國立虎尾科技大學 |

Effects of Information Visualization and Task Type on Cognition and Judgment in Human-System Interaction: A Neuroergonomic Approach.

Record Type:	Language materials, printed : Monograph/item
Title/Author:	Effects of Information Visualization and Task Type on Cognition and Judgment in Human-System Interaction: A Neuroergonomic Approach./
Author:	Nuamah, Joseph Kwadwo Gyamfi.
Published:	Ann Arbor : ProQuest Dissertations & Theses, : 2018,
Description:	209 p.
Notes:	Source: Dissertation Abstracts International, Volume: 79-10(E), Section: B.
Contained By:	Dissertation Abstracts International79-10B(E).
Subject:	Industrial engineering. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10785655
ISBN:	9780355982589

Effects of Information Visualization and Task Type on Cognition and Judgment in Human-System Interaction: A Neuroergonomic Approach.
Nuamah, Joseph Kwadwo Gyamfi.

Effects of Information Visualization and Task Type on Cognition and Judgment in Human-System Interaction: A Neuroergonomic Approach. - Ann Arbor : ProQuest Dissertations & Theses, 2018 - 209 p.

Source: Dissertation Abstracts International, Volume: 79-10(E), Section: B.

Thesis (Ph.D.)--North Carolina Agricultural and Technical State University, 2018.

The state of the world in many human system interactions may be modeled as an environmental criterion and cues, with cues providing information about the criterion. Observability of state of the world depends in part on how these cues are mapped to the interface that mediates interaction between the human operator and the environment. Cues need to be presented in a manner that leverages perception and cognitive performance of the human operator's visual system. By adopting a neuroergonomic approach, we combined the Cognitive Continuum Theory with the Cognitive Fit Theory to enhance understanding of how information presentation and task type impact decision performance. We run two separate within-subject experiments. For each experiment, we recorded electroencephalography (EEG) signals of 15 participants while each participant made judgments about water quality based on presented cues. In Experiment 1, we presented an analytical-inducing task in two formats -- graphical and numerical. We used linear Logistic Regression model and traditional Lens Model approach to analyze data from Experiment 1. In Experiment 2, we presented an intuitive-inducing task in two formats -- graphical and numerical. Here, we extended the body of knowledge in judgment analysis research by adopting the Rule-Based Lens Model and employed machine learning models to infer noncompensatory strategies from judgment data. We did not select the machine learning model beforehand. We explored nonlinear machine learning algorithms -- random forest, k-nearest neighbors, C5.0, support vector machine, artificial neural network, and two fuzzy rule-based classification systems: GFS.GCCL and FRBCS.CHI, applied the 10-fold cross validation technique to estimate candidate models' test errors, and selected the model with least error. Results suggest that (1) not all tasks may afford system designers opportunity to display task variables in a way that elicits a particular mode of cognition, (2) EEG laterality index may be a more reliable index of cognitive mode than traditional behavioral and subjective measures, (3) system designers can use psychophysiological information along with behavioral and subjective assessments to decide whether a modified interface produces equivalent human responses as compared to a baseline or whether a modified interface leads to an improvement in human responses, and (4) EEG could be used to complement usability testing by assessing participants' working memory while they work with different visualizations. Overall, we show that the extent to which the potential for intuitive and analytical cognition can be tapped for optimal judgment performance is contingent on the compatibility between the judge's dominant cognitive mode and the nature of the task.

ISBN: 9780355982589Subjects--Topical Terms:

679492
Industrial engineering.

Effects of Information Visualization and Task Type on Cognition and Judgment in Human-System Interaction: A Neuroergonomic Approach.
LDR:03798nam a2200289 4500 001 931650
005 20190716101634.5
008 190815s2018 ||||||||||||||||| ||eng d
020 $a 9780355982589
035 $a (MiAaPQ)AAI10785655
035 $a (MiAaPQ)ncat:10840
035 $a AAI10785655
040 $a MiAaPQ $c MiAaPQ
100 1 $a Nuamah, Joseph Kwadwo Gyamfi. $3 1213846
245 1 0 $a Effects of Information Visualization and Task Type on Cognition and Judgment in Human-System Interaction: A Neuroergonomic Approach.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2018
300 $a 209 p.
500 $a Source: Dissertation Abstracts International, Volume: 79-10(E), Section: B.
500 $a Adviser: Younho Seong.
502 $a Thesis (Ph.D.)--North Carolina Agricultural and Technical State University, 2018.
520 $a The state of the world in many human system interactions may be modeled as an environmental criterion and cues, with cues providing information about the criterion. Observability of state of the world depends in part on how these cues are mapped to the interface that mediates interaction between the human operator and the environment. Cues need to be presented in a manner that leverages perception and cognitive performance of the human operator's visual system. By adopting a neuroergonomic approach, we combined the Cognitive Continuum Theory with the Cognitive Fit Theory to enhance understanding of how information presentation and task type impact decision performance. We run two separate within-subject experiments. For each experiment, we recorded electroencephalography (EEG) signals of 15 participants while each participant made judgments about water quality based on presented cues. In Experiment 1, we presented an analytical-inducing task in two formats -- graphical and numerical. We used linear Logistic Regression model and traditional Lens Model approach to analyze data from Experiment 1. In Experiment 2, we presented an intuitive-inducing task in two formats -- graphical and numerical. Here, we extended the body of knowledge in judgment analysis research by adopting the Rule-Based Lens Model and employed machine learning models to infer noncompensatory strategies from judgment data. We did not select the machine learning model beforehand. We explored nonlinear machine learning algorithms -- random forest, k-nearest neighbors, C5.0, support vector machine, artificial neural network, and two fuzzy rule-based classification systems: GFS.GCCL and FRBCS.CHI, applied the 10-fold cross validation technique to estimate candidate models' test errors, and selected the model with least error. Results suggest that (1) not all tasks may afford system designers opportunity to display task variables in a way that elicits a particular mode of cognition, (2) EEG laterality index may be a more reliable index of cognitive mode than traditional behavioral and subjective measures, (3) system designers can use psychophysiological information along with behavioral and subjective assessments to decide whether a modified interface produces equivalent human responses as compared to a baseline or whether a modified interface leads to an improvement in human responses, and (4) EEG could be used to complement usability testing by assessing participants' working memory while they work with different visualizations. Overall, we show that the extent to which the potential for intuitive and analytical cognition can be tapped for optimal judgment performance is contingent on the compatibility between the judge's dominant cognitive mode and the nature of the task.
590 $a School code: 1544.
650 4 $a Industrial engineering. $3 679492
690 $a 0546
710 2 $a North Carolina Agricultural and Technical State University. $b Industrial and Systems Engineering. $3 1190051
773 0 $t Dissertation Abstracts International $g 79-10B(E).
790 $a 1544
791 $a Ph.D.
792 $a 2018
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10785655