Precursors and downstream consequences of prediction in language comprehension

Hubbard, Ryan James

Permalink

https://hdl.handle.net/2142/99396 Copy

Description

Title

Precursors and downstream consequences of prediction in language comprehension

Author(s)

Hubbard, Ryan James

Issue Date

2017-12-08

Director of Research (if dissertation) or Advisor (if thesis)

Federmeier, Kara D.

Doctoral Committee Chair(s)

Federmeier, Kara D.

Committee Member(s)

• Cohen, Neal J.
• Dell, Gary S.
• Sahakyan, Lili
• Tanner, Darren

Department of Study

Psychology

Discipline

Psychology

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Date of Ingest

2018-03-13T15:49:06Z

Keyword(s)

• Language
• memory
• prediction
• ERP

Abstract

During language comprehension, the brain rapidly integrates incoming linguistic stimuli to not only incrementally build a contextual representation, but also predict upcoming information. This predictive mechanism leads to behavioral facilitation of processing of expected words, as well as a reduction in amplitude of the N400, a neural response reflecting access of semantic memory. However, little research has identified a behavioral or neurophysiological cost of errors in prediction. Additionally, only recent work has begun to investigate neural activity related to prediction prior to encountering a predicted stimulus. Most work has focused on what happens immediately after a predicted or unpredicted stimulus is encountered. Here, I explore new avenues of research by examining downstream consequences of prediction during language comprehension on future recognition memory. Additionally, I test whether these consequences occur following any violation of predictions, or whether the semantic fit of the violation to the established context plays a role. Finally, I adapt a classic paradigm, word stem completion, to investigate electrophysiological activity following a cue that is modulated by how predictive the outcome is. With this work, I not only have discovered costs of failed and successful predictions and identified neural signals potentially related to generation of predictions, but also have researched prediction in novel ways that can continue to expand and further elucidate how this mechanism affects cognition and changes across populations.

Graduation Semester

2017-12

Type of Resource

text

Permalink

http://hdl.handle.net/2142/99396

Copyright and License Information

Copyright 2017 Ryan Hubbard

Owning Collections