System interface effects on human performance: Vehicle control tasks, speed/accuracy modeling, and advanced business visualization
Abstract
The purpose of this document is to present information gathered via experimental methods on the impact on human performance of various interface characteristics. Previous human-computer interface research in the fields of management, industrial engineering, and psychology have primarily focused on the effects of visual display attributes, and then often focusing on relatively simple representations of data. One principle outcome of these research studies has been that task completion time and/or task accuracy was improved when the display of information was related to the characteristics of the task. The first research study described in this document extended that concept to the relationship between a physical input device and completion of a motor task. The results showed that performance was enhanced when particular characteristics of a vehicle control motor task (rotational and translational control) were similar to the characteristics of the physical interface input device. The second research study described in this document attempted to justify using a multiple criteria optimization problem formulation to describe human-system interaction. Researchers in psychology have modeled the functional forms and parameters that describe the tradeoff between accuracy and speed in various situations. However, the goal of the research presented here was to model human-system interaction at a higher level that would include not only the goal of the human, but also the constraints under which they operate. The results of human performance on a vehicle control task indicated that a standard graphical analysis method used in optimization research could adequately describe the human-system interaction, while providing insights into extremely high and low levels of performance. The final research included in this document investigated the value of using the World Wide Web as an experimental tool in data visualization research, while examining the utility of advanced business data visuals. The results of the study suggested that, while useful as a data gathering tool, the WWW does not yet provide the experimental controls necessary to produce reliable data. Another result of this study was that even domain experts had difficulty deriving useful information from such complex business images.
Degree
Ph.D.
Advisors
Altinkemer, Purdue University.
Subject Area
Management|Industrial engineering|Occupational psychology
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.