We present a detailed process theory of the moment-by-moment working-memory retrievals and associated control structure that subserve sentence comprehension. The theory is derived from the application of independently motivated principles of memory and cognitive skill to the specialized task of sentence parsing. The resulting theory construes sentence processing as a series of skilled associative memory retrievals modulated by similarity-based interference and fluctuating activation. The cognitive principles are formalized in computational form in the Adaptive Control of Thought–Rational (ACT–R) architecture, and our process model is realized in ACT–R. We present the results of 6 sets of simulations: 5 simulation sets provide quantitative accounts of the effects of length and structural interference on both unambiguous and garden-path structures. A final simulation set provides a graded taxonomy of double center embeddings ranging from relatively easy to extremely difficult. The explanation of center-embedding difficulty is a novel one that derives from the model’s complete reliance on discriminating retrieval cues in the absence of an explicit representation of serial order information. All fits were obtained with only 1 free scaling parameter fixed across the simulations; all other parameters were ACT–R defaults. The modeling results support the hypothesis that fluctuating activation and similarity-based interference are the key factors shaping working memory in sentence processing. We contrast the theory and empirical predictions with several related accounts of sentence-processing complexity.

Psycholinguistic research has shown that the influence of abstract syntactic knowledge on performance is shaped by particular sentences that have been experienced. To explore this idea, the authors applied a connectionist model of sentence production to the development and use of abstract syntax. The model makes use of (a) error-based learning to acquire and adapt sequencing mechanisms and (b) meaning–form mappings to derive syntactic representations. The model is able to account for most of what is known about structural priming in adult speakers, as well as key findings in preferential looking and elicited production studies of language acquisition. The model suggests how abstract knowledge and concrete experience are balanced in the development and use of syntax.

This thesis addresses the cognitive basis of syntactic adaptation, which biases speakers to repeat their own syntactic constructions and those of their conversational partners. I address two types of syntactic adaptation: short-term priming and longterm adaptation. I develop two metrics for syntactic adaptation within a speaker and between speakers in dialogue: one for short-term priming effects that decay quickly, and one for long-term adaptation over the course of a dialogue. Both methods estimate adaptation in large datasets consisting of transcribed human-human dialogue annotated with syntactic information. Two such corpora in English are used: Switchboard, a collection of spontaneous phone conversation, and HCRC Map Task, a set of task-oriented dialogues in which participants describe routes on a map to one another. I find both priming and long-term adaptation in both corpora, confirming well-known experimental results (e.g., Bock, 1986b). I extend prior work by showing that syntactic priming effects not only apply to selected syntactic constructions that are alternative realizations of the same semantics, but still hold when a broad

Abstract not found

A thoroughly investigated area in syntactic research is what have often been considered transformationally related alternations; consider (1), (2) and (3) for examples of particle placement, dative shift and preposition stranding respectively. I

Alignment of interlocutors is a well known psycholinguistic phenomenon of great relevance for dialogue systems in general and natural language generation in particular. In this paper, we present the alignmentcapable microplanner SPUD prime. Using a priming-based model of interactive alignment, it is flexible enough to model the alignment behaviour of human speakers to a high degree. This will allow for further investigation of which parameters are important to model alignment and how the human–computer interaction changes when the computer aligns to its users. 1

Speakers tend to reproduce syntactic structures that they have recently comprehended or produced. This structural or syntactic priming occurs despite differences in the particular conceptual or event roles expressed in prime and target sentences (Bock & Loebell, 1990). In two sentence recall studies, we used the tendency of speakers to paraphrase the finite complements of object-raising verbs as infinitive complements (e.g., “John believed that Mary was nice ” as “John believed Mary to be nice”) to test whether an additional conceptual role would affect priming. Prime constructions with identical constituent orders as objectraising infinitives but an additional conceptual role (“John persuaded Mary to be nice”) resulted in fewer paraphrases. Contrasts with other constructions suggest that the critical difference between primes was this extra conceptual role. Thus, subtle differences in conceptual structures can affect how speakers grammatically encode message elements. The meaning of an utterance constrains the form of its expression. For instance, a speaker who wishes to talk about one thing affecting another thing is more likely to create a sentence with two noun phrases than a sentence with only one noun phrase. Given information about what a speaker intends to express and context of the utterance, there is a limited set of constructions that the speaker can felicitously use. Yet, despite the many systematic meaning-form correlations present in languages, the mappings between them are not always one-to-one (for discussion, see

Three experiments investigated whether speakers use constituent ordering as a mechanism for avoiding ambiguities. In utterances like ‘‘Jane showed the letter to Mary to her mother,’ ’ alternate orders would avoid the temporary PPattachment ambiguity (‘‘Jane showed her mother the letter to Mary,’ ’ or ‘‘Jane showed to her mother the letter to Mary’’). A preference judgment experiment confirmed that comprehenders prefer the latter orders for dative utterances when the former order would have contained an ambiguity. Nevertheless, speakers in two on-line production experiments showed no evidence of an ambiguity avoidance strategy. In fact, they were slightly more likely to use the former order when it was ambiguous than when it was not. Speakers ’ failure to disambiguate with ordering cannot be explained by the use of other ambiguity mechanisms, like prosody. A prosodic analysis of the responses in Experiment 3 showed that while speakers generally produced prosodic patterns that were consistent with the syntactic structure, these patterns would not strongly disambiguate the PP-attachment ambiguity. We suggest that speakers do not consistently disambiguate local PP-attachment ambiguities of this type, and in particular do not use constituent ordering for this purpose. Instead, constituent ordering is driven by factors like syntactic weight and lexical bias, which may be internal to the production system.

Syntactic priming during language production is pervasive and well-studied. Hearing, reading, speaking or writing a sentence with a given structure increases the probability of subsequently producing the same structure, regardless of whether the prime and target share lexical content. In contrast, syntactic priming during comprehension has proven more elusive, fueling claims that comprehension is less dependent on general syntactic representations and more dependent on lexical knowledge. In three experiments we explored syntactic priming during spoken language comprehension. Participants acted out double-object (DO) or prepositional-object (PO) dative sentences while their eye movements were recorded. Prime sentences used different verbs and nouns than the target sentences. In target sentences, the onset of the direct-object noun was consistent with both an animate recipient and an inanimate theme, creating a temporary ambiguity in the argument structure of the verb (DO e.g., Show the horse the book; PO e.g., Show the horn to the dog). We measured the difference in looks to the potential recipient and the potential theme during the ambiguous interval. In all experiments, participants who heard DO primes showed a greater preference for the recipient over the theme than those who heard PO primes, demonstrating across-verb priming during online language comprehension. These results accord with priming found in production studies, indicating a role for abstract structural information during comprehension as well as production.

Abstract not found