Upcoming Talks

Affordances and the Problem of Action Selection: The Endlessly Intriguing Case of Throwing for Distance and Accuracy
Dr Andrew D Wilson
Leeds Beckett University

Lecture Theatre 7, Rendall Building, University of Liverpool
Wednesday 12th November 2014, 1-2pm

Of all the many possible things I might do right now, why did I just do that? This is the problem of action selection and it's been annoying movement scientists for a long time. The current hypotheses involve top-down solutions that restrict our choices down to a manageable number, but these solutions don't work because both the 'right' response and the 'right' restriction are task specific. Action selection therefore requires the perception of action-relevant properties of tasks. Gibson called these properties 'affordances' and understanding skilled action means figuring out what affordances are and how we perceive them. In this talk I will describe how we can do some science on action selection by getting serious about tasks and their affordances, and I will lay this out in the context of throwing for long distances and accuracy.


Why throwing? First, it is that wonderful thing, the nearly uniquely human behaviour. We've been throwing things for fun and profit for thousands of years and it's been so important to our evolutionary success it has left many marks on our anatomy and perceptual systems. Second, it turns out to be the perfect test bed for developing a task dynamical analysis of affordances and action selection in an ecologically valid and interesting setting. I'll walk you through the analysis, show you data from experiments derived from the analysis and explain that data using simulations built according to the analysis as well. This work provides a foundation for lots more science on throwing (there's plenty more to do!) and expands the theoretical and empirical toolbox for embodied cognition research in general.

(Link to my poster on this for VSS 2013)

No comments:

Post a Comment