Memory for modality of presentation: Within-modality discrimination

Two experiments demonstrated that Ss are capable of making within-modality memory discriminations in both visual and auditory modalities. In Experiment I Ss studied mixed lists of pictures and labels representing common objects and were subsequently required to judge whether the original presentation was pictorial or verbal The high level of performance achieved on this task was unaffected by degree of categorical relatedness of items within method of presentation or by instructions to produce visual images when items were presented verbally. In Experiment II Ss demonstrated the ability to remember whether a sentence was originally presented by a male or a female speaker. Some strategies by which within-modality discrimination in memory might be accomplished are discussed.     

Sentence comprehension and relative pronouns

Doubly self-embedded sentences, differing only in whether the relative pronouns were present or deleted, were presented to two groups of Ss who were required to respond to a word beginning with a particular letter in each sentence and to paraphrase the sentence. The results of both the phoneme monitor task and the paraphrase task indicated that comprehension was better when the relative pronouns were present than when they were deleted. These results are consistent with earlier results for the paraphrase task but are not consistent with earlier results for the phoneme monitor task. Possible reasons for the inconsistency were considered. It was concluded that the phoneme monitor task does reflect comprehension difficulty in a manner consistent with the paraphrase task and that deleting the relative pronouns does make comprehending self-embedded sentences more difficult.     

Evidence for language recoding in autistic, retarded and normal children: a re-examination

In a replication and extension of earlier studies by Hermelin & O'Connor, language recoding abilities in autistic, retarded and normal children matched for mental age and digit span, were compared in a verbal recall task. Random word lists, sentences, and anomalous sentences, eight or 12 items in length (for high and low memory span subgroups) were presented and the number of words recalled from each type of input was scored. All low span children recalled sentences better than random lists with normal children superior to retarded and autistic children and the latter group poorer than the retarded group. Autistic children showed a recency effect with both types of input. There were no group differences amongst high span children and sentences were again better recalled than random lists. In Expt II sentences were better recalled than anomalous sentences, with autistic and retarded children equivalent in performance and poorer than normal children. Although low span autistic children were clearly deficient in recall of sentence material when compared with the two control groups, the effect of conditions showed that they were able to use structure to improve recall. Since high span autistic children did not perform differently from controls it is suggested that results from this kind of study may not be generalizable, and that claims for a specific coding deficit in autistic children need further substantiation.     

On-line and off-line reflexive resolution in Japanese logophoric sentences

This study investigates on-line and off-line resolution of a Japanese reflexive,jibun, for logophoric sentences, i.e., complex sentences involving a matrix verb that reflects one's point of view feelings, or state of consciousness, and for nonlogophoric sentences, i.e., complex sentences involving a subordinate adverbial clause. According to Kuroda's (1973) thesis and Chomsky's (1981) binding principle (for nonlogophoric sentences only), the reflexive in the logophoric sentences can be associated with the subject of both a matrix sentence and a subordinate sentence whereas that in the nonlogophoric sentences can only be associated with the subject of the subordinate sentence. In Experiment 1, 42 students were administered a probe-recognition task in which a probe was given for the subject either of the matrix sentence or of the subordinate sentence immediately after the end of the subordinate clause or at the end of a sentence following the matrix verb. Recognition times were faster for a matrix-subject probe than for a subordinatesubject probe regardless of the sentence type and probe position. In Experiment 2, 40 students were administered an on-line antecedent identification task in which they were required to quickly and accurately identify, when given a probe, the antecedent of the reflexive, with the probe given after the reflexive or at the end of a sentence. Regardless of the sentence type, matrix-subject was judged to be the antecedent of the reflexive more often than subordinate-subject, with the effect of probe position being negligible. An offline study required 136 students to indicate the antecedent(s) for the two types of sentence given in their entirety. No effect of sentence type was found. Findings indicate that neither Kuroda's thesis nor Chomsky's binding principle is applied when Japanese speakers parse logophoric and nonlogophoric sentences.Part of this study was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture, No. 05610067. Deep gratitude is due to Professor D. D. Steinberg for reading an earlier version of this paper. Thanks are also due to Ms. S. Ogawa for her help in collecting part of the off-line data.     

The bizarre mnemonic

Summary Five experiments measured effects of bizarre contexts on the free recall of noun triplets after brief retention intervals. More triplets were remembered from bizarre than from common contexts in short mixed lists (12 sentences) when the sentences were presented at a controlled (10 seconds/sentence) rate, regardless of incidental task (rating images for bizarreness, vividness, or memorability). The average number of words/sentence recalled, however, tended to be higher for common than for bizarre contexts. No memory benefit from bizarreness was found for pure lists nor for lists containing more than six triplets in bizarre contexts. The bizarreness effect was less when the subject controlled the rate of presentation. A sixth experiment, which tested recall after immediate and two-day retention intervals, found that the Bizarre/Common Context by Pure/Mixed List interaction increased over longer retention intervals.     

The two meanings of a homophone

Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 Ss were presented with sentences whose main noun was either a homophone or not. Following each sentence, Ss were presented with a single noun and they had to indicate whether it was relevant to the sentence or not. More specifically, irrelevant probes following sentences containing homophones were either irrelevant with respect to both interpretations of the homophone or irrelevant with respect to the present interpretation of the homophone but relevant with respect to its other interpretation. The data suggest that people compare the main noun and the probe, and that in doing so they compute both entries of the homophone noun. No effect was found due to a variation in the delay between the presentation of the sentence and the presentation of the probe.     

Imagery effects and semantic similarity in sentence recognition memory

The idea that subjects often use imagery to discriminate semantically similar sentences was tested in three experiments. In the first experiment, subjects heard subject-verb-object sentences in the context of either a comprehension task or an image-generation task. Their memory for the sentences was tested using a two-alternative forced-choice recognition test in which different types of distractor sentence were used. A sentence semantically similar to the target sentence was one type; a sentence with the same subject and object nouns as the target sentence, but dissimilar in meaning, was another type; and a sentence similar in meaning to one of the stimulus sentences, but not to the target sentence, was a third type. The results showed that the image-generation instructions enhanced later recognition performance, but only for semantically similar test items. A second experiment showed that this finding only holds for high-imagery sentences containing concrete noun concepts. A third experiment demonstrated that the enhanced recognition performance could not be accounted for in terms of a semantic model of test-item discrimination. Collectively, the results were interpreted as providing evidence for the notion that subjects discriminate the semantically similar test items by elaborating the sentence encoding through image processing.     

Short term retention of sentences

A series of experiments was performed on the interaction between the short-term retention of sentences and of digits. In Experiment I a digit span method was used whereby subjects were presented with a sentence followed by a sequence of digits and were required either (a) to recall the sentence first and then the digits or (b) to recall the digits followed by the sentence. Under condition (a) prior recall of the sentence reduced the percentage of digit sequences correctly recalled, while under condition (b) retention of the sentence appeared to have no effect on digit recall. This last finding was confirmed in Experiment II, where the sentences varied both in grammatical complexity and length.

In Experiment III the effect of prior recall of a sentence on the recall of digits was found to depend on the type of sentence used. A correlation was observed between the size of this effect and the time taken to recall a sentence. The rate of forgetting suggested by this observation was comparable to that obtained in Experiment IV, where subjects performed an intervening task that did not involve immediate memory for sentences in the interval between the presentation and recall of a six-digit sequence.

It was concluded from these results that the short-term retention of sentences and of lists of items cannot be explained in terms of some general store of limited capacity.     

Semantic binding,not attentional control,generates coherent global structure in children's narrative memory

Encoding and retrieving global narrative structure influences children's narrative recall. The influence of age and attentional/executive resources on binding processes during sentence list recall was examined in 5- to 6- and 8- to 9-year-old children. Older and younger children showed superior recall of lists, and achieved higher scores on two metrics of chunking; access to different sentences (i.e., number of chunks) and sentence completion (i.e., chunk size), when lists were presented within a coherent global structure. Children's list recall and sentence access, but not their sentence completion scores, were affected by a concurrent self-paced attention-demanding task. Children, unlike adults, engage in active storage of verbal information in thematically related sentence lists. The coherence-advantage effect was stable across age groups and insensitive to the secondary task. Overall, findings imply that semantic binding generates stronger memory representations and superior recall for sentences within a story context than for sentence sets that lack global narrative structure.     

Voluntary forgetting before and after learning has been accomplished

Two experiments, in which voluntary forgetting was measured, are reported. In both experiments, lists of randomly mixed syllables and numbers were projected on a screen, and Ss were instructed before projection which part of the lists to remember or to forget. In the first experiment, the syllables were cued as items to be forgotten in one group, and to be remembered in the other one. At the end of the projection, Ss were instructed to recall freely only the syllables. For the former group, this was an unexpected test. The same list was projected for different groups: one, three, five, seven, and fourteen times, with new randomly mixed items each time. Syllable recall increased with an increase of times of projection of the list in the second group, but not in the first one. Recall of syllables in the first group was, however, above the chance level. These results were best interpreted by assuming a voluntary forgetting mechanism, which operates after the verbal items leave the short term store. The second experiment, a variation of the first, was designed to find the locus of voluntary forgetting mechanisms. The question is whether Ss can voluntarily forget syllables only before they are learned, or also after learning has already taken place. Voluntary forgetting after learning has been achieved was found to be ineffective. These results tend to support the hypothesis that voluntary forgetting processes do not operate in the long-term store and that Ss can block the transmission of to be forgotten items from the short-term store to the long-term store.     

The role of reference fields in learning of artificial language expressions

Abstract.— The learning of artificial language "sentences" as a function of reference fields in different modalities was studied in a concept formation situation. With verbal sentences a visual reference field was found to be more expedient than a verbal reference field. When the sentences were represented in a visual sign-language form, this effect was reversed, thus revealing an interaction effect between sentence modality and reference field modality. Each sentence contained three nonsense items and was presented together with a contextual reference field. However, the denotative content of the items was not specified in advance. In solving the task, the subjects were thus left free to establish their own relations between the sentence items and the reference fields, instead of having to discover predetermined denotative relations between items and context. The findings were taken to support the notion that concept formation and association formation are basically aspects of the same underlying process.     

Varieties of frequency interference in children's recognition of sentence information

The memory of fourth-grade children (9 years, 10 months of age) for information contained in target sentences presented 24 hours earlier was tested by multiple-choice recognition items each consisting of a question about a target sentence followed by three response options. The options consisted of the correct word from the sentence, a new word, and an old, incorrect word that was contained in another sentence on the study trial. The results indicated that the amount of interference produced by old information depended upon the relationship between the sentence in which the old information occurred and the target sentence—being the least when the two were unrelated; the most when the two shared the same subject noun (which also occurred in the stem of the test item for the target sentence); and intermediate when the subjects of the two sentences were synonyms. The interference produced by these manipulations was moderated by instructional strategies and the distance between the target sentence and its variation. The experiment establishes the validity of contextual frequency accrual, a construct logically necessary in extending frequency theory to recognition of information contained in sentences.     

Event-related potential sensitivity to acoustic and semantic properties of terminal words in sentences.

This experiment was concerned with the effects of phonologically correct masking on the electrophysiological responses to terminal words of spoken sentences differing in contextual constraint. Two event-related potential (ERP) components, the N400 and N200, were recorded to the terminal words of high and low constraint sentences in four conditions. In the Control condition, subjects (Ss) simply attended to the sentences with no explicit task instructions. In the Semantic condition, Ss were required to listen to the stimuli in order to make semantic judgements about the terminal word of each sentence. The Control+Masking condition was identical to the Control condition except for the simultaneous presentation of a masking stimulus. The Semantic+Masking condition had Ss listening to sentences in the presence of masking with the task of making semantic judgements about the terminal word of each sentence. ERPs were recorded from Fz, Cz, Pz, T3, and T4 in 10 subjects. Amplitudes of both the N200 and the N400 were sensitive to contextual constraint with larger responses elicited by the terminal words of low constraint sentences. In addition to demonstrating the co-occurrence of the N200 and N400, this experiment highlighted a functional separation between the two components. Masking had no statistically significant effect on N200 latency but N400 latency was delayed in the masked conditions relative to those in the unmasked conditions. It is proposed that the N200 and N400 are manifestations of two different processes; the N200 reflects the acoustic/phonological processing of the terminal word while the N400 reflects the cognitive/linguistic processing. The relationship between the N200 recorded in this experiment and the discrimination N200 is discussed.     

Memory for the pragmatic implications of sentences

A sentence pragmatically implies another sentence when information in the first sentence leads the hearer to expect something that is neither explicitly stated nor necessarily implied by the original sentence. Thus,The safe-cracker put the match to the fuse pragmatically impliesThe safe-cracker lit the fuse. In a cued recall task with sentences containing pragmatic implications, 19% of the items were recalled correctly while 26% of the responses consisted of the pragmatic implications of the original sentences. The data were interpreted as demonstrating the strong interaction of the subjects’ long-term knowledge with the episodic memory for sentences task.     

A discourse on semantic priming

The processing of a word is sometimes affected if earlier words are semantically related to it (semantic “priming”). Priming phenomena have generally been interpreted as reflecting the organization of the mental lexicon. Past studies have shown that priming effects have a very fast rise time and a relatively rapid decay time. This paper investigates the rate of decay of semantic facilitation in both sentences and lists. It was hypothesized that sentence processing involves the construction of a discourse model in which the main topics stay active. If true, then words referring to related objects or events will be processed rapidly even if they occur later in the input, i.e., there will be no decay of facilitation. Three experiments with college students examined the relative time to process a critical word when it was preceded by either a pair of semantically related words or more neutral words. The materials occurred in either sentences or lists, the latter being word-level anagrams of the sentences. Subjects carried out the phonememonitoring task, responding to a word-initial target phoneme that occurred immediately after the critical word. In Experiments I (N = 58) and II (N = 40) 12 words separated the related/neutral words and the critical items. Facilitation in processing the critical word was present in sentences but not in lists. Experiment III (N = 128) showed that the amount of facilitation in sentences was the same when 12 words separated the related/neutral and critical words as when 1.5 words separated them. Thus, there was no evidence obtained here for decay of facilitation in sentences. The results are taken to be consistent with a discourse-model interpretation of semantic facilitation in sentences.     

Similarity information versus relational information: differences in the time course of retrieval

Two experiments are reported that examine the time course of retrieval in a sentence matching procedure. Subjects learned lists of active and passive sentences and were tested with sentences in active or passive, correct or incorrect versions; for example, if "John hit Bill" was a studied sentence, "Bill hit John" would be an incorrect active test sentence. A response signal procedure was used so that accuracy could be measured as a function of time. The data show that sentences containing words from studied sentences are discriminable early in processing from sentences containing all new words, but discrimination of correct from incorrect versions of studied sentences occurs only later in processing (after 600-700 ms). These results demonstrate that different kinds of information are available at different points during the time course of retrieval and so suggest that modifications are required of models that provide only a unitary value for the amount of match between a test probe and information in memory. Early in processing, the growth of accuracy can be explained by a simple model that assumes independent contributions to total amount of match for each of the content words of a sentence, but this independent processing model cannot account for discrimination later in processing. Several, more general, memory models are examined with respect to their abilities to produce independent item information early in processing and relational information later in processing.     

The construction of spatial situation models during reading

Psycholinguistic research faces a major challenge in describing the mental representations readers construct from a text. It is now widely accepted that readers end with a representation of the situation described in the text. However, it is unclear whether this representation allows the activation of elements in accordance with their situation proximity. To answer this question, two experiments were conducted. Participants read texts, sentence by sentence, which gave them instructions about how to arrange items in a layout; they then performed a recognition task. By manipulating the spatial proximity between prime and probe items, this task allowed the measurement of a spatial priming effect. In the first experiment, a larger priming effect was observed for closer items on the spatial layout. The second experiment replicated these findings and showed that the priming effects are better explained by categorical distance than by Euclidean distance.     

The comprehension and verification of ambiguous sentences

When Ss are presented with an ambiguous sentence they tend to interpret it in only one way. If later events warrant, Ss can recover the other meaning, a process which takes time. These conclusions follow from the results of a study in which 40 undergraduate Ss verified whether or not pictures shown at the end of a sentence represented the meaning of the sentence. When ambiguous sentences were presented, the verification time (VT) was no slower than for unambiguous sentences if the picture represented the “expected” meaning (as determined on a pre-test) of the ambiguity. The VT to the picture representing the “unexpected” meaning of the ambiguity was longer than VT to corresponding control sentences.     

Time to read semantically related sentences*

Ss read sentences specifying relations among concepts, with the sentences related to each other in different hierarchical structures. In the case where sentences had implicative relationships, time to read sentences low in the hierarchy was less if sentences higher in the hierarchy had been read previously, and sentences inconsistent with expectations based on earlier items took longer. However, when structural relations did not involve implication about content, time to read the sentences was not very different regardless of whether related sentences had been read previously.