N400 amplitudes reflect change in a probabilistic representation of meaning: Evidence from a neural network model

Milena Rabovsky

Freie Universität Berlin, Department of Psychology

The N400 component of the event-related brain potential has aroused much interest because it is thought to provide an online measure of meaning processing in the brain. Yet, despite large amounts of data on the N400, its functional basis remains incompletely understood and actively debated. In the talk, I will present a computationally explicit account of the process underlying N400 amplitudes. Specifically, we simulate N400 amplitudes as the change induced by an incoming stimulus in an implicit and probabilistic representation of meaning captured by the hidden unit activation pattern in a neural network model of sentence comprehension. This measure reflects surprise at the semantic level of representation, i.e. the change in implicitly represented conditional probabilities of semantic features of the event described by a sentence, and accounts for a broad range of empirically observed N400 effects.

If you would like to meet with the speaker, please contact Alessandra Zarcone.

Reference:

Rabovsky, M., Hansen, S.S., & McClelland, J.L. (2016). N400 amplitudes reflect change in a probabilisic representation of meaning: Evidence from a connectionist model. Proceedings of the 38th Annual Meeting of the Cognitive Science Society, pp 2045-2050. Cognitive Science Society: Austin, TX.