Introduction 2 Cognitive styles heterogeneous visual multimodal We have been carrying out an evaluation of the effects of teaching logic with Barwise and Etchemendy's Hyperproof [1]. One of our major findings has been that individual differences between students have a significant effect on students' responses to Hyperproof: their prior cognitive style influences both the overall effectiveness of the teaching regime, and the actual proof structures that students produce under exam conditions [9, 7, 8]. Now, Barwise and Etchemendy designed Hyperproof to support reasoning, in which information from differing modalities---sentential and graphical---is combined, or transferred from one modality to another. It is obviously, therefore, a multimodal system containing a visual sub-system. But given that one group of students benefits particularly from being taught with Hyperproof, we can ask: do they do well because it is a logical system, or do they do well because it is

Computer-based logic proofs are a form of `unnatural' language discourse, but the structure and process of proof generation can be observed in considerable detail, and analysis is leading to a number of general insights. We have been studying how students respond to multimodal logic teaching. Performance measures have already indicated that students' pre-existing cognitive styles have a significant impact on teaching outcome. Furthermore, a large corpus of proofs has been gathered via automatic logging of proof development. This paper applies a series of techniques, including corpus statistical methods, to the proof logs. The results indicate that students' cognitive styles influence the structure of their logical discourse, via their differing methods of handling abstract information, and transferring information between modalities. As well as uncovering different thinking styles in this artificial domain, the observations raise the issue of the importance of individual differences in...

Hyperproof is a logic course that uses a multimodal computer interface in which proofs are constructed. Students learn differently both from the course and, relatedly, within the course. The nature of these differences has been examined in detail, but not without specific reference to established individual difference dimensions. This paper shows that the differences between learning within Hyperproof are concordant with the serialist-holist distinction in strategy use. Equating these differences to an established educational measure provides a means of assessing students' learning from the course as a result of learning within the course, and, conversely, provides a framework within which 2 Holist and Serialist Strategies in Complex Reasoning Tasks: Cognitive Style and Strategy Change the serialist-holist distinction can be cast in more computational terms. The different strategies for constructing proofs in Hyperproof relate to different uses of external structure and working memo...