thecenter

Activities and Seminars

Sachiko Kinoshita, Macquarie University, Sydney, Australia
 
Date: Oct 13, 2011

What: “Masked priming in the Bayesian Reader”


What: “Masked priming in the Bayesian Reader”

When: Thursday, October 13, 11am

Who: Sachiko Kinoshita, ARC Centre for Excellence in Cognition and its Disorders and Department of Psychology, Macquarie University, Sydney, Australia.

Where: BCBL, Paseo Mikeletegi 69, Floor 2.

Summary:

Masked priming has been used to study automatic – unconscious – processing of words (and other stimuli). In this talk, I present a novel theoretical framework for explaining masked priming effects, as an alternative to the long-held “pre-activation” view of priming. The framework (Norris & Kinoshita, 2008, JEP: General) is based on the Bayesian Reader model of visual word recognition put forward by Dennis Norris (2006, 2009, Psychological Review). The key assumption of the model is that in the recognition of words (and other objects), readers make decisions by combining evidence accumulated by noisy sampling from perceptual input, and prior probability (“priors”) in an optimal, Bayesian fashion. In masked priming, because of spatial and temporal proximity, people do not realize the prime and the target are distinct perceptual events, and hence evidence from the prime is not discounted. This account explains findings that are otherwise puzzling, and provides novel insights into the early stages of word recognition.