Child-computer interaction and the value of help facilities in promoting logical reasoning performance

Cockerton-Turner, Tracey Anne (1990) Child-computer interaction and the value of help facilities in promoting logical reasoning performance. University of Southampton, Doctoral Thesis.

Abstract

The initial aim of this thesis was to explore child-computer interaction through the use of various `help' facilities and to determine the nature of help facilities provided by the computer, which are successful in promoting logical reasoning performance. An initial observational study focused on children's collaborative problem solving behaviour with a manual version of a physics task. Peer-interaction was analysed to reveal particular processes influencing their problem solving performance. From this basis a Help facility was designed to accompany a computer-presented version of the task, which would have some of the attributes of a `social other'. Only paired subjects who did not have Help available performed significantly better than those who worked alone and/or with Help. Analysis of video data and questionnaire data revealed that the poor performance of the Help users could be attributed to their understanding of the nature of Help facilities, together with the design of the interface and the learning ethos of the classroom. In a second empirical investigation, a `scaffolding' approach to learning was applied to the design of a context-sensitive Help facility. This appeared to benefit both pairs and individuals in terms of performance, over and above that of a comparable facility, where subjects had to request Help. The problem of not knowing when one needs help appeared to be a major influential factor. Analyses of such child-computer interactions in the educational context, appeared directly relevant to identifying and optimizing the conditions of computer use conducive to cognitive development. When a `user-sensitive' Help facility is designed on the basis of an analysis of children's errors, again the system-initiated version promotes logical reasoning performance, over and above that of a control (no-Help) group and a Request-Help group. In the final study adult performance on the task was found to be significantly different from that of the 10-11 year olds, although their performance is enhanced when a `pseudo intelligent Help facility' is provided. However, when adults are required to collaborate on the task via a computer, success is not guaranteed. Whether people communicate at, with or via the computer is a crucial issue, and as in this case, may determine the success of one's performance, both for adults and children.

This record has no associated files available for download.

Contributors

Download statistics