Learning Sentence Research Articles

Individual Differences in Learning Arbitrary Sentences Presented at Different Rates

This study investigated individual differences in students′ learning of arbitrary sentences when the sentences were presented at different rates. Undergraduate students were shown arbitrary sentences on a computer monitor at various presentation rates. The students were then asked to answer questions about the content of the sentences. The results indicated that (a) both successful and less successful learners profited from additional study time in cued-recall performance; and (b) the cued-recall performance of less successful learners was comparable with that of the successful learners after a longer presentation rate was given to the less successful learners.

One method to increase spontaneous use of an assistive communication device: case study

This report presents a case study of a 19-year-old adolescent male with a severe physical impairment and a mild-to-moderate cognitive deficit. Based upon classroom observations, it was noted that he infrequently used his assistive communication device (Touch Talker) for conversational interaction. A training program was designed to teach (1) accessing of specific three- to six-word sentences and (2) production of the learned sentence following spoken cues. Observations of communication behavior within his daily environment were made to determine use of his communication device. Results indicated rapid learning of the targeted sentences and subsequent spontaneous production of the sentences following spoken cues. A concomitant increase in spontaneous use of his device for daily activities within the classroom and community was also observed.

Picture versus question elaboration on young children's learning of sentences containing high- and low-probability content

The main purpose of the two experiments reported here was to compare the potency of two types of elaboration on children's learning of sentence content: The effects of partial picture adjuncts were compared to the effects produced by answering “why”-questions about the relationships specified in the sentences. Five- to seven-year-old children heard sentences of the form, subject/verb/direct object/preposition/instrument. Sentences contained either a high-probability or a low-probability instrument given the semantic context. In Experiment 1, sentences either were accompanied by a partial picture depicting the sentence action but omitting the instrument or were presented without a partial picture accompaniment. Recall was improved by provision of partial pictures at study. In Experiment 2, the sentences were accompanied by complete pictures depicting the sentence content. In both experiments, questioning significantly reduced recall of high-probability sentences, with recall of instruments affected especially negatively. Evidence is presented that insufficient attention to instruments may have been one mechanism mediating depressed recall of high- compared to low-probability instruments in the questioned conditions. In summary, partial pictures improved cued recall of sentences in this study; in contrast, all significant effects produced by answering why-questions were negative ones (i.e., later recall was reduced following interrogation at study).

Short-term memory, verbal competence, and age as predictors of imagery instructional effectiveness

Children 6 to 11 years of age heard concrete sentences that they were asked to learn. Half the subjects were instructed to construct images representing the sentence meanings. The remaining participants were provided no strategy instructions (control condition). Consistent with previous outcomes, the older children in the imagery condition learned signficantly more than did the older control subjects. There was a slight trend favoring imagery for younger children in the sample. Individual differences in short-term memory and verbal competence were more highly associated with performance in the imagery than in the control condition, such that greater short-term memory and verbal competence predicted better sentence learning in the imagery condition. In particular, short-term memory and verbal competence made unique contributions (relative to age and to each other) to prediction of sentence learning in the imagery condition, but not in the control condition. In short, imagery instruction is more effective with children who are more intellectually competent.

Second language proficiency differences in the learning of semantically-equivalent bilingual sentences

ABSTRACTThree groups of subjects who used English as a second language and who were considered to be at different levels of proficiency in English participated in a study of transfer of learning from English to Yoruba, their native language, and from Yoruba to English. It was predicted that total transfer from one language to the other would decrease with increasing proficiency in English and that transfer from Yoruba to English would be higher than from English to Yoruba at lower levels of proficiency in English. Findings showed rather that total transfer increased with increasing proficiency in English and that transfer from English to Yoruba was higher than from Yoruba to English for all groups. It is concluded that on a verbal transfer task, bilinguals show development from independent to interdependent language systems with increasing proficiency in a second language.

Perceptual learning of synthetic words and sentences

Previous research has shown that the intelligibility of synthetic speech can be improved with training [E. C. Schwab, H. C. Nusbaum, and D. B. Pisoni, Human Factors (in press)]. In the present study, we investigated the relationship between the type of training subjects receive and the pattern of perceptual learning that occurs. Three groups of subjects were given word recognition tests consisting of synthetic words and sentences generated by the Cottax Type‐'n‐Talk, before and after a training period. One group of subjects was trained on isolated synthetic words. A second group was trained on fluent synthetic sentences. A control group received no training at all. Recognition of isolated synthetic words improved equally for both groups trained on synthetic speech compared to the control group. However, overall word recognition in sentences only improved for subjects that were trained on synthetic sentences. These results demonstrate that performance with synthetic sentences will predict performance with isolated words, but the converse may not always hold. [Work supported by AFOSR and NIH.]

Depth of processing and interference effects in the learning and remembering of sentences.

AUTHOR Kane, Janet fadde; Anderson, Richard C. . TITLE Depth of Processing and Interference Effects in the Learning and Remembering of -Sententes. Technical Report No. 21. ,INSTITUTION Bolt, Beranek and.Newman, Inc., Cambridge, Nags.; Illinois Univ., Urbana. Center for the Study o4 Reading. _ SPONS AGENCY National Inst. of Education (DHEW), Washington, D.C. PUB DATE Feb 77 CONTRACT 400-76-0116 GRANT NIE-G-74-0007 NGTE 29p.

Depth of processing and interference effects in the learning and remembering of sentences.

Depth of processing and interference effects in the learning and remembering of sentences.

Convergence‐divergence and the learning of concrete and abstract sentences

School pupils (mean age 17 years) were classified as convergers or divergers on the basis of their performance on the AH5 Intelligence Test and the Uses of Objects test. They were presented with a set of concrete and abstract sentences, followed immediately by a forced‐choice recognition test. There was a significant interaction between abstractness‐concreteness and convergence‐divergence, in which the effect of the abstractness‐concreteness variable was highly significant for convergers, but was non‐significant for divergers. This interaction was attributed to the superior ability of divergers to discover those different interpretations of a single nominal stimulus held to be inherent in abstract, though not in concrete, sentences. In addition, some evidence against Paivio's (1971) dual‐coding model was presented.

BETAHISTINE HYDROCHLORIDE (SERC) IN CEREBROVASCULAR DISEASE: A PLACEBO-CONTROLLED STUDY

A double-blind, placebo-controlled, clinical study was performed to assess the effects of oral betahistine hydrochloride (Serc) on mental impairment and physical disability in patients with established cerebrovascular disease. Fifty-three patients were admitted to the study during 18 months. Forty-five patients completed the study. They received either betahistine 24 mg daily or placebo for eight weeks. Clinical assessments of general functional activity were done and a battery of nine mental function tests was administered pretreatment and at two-weekly intervals during therapy. The results were analysed statistically using distribution-free tests. Significant differences were demonstrated between betahistine and placebo at week 8 of treatment for associate learning, digit retention, general knowledge, orientation, sentence learning and simple arithmetic. These differences were consistently in favour of betahistine at or close to the 5% level of significance. The results of the remaining mental function tests showed no significant differences between betahistine and placebo, but trends were in favour of the former. General functional activity assessments also demonstrated that betahistine-treated patients were significantly better than those on placebo (P less than or equal to 0.05). No untoward side-effects were noted during the study.

Cumulative learning and long-term retention of sentences

The purpose of this experiment was to study the role of the number of simultaneously presented words in the formation of higher-order units. The variable was the number of modifying elements presented over successive trails in the learning of a set of eight sentences. Results indicated that a S can integrate one new element into an already acquired structure, while he cannot process two or four new elements. After a one-week delay, the learning conditions determine the amount of retention: the progressive introduction of one or two new elements into the kernel group which makes for the organization of very strong structures, leads to a better retention.

Computer analysis of EEG wakefulness-sleep patterns during learning of novel and familiar sentences

In learning experiments novel and familiar sentences were presented acoustically to human subjects during EEG slow wave sleep. Frequency spectra of parieto-occipital and temporo-occipital EEGs were analyzed 25 sec before and 5–30, 30–55 and 55–80 sec after presentation. Statistical comparisons between spectra were performed, using the distribution means. Three qualities of learning were distinguished in tests after the termination of sleep: “spontaneous recall”, “recognition” and “no recall-no recognition”. The presentation caused EEG activations of different level and durations. The cases of “no recall-no recognition” were associated with the lowest EEG activations. Significantly higher and longer activations were found when learning was successful, indicating a systematic relationship between level of EEG activation after presentation of the material, and quality of learning. Further, for the same quality of learning, significantly higher activations were associated with novel than with familiar learning material. Thus, the EEG pattern which exists after the input of learning material to the brain reflects the brain functional state whose level (as characterized by EEG wave frequency) and duration (in conjunction with the factors: interference and difficulty of the learning task) determines the possibility of later recall.

Deep and surface structure in children's sentence learning

Fifth-graders were asked to learn 32 syntactically varied, semantically unrelated sentences containing combinations of agentive, objective and instrumental case relations. Five learning trials, each followed by a noun-prompted recall test, were provided. Recall patterns indicated that variations in surface structure complexity (i.e., verb voice and sentential position of noun prompts) exerted less influence on Ss' memory for sentences than deep structures. Several trials were required for Ss to reproduce syntactic details accurately and many transformational errors were observed. Even then, children were unable to learn one syntactic form. Although patterns of recall were more consistent with deep structure predictions, results were not completely supportive of Fillmore's analysis of deep structure in terms of case relations, and the possibility that other non-linguistic means were used to store sentences loomed as an alternative to the deep structure view.

Concretization and sentence learning.

Concretization and sentence learning.

The effect of violating phrase structure rules and selectional restrictions on TEP patterns

Two experiments were conducted to investigate the effects of violating phrase-structure rules and selectional restrictions on the TEP patterns in sentence learning tasks. It was found that violating selectional restrictions affected the between- and within-phrase pattern of responding. When phrase-structure rules were violated, the strings were found to be learned word by word. These findings were related to the problem of surface structure and deep structure analysis of sentence processing.

Feedback and sentence learning

The theoretical functions of external feedback in S-R and close-loop models of verbal learning are presented. Contradictory predictions from the models were tested with a 3 × 3 factorial experiment including three types of feedback and three amounts of rehearsal. Ninety adult Ss were tested individually and were required to learn 39 sentences verbatim. The results were: (a) Feedback facilitated learning when it followed wrong responses. (b) Feedback had no effect on learning following correct responses. (c) Feedback consisting of both the stimulus and the response was superior to no feedback, whereas feedback consisting of only the response did not differ from no feedback.

Imagery and sentence learning.

Imagery and sentence learning.

Learning in dysphasia

The performance of a group of dysphasic patients and of a group of non-dysphasic patients (many of whom had verified unilateral cerebral lesions) was compared on tests of: sentence learning, sentence recall, rhythm learning, rhythm recall, word finding and of non-verbal intelligence. In Experiment II a small group of dysphasic subjects was taught to repeat a sequence of 5 names of objects (immediately after failing to name these 5 objects), and was thereafter tested for the ability to name the objects which previously could not be named. It was found that the dysphasic patients were significantly impaired at sentence learning, sentence recall and word finding, but showed only a trend towards impairment at rhythm learning and not even a trend at rhythm recall nor on the test of non-verbal intelligence. In Experiment II it was found that learning a sequence of object names facilitates, but generally does not ensure perfect performance at, object naming. The findings suggest that there might co-exist two disabilities in dysphasia: verbal learning and verbal recall.

Transfer from free-recall pair learning to free-recall sentence learning

Transfer from free-recall pair learning to free-recall sentence learning was studied in a 3 by 3 factorial experiment The first variable was the intrapair sequence of the List 1 pairs; noun-verb, verb-noun, or noun-noun. Jlie pairs in a given list entered an A-B, A-B, an A-B, CD, or an A-B, A-Br transfer paradigm with the corresponding sequence in the noun-verb-noun sentences in List 2, therefore transfer paradigm was the second variable. The major finding was that, in general, sentence learning was fastest in the A-B, A-B paradigm, intermediate in the A-B, C-D paradigm, and slowest in the A-B, A-Br paradigm These transfer results paralleled the findings of other transfer studies, and were interpreted in terms ofSs using the learned sequences from pair learning to integrate the sentences. Analysis of intrasentence sequence acquisition suggested that the manner in which the sequences are used to integrate the sentence may be quite complex.

Recall of word lists following learning of sentences and of anomalous and random strings

Three experiments examined the recall of word-lists which had been preceded, during learning, by memorization of various strings, sentences, anomalous and random strings (Exps. I and II), kernel sentences, and ten grammatical transformations (supplementary experiment). In Exp. I, recall of the list always followed recall of the string. Experiment II compared four recall conditions: (1) string-recall first, (2) list-recall first, (3) list-recall only, (4) string-recall only. Each recall output was solicited equally often. The S did not know which output would be required until after the learning trial was concluded. The supplementary experiment tested recall under Conditions 1 and 2. Under Recall Condition 1 differences in list-recall were related systematically to variations of the type of preceding string. But when the list was recalled first, or alone, no systematic differences were present. The results were used to evaluate two accounts of the dependency of list recall, under Condition 1, on the type of preceding string.