Development of prediction in the ventral visual stream
Keywords
No Thumbnail Available
Authors
Issue Date
2019-07-18
Language
en
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Predicting future states of the environment enables us to act rapidly and efficiently in the world around us. According to predictive coding theories, mismatches between our expectations and incoming sensory inputs are represented by the amount of neural activation. Therefore, predicted events lead to relatively reduced neural activation compared to unpredicted events. Events become more predictable as they often follow a preceding event. It has been established that expected stimuli lead to attenuated neural response throughout the ventral visual stream after statistical learning, although it remains unclear what happens during the acquisition of these associations. By using a two-session MRI experimental setup, the effect of expectation on visual processing is investigated during an initial learning session (session one) and after statistical learning has taken place (session two). Unexpectedly, results indicate no effect of expectation during session two. A trend of expectation enhancement is visible in session one. Interestingly, a general decrease in neural response was found in session two compared to session one. Several explanations are proposed to account for the obtained results, ranging from the multimodal quality of the stimulus used, to the design of the study. Further research is necessary to determine which explanation is driving the absence of an effect of expectation.
Description
Citation
Supervisor
Faculty
Faculteit der Sociale Wetenschappen