Structural priming in question-answer dialogues.

Area of Science:

• Psycholinguistics
• Computational Linguistics
• Cognitive Science

Background:

• Structural priming is a phenomenon where exposure to a linguistic structure increases the likelihood of using that same structure.
• The mechanisms underlying structural priming, specifically whether speakers 'self-prime' (i.e., prime themselves), are debated.
• Existing theories propose either activation-based or error-based accounts for priming.

Purpose of the Study:

• To investigate whether speakers exhibit self-priming in naturalistic question-answer dialogues.
• To determine if a speaker's use of a linguistic structure in one utterance influences their subsequent use of the same structure.
• To differentiate between activation-based and error-based theories of structural priming.

Main Methods:

• Conducted experiments involving simulated phone calls to restaurants.
• Experimenters asked two distinct questions: one about menu serving times and another about closing times.
• Analyzed participants' responses for the use of prepositions, specifically in relation to the experimenter's question structure.

Main Results:

• Participants were more likely to use a preposition in their response when the experimenter's question also contained a preposition.
• Crucially, a speaker's use of a preposition in their first response did not influence their use of a preposition in their second response.
• This indicates a lack of self-priming in the observed dialogues.

Conclusions:

• The findings provide evidence against self-priming in this dialogue context.
• The results support error-based theories of structural priming, suggesting that errors or deviations from expectations drive priming.
• Conversely, the data challenge activation-based accounts, which predict that repeated exposure should lead to self-priming.