We discuss and debunk five common assumptions about the interrelation of semantics, syntax, and f... more We discuss and debunk five common assumptions about the interrelation of semantics, syntax, and frequency during sentence processing. In the course of this, we explore the implications of the view that syntax is assigned as the last stage of comprehension rather than the first: Statistically based perceptual strategies propose an initial semantic representation, which then constrains the assignment of syntactic representations. This view accounts for a variety of facts, as well as suggesting some surprising new ones.
This research disproves the hypothesis that less-skilled comprehenders are less able to take adva... more This research disproves the hypothesis that less-skilled comprehenders are less able to take advantage of constraints at all levels of structure. Five studies used self-paced reading, meaning probe judgment, recall, and sentence and word recognition tasks to examine the effect of supportive discourse contexts on sentence processing in skilled and average comprehenders. The results support a model in which comprehenders develop distinct sentence and discourse representations simultaneously, with limited sharing of information and processing resources. The studies suggest three conditions under which discourse information may influence on-line comprehension. To influence discourse integration, comprehenders must have produced a conteptual representation on which discourse processes can operate (the natural unit hypothesis). To influence syntactic processing, comprehenders must assign a sentence representation to the relevant discourse-based prediction (the linguistic prediction hypoth...
Page 1. 238 Computational Linguistics Volume 28, Number 2 Sentence Comprehension: The Integration... more Page 1. 238 Computational Linguistics Volume 28, Number 2 Sentence Comprehension: The Integration of Habits and Rules David J. Townsend and Thomas G. Bever (Montclair State University and University of Arizona) Cambridge ...
Page 1. 238 Computational Linguistics Volume 28, Number 2 Sentence Comprehension: The Integration... more Page 1. 238 Computational Linguistics Volume 28, Number 2 Sentence Comprehension: The Integration of Habits and Rules David J. Townsend and Thomas G. Bever (Montclair State University and University of Arizona) Cambridge ...
... dodgingpunches... 2. Adjectival structure: If the instructor gets all the students involved, ... more ... dodgingpunches... 2. Adjectival structure: If the instructor gets all the students involved, dancing classes... If. Pluralmorphological bias 1. Gerund structure: If the boxers wantto avoid unnecessary injuries, dodging punches . . . 2 ...
Http Dx Doi Org 10 1080 01690969108406938, Dec 13, 2007
... were presented at the annual meetings of the Psychonomics Society and the City University of ... more ... were presented at the annual meetings of the Psychonomics Society and the City University of ... If increased probability of a sentence in discourse reduces the number of choices that the ... must consider at the acoustic level, there should be greater accuracy in detecting a change ...
ABSTRACT Showed 48 kindergarten, 48 2nd-, and 48 4th-grade Ss word-pairs (e.g., cow–horse) and as... more ABSTRACT Showed 48 kindergarten, 48 2nd-, and 48 4th-grade Ss word-pairs (e.g., cow–horse) and asked them to generate a 1-word clue that did not rhyme with the target word (cow) so that another person could guess which was the underlined word. Each S was assigned to a good-communicator or poor-communicator group and was then assigned to a training condition or a no-training condition. Ss in the training condition were taught how to generate effective clues. The results indicate that training fosters an increase in clue generation ability and that communication ability in a word-pair task is modifiable. (6 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Page 1. 238 Computational Linguistics Volume 28, Number 2 Sentence Comprehension: The Integration... more Page 1. 238 Computational Linguistics Volume 28, Number 2 Sentence Comprehension: The Integration of Habits and Rules David J. Townsend and Thomas G. Bever (Montclair State University and University of Arizona) Cambridge ...
We evaluated specification and under-specification hypotheses of verb representation by measuring... more We evaluated specification and under-specification hypotheses of verb representation by measuring eye-fixations during reading. Participants read two sets of verb phrases that differed in object definiteness and in preference for in- vs. for-adverbial phrases. Total fixation time on the adverbial phrase depended on a predicted interaction between verb phrase preference and object definiteness. These results are consistent with claims that verbs specify boundedness information.
Canadian Journal of Psychology Revue Canadienne de Psychologie, 1975
ABSTRACT Previous research has shown that increased Thorndike-Lorge frequency (F) of stimulus-ter... more ABSTRACT Previous research has shown that increased Thorndike-Lorge frequency (F) of stimulus-terms leads to poorer paired-associate learning (PAL), while increased F-level of response-terms facilitates PAL. Therefore, if a word pair is learned in the order A-B, then later tested in reverse (B-A), there are two possibilities. (1) Error level could be the same on the reverse test, implying that performance on the A-B learning trials determined the strength of the B-A association. (2) Error level on the reverse test could revert to what it would have been if the B-A pair had been learned originally. In the present study the experimental group learned A-B word pairs and was tested on corresponding B-A versions. Results showed that, for the experimental group, the relationship between errors and F-level of the A-terms shifted dramatically between the last A-B trial and B-A test trial. On the B-A test trial, this group performed in a manner comparable to that of a control group trained on B-A throughout. Thus, for pairs which differed in F-level of their stimulus-terms, relative performance during the A-B learning trials was not a crucial factor in later B-A performance.
Researchers frequently use data from monitoring tasks to argue that constraints on meaning facili... more Researchers frequently use data from monitoring tasks to argue that constraints on meaning facilitate lower-level processes. An alternate hypothesis is that the processing level that a monitoring task requires interacts with discourse-level processing. Subjects monitored spoken sentences for a synonym (semantic match), a nonsense word (phonological match), or a rhyme (phonologically and semantically constrained matching). The critical targets appeared at the beginning of the final clause in two-clause sentences that began with if, which signals a semantic analysis at the discourse level, or with though, which maintains a surface representation. Synonym-monitoring times were faster for if than for though, nonsense word-monitoring times were faster for though than for if, and rhyme-monitoring times did not differ for if and though. The results show that conjunctions influence how listeners allocate attention to semantic versus phonological information, implying that listeners form the...
Bilingual college students heard "mixed" passages in English and Spanish. After each pa... more Bilingual college students heard "mixed" passages in English and Spanish. After each passage, the subject heard a test sentence which was either identical to one that had appeared in the passage or a modified version of one that had appeared in the passage. Modifications of test sentences involved either (a) grammatical subject substitution or negation so that the meaning of
We discuss and debunk five common assumptions about the interrelation of semantics, syntax, and f... more We discuss and debunk five common assumptions about the interrelation of semantics, syntax, and frequency during sentence processing. In the course of this, we explore the implications of the view that syntax is assigned as the last stage of comprehension rather than the first: Statistically based perceptual strategies propose an initial semantic representation, which then constrains the assignment of syntactic representations. This view accounts for a variety of facts, as well as suggesting some surprising new ones.
This research disproves the hypothesis that less-skilled comprehenders are less able to take adva... more This research disproves the hypothesis that less-skilled comprehenders are less able to take advantage of constraints at all levels of structure. Five studies used self-paced reading, meaning probe judgment, recall, and sentence and word recognition tasks to examine the effect of supportive discourse contexts on sentence processing in skilled and average comprehenders. The results support a model in which comprehenders develop distinct sentence and discourse representations simultaneously, with limited sharing of information and processing resources. The studies suggest three conditions under which discourse information may influence on-line comprehension. To influence discourse integration, comprehenders must have produced a conteptual representation on which discourse processes can operate (the natural unit hypothesis). To influence syntactic processing, comprehenders must assign a sentence representation to the relevant discourse-based prediction (the linguistic prediction hypoth...
Page 1. 238 Computational Linguistics Volume 28, Number 2 Sentence Comprehension: The Integration... more Page 1. 238 Computational Linguistics Volume 28, Number 2 Sentence Comprehension: The Integration of Habits and Rules David J. Townsend and Thomas G. Bever (Montclair State University and University of Arizona) Cambridge ...
Page 1. 238 Computational Linguistics Volume 28, Number 2 Sentence Comprehension: The Integration... more Page 1. 238 Computational Linguistics Volume 28, Number 2 Sentence Comprehension: The Integration of Habits and Rules David J. Townsend and Thomas G. Bever (Montclair State University and University of Arizona) Cambridge ...
... dodgingpunches... 2. Adjectival structure: If the instructor gets all the students involved, ... more ... dodgingpunches... 2. Adjectival structure: If the instructor gets all the students involved, dancing classes... If. Pluralmorphological bias 1. Gerund structure: If the boxers wantto avoid unnecessary injuries, dodging punches . . . 2 ...
Http Dx Doi Org 10 1080 01690969108406938, Dec 13, 2007
... were presented at the annual meetings of the Psychonomics Society and the City University of ... more ... were presented at the annual meetings of the Psychonomics Society and the City University of ... If increased probability of a sentence in discourse reduces the number of choices that the ... must consider at the acoustic level, there should be greater accuracy in detecting a change ...
ABSTRACT Showed 48 kindergarten, 48 2nd-, and 48 4th-grade Ss word-pairs (e.g., cow–horse) and as... more ABSTRACT Showed 48 kindergarten, 48 2nd-, and 48 4th-grade Ss word-pairs (e.g., cow–horse) and asked them to generate a 1-word clue that did not rhyme with the target word (cow) so that another person could guess which was the underlined word. Each S was assigned to a good-communicator or poor-communicator group and was then assigned to a training condition or a no-training condition. Ss in the training condition were taught how to generate effective clues. The results indicate that training fosters an increase in clue generation ability and that communication ability in a word-pair task is modifiable. (6 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Page 1. 238 Computational Linguistics Volume 28, Number 2 Sentence Comprehension: The Integration... more Page 1. 238 Computational Linguistics Volume 28, Number 2 Sentence Comprehension: The Integration of Habits and Rules David J. Townsend and Thomas G. Bever (Montclair State University and University of Arizona) Cambridge ...
We evaluated specification and under-specification hypotheses of verb representation by measuring... more We evaluated specification and under-specification hypotheses of verb representation by measuring eye-fixations during reading. Participants read two sets of verb phrases that differed in object definiteness and in preference for in- vs. for-adverbial phrases. Total fixation time on the adverbial phrase depended on a predicted interaction between verb phrase preference and object definiteness. These results are consistent with claims that verbs specify boundedness information.
Canadian Journal of Psychology Revue Canadienne de Psychologie, 1975
ABSTRACT Previous research has shown that increased Thorndike-Lorge frequency (F) of stimulus-ter... more ABSTRACT Previous research has shown that increased Thorndike-Lorge frequency (F) of stimulus-terms leads to poorer paired-associate learning (PAL), while increased F-level of response-terms facilitates PAL. Therefore, if a word pair is learned in the order A-B, then later tested in reverse (B-A), there are two possibilities. (1) Error level could be the same on the reverse test, implying that performance on the A-B learning trials determined the strength of the B-A association. (2) Error level on the reverse test could revert to what it would have been if the B-A pair had been learned originally. In the present study the experimental group learned A-B word pairs and was tested on corresponding B-A versions. Results showed that, for the experimental group, the relationship between errors and F-level of the A-terms shifted dramatically between the last A-B trial and B-A test trial. On the B-A test trial, this group performed in a manner comparable to that of a control group trained on B-A throughout. Thus, for pairs which differed in F-level of their stimulus-terms, relative performance during the A-B learning trials was not a crucial factor in later B-A performance.
Researchers frequently use data from monitoring tasks to argue that constraints on meaning facili... more Researchers frequently use data from monitoring tasks to argue that constraints on meaning facilitate lower-level processes. An alternate hypothesis is that the processing level that a monitoring task requires interacts with discourse-level processing. Subjects monitored spoken sentences for a synonym (semantic match), a nonsense word (phonological match), or a rhyme (phonologically and semantically constrained matching). The critical targets appeared at the beginning of the final clause in two-clause sentences that began with if, which signals a semantic analysis at the discourse level, or with though, which maintains a surface representation. Synonym-monitoring times were faster for if than for though, nonsense word-monitoring times were faster for though than for if, and rhyme-monitoring times did not differ for if and though. The results show that conjunctions influence how listeners allocate attention to semantic versus phonological information, implying that listeners form the...
Bilingual college students heard "mixed" passages in English and Spanish. After each pa... more Bilingual college students heard "mixed" passages in English and Spanish. After each passage, the subject heard a test sentence which was either identical to one that had appeared in the passage or a modified version of one that had appeared in the passage. Modifications of test sentences involved either (a) grammatical subject substitution or negation so that the meaning of
Uploads
Papers by David Townsend