanisms or cognitive architectures. But, these two modeling paradigms are limited in the domains of multiple monitor environments. Although the schematic model based on neural mechanisms can represent human visual systems in multiple monitor environments by providing a detailed account of eye and head movements, these models cannot easily be applied in complex cognitive interactions. On the other hand, the cognitive architectures can model the interaction of multiple aspects of cognition, but these architectures have not focused on modeling the visual orienting behavior of eye and head movements. Thus, in this study, a specific cognitive architecture, which is ACT-R, is extended by an existing schematic model of human visual systems based on neural mechanisms in order to model human performance in multiple monitor environments more accurately. And, this study proposes a method of modeling human performance using the extended ACT-R. The proposed method is validated by an experiment, confirming that the proposed method is able to predict human performance more accurately in multiple monitor environments. Relevance to industry: Predicting human performance with a computational model can be used as an alternative method to implementing iterative user testing for developing a system interface. The computational model in this study can predict human performance in multiple monitor environments, so that the model can be applied early on in the design phase, to evaluate the system interface in multiple monitor environments.

Original languageEnglish
Pages (from-to)857-865
Number of pages9
JournalInternational Journal of Industrial Ergonomics
Volume44
Issue number6
DOIs
Publication statusPublished - 2014 Nov 1

Fingerprint

performance
Schematic diagrams
Head Movements
Eye Movements
Visual Perception
visual perception
Cognition
interaction
Industry
cognition
paradigm
industry
experiment
Research
Testing
Experiments

ASJC Scopus subject areas

  • Human Factors and Ergonomics
  • Public Health, Environmental and Occupational Health

Cite this

Computational modeling of human performance in multiple monitor environments with ACT-R cognitive architecture. / Oh, Hyungseok; Jo, Seongsik; Myung, Rohae.

In: International Journal of Industrial Ergonomics, Vol. 44, No. 6, 01.11.2014, p. 857-865.

Research output: Contribution to journalArticle

@article{a0317b618019498394b55a2eeff5a7d0,
title = "Computational modeling of human performance in multiple monitor environments with ACT-R cognitive architecture",
abstract = "Much research on modeling human performance associated with visual perception is formulated by schematic models based on neural mechanisms or cognitive architectures. But, these two modeling paradigms are limited in the domains of multiple monitor environments. Although the schematic model based on neural mechanisms can represent human visual systems in multiple monitor environments by providing a detailed account of eye and head movements, these models cannot easily be applied in complex cognitive interactions. On the other hand, the cognitive architectures can model the interaction of multiple aspects of cognition, but these architectures have not focused on modeling the visual orienting behavior of eye and head movements. Thus, in this study, a specific cognitive architecture, which is ACT-R, is extended by an existing schematic model of human visual systems based on neural mechanisms in order to model human performance in multiple monitor environments more accurately. And, this study proposes a method of modeling human performance using the extended ACT-R. The proposed method is validated by an experiment, confirming that the proposed method is able to predict human performance more accurately in multiple monitor environments. Relevance to industry: Predicting human performance with a computational model can be used as an alternative method to implementing iterative user testing for developing a system interface. The computational model in this study can predict human performance in multiple monitor environments, so that the model can be applied early on in the design phase, to evaluate the system interface in multiple monitor environments.",
keywords = "ACT-R, Cognitive architecture, Eye-head coordination, Gaze control model, Human performance, Multiple monitor environments",
author = "Hyungseok Oh and Seongsik Jo and Rohae Myung",
year = "2014",
month = "11",
day = "1",
doi = "10.1016/j.ergon.2014.09.004",
language = "English",
volume = "44",
pages = "857--865",
journal = "International Journal of Industrial Ergonomics",
issn = "0169-8141",
publisher = "Elsevier",
number = "6",

}

TY - JOUR

T1 - Computational modeling of human performance in multiple monitor environments with ACT-R cognitive architecture

AU - Oh, Hyungseok

AU - Jo, Seongsik

AU - Myung, Rohae

PY - 2014/11/1

Y1 - 2014/11/1

N2 - Much research on modeling human performance associated with visual perception is formulated by schematic models based on neural mechanisms or cognitive architectures. But, these two modeling paradigms are limited in the domains of multiple monitor environments. Although the schematic model based on neural mechanisms can represent human visual systems in multiple monitor environments by providing a detailed account of eye and head movements, these models cannot easily be applied in complex cognitive interactions. On the other hand, the cognitive architectures can model the interaction of multiple aspects of cognition, but these architectures have not focused on modeling the visual orienting behavior of eye and head movements. Thus, in this study, a specific cognitive architecture, which is ACT-R, is extended by an existing schematic model of human visual systems based on neural mechanisms in order to model human performance in multiple monitor environments more accurately. And, this study proposes a method of modeling human performance using the extended ACT-R. The proposed method is validated by an experiment, confirming that the proposed method is able to predict human performance more accurately in multiple monitor environments. Relevance to industry: Predicting human performance with a computational model can be used as an alternative method to implementing iterative user testing for developing a system interface. The computational model in this study can predict human performance in multiple monitor environments, so that the model can be applied early on in the design phase, to evaluate the system interface in multiple monitor environments.

AB - Much research on modeling human performance associated with visual perception is formulated by schematic models based on neural mechanisms or cognitive architectures. But, these two modeling paradigms are limited in the domains of multiple monitor environments. Although the schematic model based on neural mechanisms can represent human visual systems in multiple monitor environments by providing a detailed account of eye and head movements, these models cannot easily be applied in complex cognitive interactions. On the other hand, the cognitive architectures can model the interaction of multiple aspects of cognition, but these architectures have not focused on modeling the visual orienting behavior of eye and head movements. Thus, in this study, a specific cognitive architecture, which is ACT-R, is extended by an existing schematic model of human visual systems based on neural mechanisms in order to model human performance in multiple monitor environments more accurately. And, this study proposes a method of modeling human performance using the extended ACT-R. The proposed method is validated by an experiment, confirming that the proposed method is able to predict human performance more accurately in multiple monitor environments. Relevance to industry: Predicting human performance with a computational model can be used as an alternative method to implementing iterative user testing for developing a system interface. The computational model in this study can predict human performance in multiple monitor environments, so that the model can be applied early on in the design phase, to evaluate the system interface in multiple monitor environments.

KW - ACT-R

KW - Cognitive architecture

KW - Eye-head coordination

KW - Gaze control model

KW - Human performance

KW - Multiple monitor environments

UR - http://www.scopus.com/inward/record.url?scp=84919396632&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84919396632&partnerID=8YFLogxK

U2 - 10.1016/j.ergon.2014.09.004

DO - 10.1016/j.ergon.2014.09.004

M3 - Article

AN - SCOPUS:84919396632

VL - 44

SP - 857

EP - 865

JO - International Journal of Industrial Ergonomics

JF - International Journal of Industrial Ergonomics

SN - 0169-8141

IS - 6

ER -