Interaction with social artefacts

I have a long-running research interest in how people interact with and perceived mentality in non-human systems, particularly animals, computer programs like 'intelligent agent' systems, and models of psychological phenomena like those developed in Hank. 'Social artifacts', like these are increasingly common but tend to be rather pernicious: the animated paper-clip in Microsoft's Office is a good example of the problems in the design of social artifacts like this. In this field, I draw together work on the theory and practice of the Turing test, with 'role-oriented' models of interaction from social psychology, and work in human-computer interaction.

Information retrieval and genre in text

My current work is applying cognitive theories of classification to information retrieval, and in particular to text classification. Genre is an important part of this - documents are not streams of letters in words, they use structure to emphasise certain sections, and rely on human perception as part of this to refine the way people use these documents. In this area, I am looking at how we can use cognitive and social models to refine and improve the quality of information retrieval tools and techologies.

Enhancing learning and assessment through new technology

In Scotland, learning enhancement is an important part of our approach to learning, looking at how we can improve the quality of the learning, and of the students learning experience. Current research projects in this field include Open Mentor (http://www.openmentor.org.uk). In collaboration with Denise Whitelock of the Open University's Institute of Educational Technology, this is a platform which can be used to provide reflective mentoring feedback for tutors providing written feedback on students' assessments.

Cognitive modelling

I (with colleague Paul Mulholland) designed and implemented the Hank visual cognitive modelling language. Within this framework, we developed cognitive models of a number of psychological theories. I developed a range of models in theory of mind, and used the modelling approach to allow comparisons between the different theoretical interpretations in the field. Clayton Lewis of Colorado visited us for three months in 1999, and developed a number of models of learning (both of language, and of concepts) in Hank. For more details, see the Hank web page: http://kmi.open.ac.uk/projects/hank/.

Published articles

• See a full list of publications
• See a few pieces of unpublished work