Summary: New research shows that repeating non-verbs as well as verbs increases syntactic priming, the tendency to reuse recently heard sentence structures.
Source: HSE
Researchers from the University of Glasgow and HSE/Northumbria report that repeating content words—both verbs and non-verbs—can strengthen syntactic priming, making speakers more likely to reproduce the sentence structure they have just heard.
Human language processing often favors repetition: people tend to reuse syntactic frames they have recently heard or produced. In psycholinguistics this tendency is known as syntactic priming. Earlier studies emphasized the special role of verb repetition in amplifying this effect. New experiments by Christoph Scheepers and Claudine Raffray (University of Glasgow) in collaboration with Andriy Myachykov (HSE and Northumbria University) demonstrate that repetition of other content words—such as nouns—can equally boost syntactic priming. Their findings were published in the Journal of Memory and Language in the article “The lexical boost effect is not diagnostic of lexically-specific syntactic representations.”
Priming is a well-documented psychological phenomenon in which exposure to a stimulus influences responses to subsequent stimuli. It is visible in many everyday behaviors: people unconsciously mirror gestures, adopt similar intonation, or repeat phrasing. Perceptual priming also makes related tasks easier—after working on math problems, for example, a person tends to solve another math problem faster than someone who has been engaged in an unrelated activity.
Classic priming experiments often contrasted simple perceptual tasks—such as recognizing visually similar shapes—with higher-level linguistic tasks that involve meaning and structure. In language, priming operates at multiple levels: lexical (word choice), semantic (meaning), and syntactic (sentence structure). The interactive alignment theory frames priming as a fundamental, largely automatic mechanism that supports communication, reduces cognitive effort, and plays a role in children’s language acquisition and social coordination.
The syntactic priming effect was first documented in the 1980s. For example, after encountering a sentence with a specific syntactic frame, a person is more likely to process and produce the next sentence using the same frame. Scheepers, Raffray, and Myachykov illustrate this with a common ditransitive example: hearing “the girl gave the boy a ball” makes a listener more likely to produce “the artist showed the child an easel” (double object frame) rather than “the artist showed an easel to the child” (prepositional object frame). Such tendencies do not occur every time but are robust across studies.
For many years it was assumed that syntactic priming was independent of lexical repetition. In the late 1990s researchers reported a “lexical boost”: when the same verb appears in both prime and target sentences, syntactic priming strengthens. That finding suggested a privileged role for verbs in linking lexical items to syntactic frames. However, the possibility that repeated nouns or other content words could produce similar boosts had received relatively little attention.
The Glasgow–HSE/Northumbria team systematically tested how repeating different content words affects priming. In their experiments participants first read full sentences (primes) and then produced target sentences from jumbled words presented on a screen. The researchers varied how many and which content words—verbs, agent nouns, recipient nouns, or theme nouns—were repeated between prime and target.
Results showed that repeating nouns (either agent or recipient nouns, and to a lesser degree theme nouns) produced lexical boosts comparable to those produced by repeating verbs. Moreover, when more content words were shared between prime and target, the magnitude of syntactic priming increased accordingly. Across experiments, no special or exclusive role for verbs was observed in producing the lexical boost.
These findings carry implications for accounts of syntactic representation. While verbs are widely acknowledged to influence sentence structure, the current results indicate that the lexical boost is not diagnostic of lexically-specific syntactic representations—that is, the boost does not uniquely reflect verb-linked syntactic frames. Instead, repetition of content words more generally appears to elevate the accessibility of a syntactic frame and thus increase the chance of its reuse.
Source: Liudmila Mezentseva, HSE
Image source: Illustrative image in the public domain
Original research: Christoph Scheepers, Claudine N. Raffray, and Andriy Myachykov, “The lexical boost effect is not diagnostic of lexically-specific syntactic representations,” Journal of Memory and Language. Published online March 17, 2017. DOI: 10.1016/j.jml.2017.03.001
The lexical boost effect is not diagnostic of lexically-specific syntactic representations
Structural priming refers to speakers and listeners unconsciously reusing syntactic structures across utterances. Prior work showed that repeating lexical items can enhance priming, a phenomenon called the lexical boost. The present comprehension-to-production experiments examined whether head constituents (verbs) versus non-head constituents (argument nouns) contribute differently to boosting English ditransitive structure priming. Experiment 1 demonstrated that lexical boosts from repeated agent or recipient nouns (and, to a lesser extent, repeated theme nouns) matched boosts from repeated verbs. Experiments 2 and 3 found that increasing the number of shared content words between primes and targets increased the magnitude of priming, again with no special contribution from verb repetition. The authors conclude that lexical boost effects are not conclusive evidence for lexically-specific syntactic representations, although other evidence supports such representations.
Authors: Christoph Scheepers, Claudine N. Raffray, Andriy Myachykov. Journal of Memory and Language. Published online March 17, 2017. DOI: 10.1016/j.jml.2017.03.001
For inquiries about the article source, contact HSE communications. Please feel free to share this summary of the research.