The Relationship between Syntactic Satiation and Syntactic Priming: A First Look

• Published in: Frontiers in psychology Vol. 8; p. 1851
• Main Authors: Do, Monica L., Kaiser, Elsi
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Media S.A 25.10.2017
• Subjects: acceptability judgments; experimental syntax; island effects; processing difficulty; Psychology; satiation; syntactic priming; acceptability judgments; island effects; experimental syntax; processing difficulty; syntactic priming; satiation
• ISSN: 1664-1078; 1664-1078
• DOI: 10.3389/fpsyg.2017.01851

Syntactic satiation is the phenomenon where some sentences that initially seem ungrammatical appear more acceptable after repeated exposures (Snyder, 2000). We investigated satiation by manipulating two factors known to affect syntactic priming, a phenomenon where recent exposure to a grammatical structure facilitates subsequent processing of that structure (Bock, 1986). Specifically, we manipulated (i) (number of sentences between primes and targets) and (ii) (type of phrase repeated across primes and targets). Experiment 1 investigated whether acceptability ratings of Complex-NP Constraint (CNPC) and Subject islands improve as consequence of these variables. If so, priming and satiation may be linked. When primes were separated from targets by one sentence, CNPC islands' acceptability was improved by a preceding island of the same type, but Subject islands' acceptability was not. When prime-target pairs were separated by five sentences, we found no improvement for either island type. Experiment 2 asked whether improvements in Experiment 1 reflected online processing or offline end-of-sentence effects. We used a self-paced reading paradigm to diagnose online structure-building and processing facilitation (Ivanova et al., 2012a) during processing. We found priming for Subject islands when primes and targets were close together, but not when they were further apart. No effects were detected when CNPC islands were close together, but there was a localized effect when sentences were further apart. The disjunction between Experiments 1 and 2 suggests repetition of the structure in Subject islands facilitated online processing but did not 'spill over' to acceptability ratings. Meanwhile, results for CNPC islands suggest that acceptability rating improvements in Experiment 1 may be driven by factors distinct from online processing facilitation. Together, our experiments show that satiation may not be a one-size-fit-all phenomenon but, instead, appears to manifest itself differently for different types of structures. Priming is possible and may be linked to satiation in purportedly "unbuildable" structures (e.g., Subject islands), but not for all types (e.g., CNPC islands). Despite this, it appears that while the types of mechanisms targeting different island types are distinct, they are nevertheless similarly sensitive to the proximity between individual exposures.