Aims and objectives. The aim of this study is to investigate cognitive ability-demand gap in human-system (agent) interaction using psychometric item response theory. Objectives are - to define ability-demand gap with item and person parameters; to identify gap representative latent parameters; and to investigate gap induced cognitive performances in robust interaction design.
Background. Item response models are widely applied in different research fields including educational psychology , nursing , medical research  and more recently in user interaction study . Most of the study applied item characteristic functions, item information function, test information function and item fit statistics to well perform a meaningful psychometric test with dichotomous and polytomous responses. Recent research identifies a new way, response latitude, the magnitude of inconsistency around a response attitude . It is argues that subjects with narrow response latitude are highly involved in task execution (eg., rating process)than subject with wider response latitude. Subjects with cognitive conflict situation might behave poorly showing wider response latitude. Cognitive dissonance and overload are two such conflict situations effects from higher order cognitive load situation . Although, dissonance is considered as a natural phenomenon, cognitive overload is sometime explained as extreme unsuccessful feeling. Another way to look at this situation is in terms of cognitive lock-up/in. Ergonomists generalizes the term dissonance to lock-up and overload to techno-stress . Conversely, the business literature terms the same situation as cognitive lock-in .We propose that with a latent response model the system should be able to identify cognitive ability-demand gap and higher order cognitive load effects.
Design. We performed a methodological and comparative design. The main proposition in methodological design is that, the highly involved subjects might have low cognitive load than low involved subjects. Therefore, response latitude may be an index of cognitive load. In broader picture, response latitude may be useful in identifying ability-demand gaps in assistive thinking and collaborative sensemaking. Initially the latitude assessment is performed in dichotomous responses with one parameter Rasch model. The later analysis is applied over polytomous responses with extended Rasch modeling. Mental multiplication task interaction data and reconfigured mobile android phone data are used in the evaluation.
Methods. Item responses theories are applied on 24 subjects 30...