Browse/search for people

Professor Colin Davis

Modelling the mind

My research attempts to better understand the mechanisms and codes underlying human cognition. I am particularly interested in reading, but I have also conducted research on speech perception and production, memory and social cognition. I use computational modelling to investigate these aspects of cognition. Running simulations of computational models makes it possible to generate predictions that can be tested in behavioural experiments. This work has been supported by the ESRC, the British Academy, and most recently, by a research fellowship from the Leverhulme Trust to explore the self-organisation of the mental lexicon.

Much of my computational modelling has focussed on visual word recognition (e.g., Davis, 2010). This domain is interesting both because it is central to the ability to read (and to understanding why some people have difficulty learning to read) and because our ability to rapidly identify printed words provides a central testing ground for theories about the nature of mental representations (e.g., are they symbolic or subsymbolic, localist or distributed?; how is information about order coded?) and the processes that operate on these representations (e.g., how do we learn both invariant and context-sensitive representations?; what role does top-down feedback play?; how are competitive processes implemented?; to what extent is processing “modular”?).

My experimental work aims to test the predictions of computational models. One technique that I’ve used frequently is masked priming, which exploits subliminal processing to test visual identification processes. Based on simulations of a computational model, I developed a more sensitive version of the masked priming methodology, called “sandwich priming” (Lupker & Davis, 2009). Together with Dr James Adelman (Warwick) I’ve recently coordinated a masked priming experiment with over 1000 participants tested across 14 research institutions. The results of this study provide an empirical database that will be invaluable in testing models of visual word recognition.

Other work, particularly with Jeff Bowers (Bristol) has tackled more general questions about the nature of cognitive representations (Bowers et al., 2009, submitted) and mechanisms (Bowers & Davis, 2004) and the merits of alternative theoretical approaches to studying cognition (Bowers & Davis, 2011, 2012).

Research keywords

  • Computational modelling
  • Reading
  • Visual word recognition
  • Language processing
  • Memory