We'll go over the preliminary work on modeling adaptation of phonetic categories via Bayesian belief updating that Florian and I submitted to the ACL CMCL workshop. I'll also talk a bit about how this framework might be extended to capture patterns of generalization of perceptual learning for one speaker's phonetic categories to another's, and other sorts of generalization.
What would be really helpful from all y'all is suggestions for how this framework might make predictions that are different from existing theories (which are pretty meager) or just straight-up surprising/interesting, and how we could test those experimentally.
Also, if y'all could read the paper and think of ways we could improve/change the framing of it, that would also be very helpful.
Dave
attachment:KleinschmidtJaeger2011.pdf