Priming Word Order in Sentence Production

Abstract

When producing a sentence, the speaker needs to place words in linear order. We hypothesized the existence of a linearization process, which imposes order on a constituent structure. T his structure is assumed to be speci® ed with respect to hierarchial relations between constituents but not with respect to word order. We tested this hypothesis in a primed picture description experiment. Speakers of Dutch repeated prime sentences and described target pictures. Word order of prime sentences was manipulated (e.g.``On the table is a ball'' vs.``A ball is on the table''). Both alternatives could be used in the description of unrelated target pictures. In support of our hypothesis, word order was``persistent'': Speakers were more likely to use a given word order, when the prime sentence had that same word order. We argue that our results support the notion of a linearization process and reject the alternative explanation that the results should be attributed to persistent selection of a fully speci® ed syntactic frame.

Syntactic priming in spoken production: Linguistic and temporal interference

“…Experimental studies of syntactic priming have demonstrated that the effect cannot be explained by nonsyntactic factors and have established some of the circumstances under which it occurs: It cannot be explained by repetition of closed-class lexical items (Bock, 1989), or metrical structure or thematic roles (Bock & Loebell, 1990); it is found in Dutch (Hartsuiker & Kolk, 1998;Hartsuiker, Kolk, & Huiskamp, 1999) as well as English; it occurs in a range of construction types (Bock, 1986;Hartsuiker et al, 1999); and it occurs whether the verb is repeated between prime and target or not, but it is stronger if the verb is repeated (Pickering & Branigan, 1998). Researchers have also employed three different methods: spoken picture description (e.g., Bock, 1986;Bock, Loebell, & Morey, 1992), written sentence completion (Pickering & Branigan, 1998), and sentence recall (Potter & Lombardi, 1998).…”

“…Hartsuiker and Kolk (1998) and Hartsuiker et al (1999) also argued for long-term priming on the basis ofa comparison between descriptions produced at the beginning of the experiment following intransitive baseline primes and descriptions produced during the experiment. For example, Hartsuiker and Kolk found that the combined proportion of PO and DO responses (vs. other responses) was higher during the experiment than at the beginning.…”

The functions of structural priming

“…A number of observations in the literature suggest that structural priming exhibits an inverse-preference pattern (Bock, 1986;Ferreira, 2003;Hartsuiker & Kolk, 1998;Hartsuiker, Kolk, & Huiskamp, 1999;Scheepers, 2003). Most of these have shown that structures that were in general less preferred or less common exhibited greater structural priming relative to a neutral baseline, whereas structures that were in general more preferred or more common exhibited less structural priming relative to a neutral baseline.…”

Speech Accommodation Without Priming: The Case of Pitch

“…The IAM predicts that " [a]s the conversation proceeds, it will become increasingly common to use exactly the same set of computations," (Garrod & Pickering, 2004, p. 10). Repeated priming of a representation over the course of a conversation should incrementally heighten its activation, leading to incrementally increasing alignment (Hartsuiker, Kolk, & Huiskamp, 1999). Second, if priming is the mechanism that drives alignment, then alignment effects should persist beyond the local exposure context.…”