Shari R. Speer

Intonational Disambiguation in Sentence Production and Comprehension

Speakers prosodic marking of syntactic constituency is often measured in sentence reading tasks that lack realistic situational constraints on speaking. Results from such studies can be criticized because the pragmatic goals of readers differ dramatically from those of speakers in typical conversation. On the other hand, recordings of unscripted speech do not readily yield the carefully controlled contrasts required for many research purposes. Our research employs a cooperative game task, in which two speakers use utterances from a predetermined set to negotiate moves around gameboards. Results from a set of early versus late closure ambiguities suggest that speakers signal this syntactic difference with prosody even when the utterance context fully disambiguates the structure. Phonetic and phonological analyses show reliable prosodic disambiguation in speakers productions; results of a comprehension task indicate that listeners can successfully use prosodic cues to categorize syntactically ambiguous fragments as portions of early or late closure utterances.

The influence of prosodic structure on the resolution of temporary syntactic closure ambiguities

This paper investigates the influence of prosodic structure on the process of sentence comprehension, with a specific focus on the relative contributions of syntactic and prosodic information to the resolution of temporary syntactic closure ambiguities. We argue that prosodic structure provides an initial memory representation for spoken sentences, and that information from this prosodic representation is available to inform syntactic parsing decisions. This view makes three predictions for the processing of temporary syntactic ambiguity: 1. When prosodic and syntactic boundaries coincide, syntactic processing should be facilitated. 2. When prosodic boundaries are placed at misleading points in syntactic structure, syntactic processing should show interference effects. 3. The processing difficulties that have been reliably demonstrated in reading experiments for syntactically complex sentences should disappear when those sentences are presented with a felicitous prosodic structure in listening experiments. These predictions were confirmed by series of experiments measuring end-of-sentence comprehension time and cross-modal naming time for sentences with temporary syntactic closure ambiguities. Sentences with coinciding or conflicting prosodic and syntactic boundaries were compared to a prosodic baseline condition.

Plausibility and argument structure in sentence comprehension

In two experiments, we investigated how reading time was affected by the plausibility of the prepositional phrase in subject-verb-noun-phrase-prepositional-phrase sentences, and the status of the prepositional phrase as argument versus adjunct of the verb. Highly plausible prepositional phrases were read faster than less plausible ones, and argument prepositional phrases were read faster than adjuncts. These effects appeared both in a self-paced reading experiment and in an experiment that measured eye movements during normal reading. The effects of plausibility were substantially larger and longer lasting than the effects of argument status, but both appeared very early in the reading of the prepositional phrase. The implications of these effects for models of parsing and sentence interpretation are discussed.

Effects of speaking rate and paragraph structure on the production of sentence prosody

Four speakers read each of eight paragraphs at four different speaking rates: Normal speaking rate and half, twice, and three times the normal rate. Each paragraph contained five sentences, one each of five types of syntactic structure: Early major syntactic break, early minor syntactic break, late major syntactic break, late minor syntactic break, or syntactic break midway through the sentence. Across paragraphs, sentences of the same syntactic type had the same prosodic foot structure, and the order of syntactic structures within paragraphs was counterbalanced. Care was also taken to make the paragraphs semantically coherent. Phonetic analyses of the prosodic structures produced for these paragraphs are presented, including absolute and relative duration measures for words and silences. In addition, listener judgments of perceived prosodic phrasing are presented. Results will be compared to the durational and pausing patterns predicted by current prosodic production algorithms (Ferreira, 1993; Gee and G...

Prosodic structure in language understanding: Evidence from tone sandhi in Mandarin

Two experiments show that prosodic information plays a crucial role in the processing of sentences of Standard Mandarin Chinese, where local lexical ambiguities may occur due to the operation of a tone sandhi rule. In Chinese, each word is associated with a tone; in this paper, the term Mandarin tone sandhi refers to a phonological rule that changes the first of two consecutive low tones (Tone 3) to a rising tone (Tone 2). As a result, a two-syllable sequence with a rising tone followed by a low tone is ambiguous. In Experiment 1, listeners identified lexical tones for ambiguous, unambiguous, and nonsense words in phrasal contexts where the tone sandhi rule might have applied. Comparable results in the lexical versus nonsense conditions indicate that judgments did not rely simply on lexically stored tonal information, but also made reference to the tonal context of the phrase. In Experiment 2, subjects chose the most likely written English translation for auditory sentences of Mandarin. Global prosodic information was manipulated to create different levels of prosodic closeness between two critical items in a tone sandhi environment, while the syntactic relation between these items was held constant. Results show that listeners relied on the prosodic structure of the phrases to determine whether or not the tone sandhi rule had applied, and consequently to identify individual lexical items. The evidence is taken to support the notion that prosodic structure influences auditory language comprehension processes.

The influence of prosodic phrasal boundaries on the resolution of semantic lexical ambiguity

Two experiments explored the effect of two levels of prosodic phrasing, the intonational phrase (IPh), and phonological phrase (PPh), on the interpretation of ambiguous words in sentences. In experiment 1, polysemous words were presented in semantically neutral sentence‐initial clauses, followed by either an IPh or PPh boundary. A second clause resolved the ambiguity. End‐of‐sentence judgment times showed that reanalysis to the subordinate meaning of the ambiguous word took longer following IPh than PPh boundaries, suggesting that more extensive interpretive processing had taken place following the higher‐level boundary. In experiment 2, strongly‐biased polysemous words were presented in clauses weakly biased toward their subordinate interpretation, followed by either an IPh or PPh boundary. Following IPh boundaries, visually presented semantic associates of the subordinate meaning of the polysemous word were named faster than targets associated with the dominant meaning. This difference was not found for...

Prosodic disambiguation of syntactic ambiguity in discourse context

A cooperative boardgame task was used to examine how native speakers use prosodic structure to resolve syntactic ambiguity in discourse context. The game task required two speakers to use utterances from a predetermined set to negotiate the movement of gamepieces to goal locations. In one condition, the discourse contained two situations that had to be described using the same syntactically ambiguous word sequence. In the other condition, an identical syntactically ambiguous structure was used to describe only one situation. Sentences that could describe two situations had an ambiguous prepositional phrase attachment as in ‘‘I want to move the square with the triangle,’’ in which the move involved either a combined square‐and‐triangle piece or a triangle pushing a square to another position. Sentences describing only one situation involved only a cylinder pushing the square as in ‘‘I want to move the square with the cylinder.’’ Phonological analyses and phonetic analyses of duration and fundamental freque...

Prosodic persistence in music performance and speech production

Does the rate of melodies that listeners hear affect the rate of their performed melodies? Skilled adult pianists performed two short melodies as a measure of their preferred performance rate. Next they heard, on each trial, a computer‐generated performance of a prime melody at a slow or fast rate (600 or 300 ms per quarter‐note beat). Following each prime melody, the pianists performed a target melody from notation. The prime and target melodies were matched for meter and length. The rate of pianists’ target melody performances was slower for performances that followed a slow prime than a fast prime, indicating that pianists’ performances were influenced by the rate of the prime melody. Performance duration was predicted by a model that includes prime and preferred durations. Findings from an analogous speech production experiment show that a similar model predicts speakers’ sentence rate from preferred and prime sentence rates. [Work supported by NIMH Grant 45764 and the Center for Cognitive Science.]

Prosodic phrasal structure and lexical interpretation

In phonological theory, phonological phrases (PPhs) and intonational phrases (IPhs) are distinct levels of prosodic phrasing, with different edge tones, different constraints on well‐formedness, and different acoustic properties. Although many studies have demonstrated prosodic effects on syntactic processing, few studies have examined prosodic effects on semantic processing, and the majority of studies have not compared the effects PPhs and IPhs [but see Kjelgaard (1995)]. It is demonstrated that, with respect to the interpretation of lexical, syntactic, and semantic ambiguities, PPhs and IPhs have distinct processing effects. Results will be presented from several experiments testing lexical ambiguities of several types, using both cross‐modal naming and end‐of‐sentence tasks. For example, it will be shown that reanalysis to the subordinate meaning of a polysemous word presented in a neutral context takes longer when the word is in a preceding IPh than in a preceding PPh, for lexically and syntactically...

Asymmetric development of perception and production of lexical stress in Korean second‐language (L2) learners of English.

This study explores perception and production of lexical stress information in L2 English learners whose L1 employs a fixed rhythmic pattern at the lexical level. Nineteen English L1 speakers and 14 Korean L2 learners of English were trained to learn 16 minimal stress nonword pairs with picture referents which are segmentally disambiguating in the last syllable (/dȝakunaɪ/ vs /dȝakunɚ/). The eye‐tracking perception experiment revealed that English L1 speakers exploited lexical stress information of the first two syllables to spot the target word in the instruction, “Click on the (target word),” whereas Korean L2 learners’ identification of the target word was delayed until the last syllable. In their production of words in a carrier, “This is the (target word),” however, L2 learners used loudness and durational cue to indicate alternative prominence of the first two syllable as did English L1 speakers. These results imply the development of production and processing of lexical stress in L2 word learning c...