Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Priming Effects in Language Comprehension and Production
Poster Session E, Sunday, September 14, 11:00 am - 12:30 pm, Field House
This poster is part of the Sandbox Series.
Audrey Laun1, Ellen Lau; 1University of Maryland
This MEG study investigates the extent to which human language production and comprehension processes share neural procedures. While many researchers claim that production and comprehension involve tightly interwoven neural processes (Pickering & Garrod 2013), it must be noted that production and comprehension have fundamentally different goals. Language production is a communication problem in which a speaker aims for their listener to develop an intended mental representation. Language comprehension is a perceptual problem in which a comprehender uses a signal to make inferences about the world. Cognitive models like the dual stream model highlight the distinctions between production and comprehension, asserting that they involve diverging neural pathways, with production utilizing the bilateral dorsal stream and comprehension utilizing the left-lateralized ventral stream (Hickok & Poeppel 2007). To investigate the differences between these processes, the current study uses a repetition priming paradigm. It has been strongly established that repetition of a stimulus leads to a reduction in amplitude of neural responses (e.g. Misra & Holcomb 2003). A possible explanation of this reduction is predictive inference – the identical prime increases the perceived likelihood of the target compared to a novel percept, reducing the ‘prediction error’ and thus, the amplitude of neural responses (Grill-Spector, Henson, & Martin 2006; Summerfield et al. 2008). If language comprehension is an inferential process that involves different mechanisms than language production, we expect priming effects to be stronger in comprehension trials because the identical prime provides information to facilitate the perception of a target rather than the endogenous planning of an action. Additionally, if the processes involve distinct physical structures, we expect the locus of priming effects to be different across comprehension and production trials. In a 2020 study, Dirani and Pylkkännen observed that evoked response amplitude differences between primed and unprimed word reading trials were larger than the differences between primed and unprimed object naming trials. However, in this study, prime and target were presented in the same modality in word reading trials, but not in object naming trials, making it challenging to isolate the effects of task. The current study extends their work using a within-subjects experimental design involving two counterbalanced experimental blocks: a primed picture naming block for production and a primed word reading block for comprehension. Trials begin with a spoken word prime in both blocks, meaning that the prime is always presented in a different modality than the target. In the comprehension block, participants read a presented word silently. Some trials are followed by comprehension questions, probing participants on whether or not they read certain words. In the production block, participants name the presented image aloud. Each block contains 150 trials. In 50 of the trials, the spoken word matches the written word/image. In the remaining 100 trials, the spoken word is unrelated to the written word/image. Data collection is already underway and MEG responses for N=8 subjects point towards differences in both the magnitude and location of priming effects across comprehension and production.
Topic Areas: Language Production,