Skip to main content
eScholarship
Open Access Publications from the University of California

Limits of Expectation-Based Memory Benefits in Hyperbolic Language Processing

Creative Commons 'BY' version 4.0 license
Abstract

Expectation-based accounts of language comprehension propose that violations of readers' expectations trigger additional processing that can strengthen memory representations. Hyperbolic language provides a useful test case for this assumption, as exaggerated utterances reliably generate expectations and disrupt online comprehension when those expectations are violated. The present study examined whether such hyperbole-induced expectation violations have downstream consequences for memory. Across three experiments, participants read narratives containing hyperbolic utterances followed by explicit quantitative information that was either expectation-consistent or expectation-inconsistent. After a brief delay, participants completed a recognition memory test. Despite reliable online processing disruption, expectation-inconsistent information showed no consistent memory advantage. Instead, memory performance favored expectation-consistent values or showed null effects, revealing limits on expectation-based memory benefits.