David Allbritton

RELIABILITY OF PROSODIC CUES FOR RESOLVING SYNTACTIC AMBIGUITY

Although previous research has shown that listeners can use prosody to resolve syntactic ambiguities in spoken sentences, it is not clear whether naive, untrained speakers in experimental situations ordinarily produce the prosodic cues necessary for disambiguating such sentences. In a series of experiments, the authors found that neither professional nor untrained speakers consistently produced such prosodic cues when simply reading ambiguous sentences in a disambiguating discourse context. Speakers who were aware of the ambiguities and were told to intentionally pronounce the sentences with one meaning of the other, however, did produce sufficient prosodic cues for listeners to identify the intended meanings.

Going Beyond the Problem Given: How Human Tutors Use Post-Solution Discussions to Support Transfer

Recent studies reveal that human tutoring sessions do not always end when the student has solved a problem. Instead, tutors and students frequently use a post-practice discussion to bring new topics to the table or to continue problem-solving discussions. One of the main roles of post-practice dialogues is to support transfer--that is, the students ability to apply concepts and adapt familiar solution strategies to unfamiliar problems. Several developers of intelligent tutoring systems have implemented post-practice modules, with similar aims. However, in contrast to the integrated instructional planning that human tutors apparently perform, automated planning of reflective activities is typically done independently of instructional planning during problem solving. We present a framework for describing reflective plans that are distributed between problem solving and debrief and evidence that reflective discussions support transfer in elementary mechanics.

METAPHOR-BASED SCHEMAS AND TEXT REPRESENTATIONS : MAKING CONNECTIONS THROUGH CONCEPTUAL METAPHORS

Four experiments were conducted to examine the role of metaphor-based schemas in text comprehension and representation. In Experiment 1, schemas facilitated recognition judgements for schema-related sentences that had been presented in a text. Similar facilitation was found for the recognition of individual words in Experiments 2, 3, and 4. The results are interpreted as evidence for the use of metaphor-based schemas to link elements within a text representation.

Research Methods Tutor: evaluation of a dialogue-based tutoring system in the classroom.

Research Methods Tutor (RMT) is a dialogue-based intelligent tutoring system for use in conjunction with undergraduate psychology research methods courses. RMT includes five topics that correspond to the curriculum of introductory research methods courses: ethics, variables, reliability, validity, and experimental design. We evaluated the effectiveness of the RMT system in the classroom using a nonequivalent control group design. Students in three classes (n = 83) used RMT, and students in two classes (n = 53) did not use RMT. Results indicated that the use of RMT yielded strong learning gains of 0.75 standard deviations above classroom instruction alone. Further, the dialogue-based tutoring condition of the system resulted in higher gains than did the textbook-style condition (CAI version) of the system. Future directions for RMT include the addition of new topics and tutoring elements.

Gender and race in predicting achievement in computer science

In the study described here, 65 prospective computer or information science majors worked through a tutorial on the basics of Perl. Eighteen students were African American. All actions were recorded and time-stamped, allowing us to investigate the relationship among six factors that we believed would predict performance in an introductory computer science (CS) course (as measured by course grade) and how much students would learn from the tutorial (as measured by the difference between pre-test and post-test performance). These factors are: preparation (SAT score, number of previous CS courses taken, and pretest score), time spent on the tutorial as a whole and on individual chapters, amount and type of experimentation, programming accuracy and/or proficiency, approach to materials that involve mathematical formalisms, and approach to learning highly unfamiliar material (pattern-matching procedures). Gender and race differences with respect to these factors were also investigated.

Tutoring Bilingual Students with an Automated Reading Tutor that Listens

Children from non-English-speaking homes are doubly disadvantaged when learning English in school. They enter school with less prior knowledge of English sounds, word meanings, and sentence structure, and they get little or no reinforcement of their learning outside of the classroom. This article compares the classroom standard practice of sustained silent reading with the Project LISTEN Reading Tutor which uses automated speech recognition to “listen” to children read aloud, providing both spoken and graphical feedback. Previous research with the Reading Tutor has focused primarily on native speaking populations. In this study 34 Hispanic students spent one month in the classroom and one month using the Reading Tutor for 25 minutes per day. The Reading Tutor condition produced significant learning gains in several measures of fluency. Effect sizes ranged from 0.55 to 1.27. These dramatic results from a one-month treatment indicate this technology may have much to offer English language learners.

Bias in Auditory Priming

Priming for previously studied words in an implicit auditory memory task has been interpreted as evidence for a presemantic perceptual representation system that encodes acoustic representations of words (B. A. Church & D. L. Schacter, 1994). In this article, 3 experiments provided evidence that such priming may result instead from a bias to respond with studied words. In forced-choice identification with similar alternative choices, there was no overall improvement in performance due to prior study. Benefits for studied test words were offset by costs for similar but nonstudied test words. Prior study had no effect when forced-choice alternatives were dissimilar. The data are discussed in relation to current models of auditory information processing and a new model (R. Ratcliff & G. McKoon, in press) for priming in visual word identification.

Irrelevant speech effects and sequence learning

The irrelevant speech effect is the finding that performance on serial recall tasks is impaired by the presence of irrelevant background speech. According to the object-oriented episodic record (O-OER) model, this impairment is due to a conflict of order information from two different sources: the seriation of the irrelevant speech and the rehearsal of the order of the to-be-remembered items. We tested the model’s prediction that irrelevant speech should impair performance on other tasks that involve seriation. Experiments 1 and 2 verified that both an irrelevant speech effect and a changing state effect would obtain in a between-subjects design in which a standard serial recall measure was used, allowing employment of a between-subjects design in subsequent experiments. Experiment 3 showed that performance on a sequence-learning task was impaired by the presence of irrelevant speech, and Experiment 4 verified that performance is worse when the irrelevant speech changes more (the changing state effect). These findings support the prediction made by the O-OER model that one essential component to the irrelevant speech effect is serial order information.

Sentential context effects on lexical decisions with a cross-modal instead of an all-visual procedure.

J. L. Nicol and D. Swinney (1989) reported facilitation in a cross-modal lexical-decision task as evidence that implicit objects of verbs (WH-traces) are reinstated during comprehension. G. McKoon and R. Ratcliff (1994) found the same priming effects in the absence of implicit objects, suggesting that the effects are attributable to some factor other than a syntactic process that would fill in implicit objects. J. L. Nicol, J. D. Fodor, and D. Swinney (1994) questioned the relevance of McKoon and Ratcliffs findings because they were obtained with all-visual rather than cross-modal presentation. In 2 experiments, the authors replicated McKoon and Ratcliffs results using cross-modal lexical decision.

A study to identify predictors of achievement in an introductory computer science course

In the study reported on here, 65 prospective computer or information science majors (47 male, 18 female) worked through a tutorial on the basics of Perl. All actions were recorded and time-stamped, allowing us to investigate the relationship between six factors that we believed would predict performance in an introductory computer science (CS) course (as measured by course grade) and how much students would learn from the tutorial (as measured by gain score from pre-test to post-test). These factors are: preparation (SAT score, number of previous CS courses taken, and pre-test score), time spent on the tutorial as a whole and on individual sections, amount and type of experimentation, programming accuracy and/or proficiency, approach to materials that involve mathematical formalisms, and approach to learning highly unfamiliar material (string manipulation procedures). Gender differences with respect to these factors were also investigated.Predictors of grade and gain score included SAT score, pre-test score (negatively correlated with gain), time (negatively correlated with gain and grade), and various measures of programming accuracy and/or proficiency---for example, the total number of program runs that contained errors (negatively correlated with grade and gain). Several measures of experimentation predicted gain score. Experimentation also predicted grade, but only as applied to the least familiar tutorial material. Although experimentation was practiced throughout the tutorial by both sexes, male and female students differed with respect to the types of tutorial topics and tasks they experimented with and the degree of experimentation---for example, male students were more likely to write programs not suggested by the tutorial. These findings suggest that a tutorial such as the one used in this study could serve as an instrument to identify students who are likely to succeed (or not) in an introductory CS course, and that instructional interventions to promote achievement should encourage experimentation, reflection on the results of experiments, care and accuracy.