Clearly, understanding how something as complex as the human mind functions, is a difficult task. Indeed, it is one which has engaged some of the greatest thinkers through history. Today's researcher, however, has a big advantage over his predecessors. The digital computer now offers a unique conceptual tool with which to envisage and model the inner workings of the mind. Indeed, since the computer's inception 50 years ago there have been considerable advances. Yet, despite the wealth of information that has been gathered, there are still very serious gaps in our understanding. Moreover, several basic philosophical questions regarding the nature of mind and its relationship to the world around it remain unresolved.
Nowadays, there is no shortage of ideas. Rather, the biggest problem seems to be the very wealth of different and often conflicting notions. What is needed most is a single coherent framework within which to view these various conceptions. This is where my major effort has so far been devoted. The inscriptor-theory, which I have developed over the past few years, appears to offer just such a framework. Although basically very simple, it naturally supports a plausible, if perhaps, still controversial view of the world.
Currently, I am extending the scope of the theory and attempting to relate it in more detail to existing models of mind. This is raising interesting questions about the origin and nature of symbols, of concepts and of language. Fundamental investigations into the nature of truth, knowledge, meaning and even computation are leading to fascinating insights into cognition, and may ultimately pave the way to a radically new form of "computing" device.
My interest in information retrieval stems primarily from the difficulties I personally faced in finding and organising material related to my A.I. research. The advent of electronic networks combined with the "publish or perish" mentality has resulted in an absolute glut of information. Wading through this mass of potential knowledge, finding useful or possibly relevant bits, storing them and later being able to locate them when needed, is rapidly becoming a full-time task in itself. While computerised information retrieval is certainly not a new idea, it is one which is gaining in importance simply because of the information overload now facing ordinary people the world over. Since the best information retrieval system we know of is the human brain, it seems appropriate to try and apply the ideas from my A.I. research to solve these problems, and hence kill two birds with one stone! Finally, computer-aided-education represents, for me, both a potential testing ground for theories of learning and an opportunity to do more practical, down-to-earth work, whose results are of immediate value.
Dr. David Davenport - Spring 1995