The contribution of phonological short-term memory to artificial grammar learning

Three experiments investigated the contribution of phonological short-term memory (STM) to grammar learning by manipulating rehearsal during study of an auditory artificial grammar made up from a vocabulary of spoken Mandarin syllables. Experiment 1 showed that concurrent, irrelevant articulation impaired grammar learning compared with a nonverbal control task. Experiment 2 replicated and extended this finding, showing that repeating the grammatical strings at study improved grammar learning compared with suppressing rehearsal or remaining silent during learning. Experiment 3 found no effects of rehearsal on grammar learning once participants had learned the component syllables. The findings suggest that phonological STM aids artificial grammar learning via effects on vocabulary learning.     

Stimulus modality and working memory performance in Greek children with reading disabilities: Additional evidence for the pictorial superiority hypothesis

This study investigated the effects of stimulus presentation modality on working memory performance in children with reading disabilities (RD) and in typically developing children (TDC), all native speakers of Greek. It was hypothesized that the visual presentation of common objects would result in improved learning and recall performance as compared to the auditory presentation of stimuli. Twenty children, ages 10–12, diagnosed with RD were matched to 20 TDC age peers. The experimental tasks implemented a multitrial verbal learning paradigm incorporating three modalities: auditory, visual, and auditory plus visual. Significant group differences were noted on language, verbal and nonverbal memory, and measures of executive abilities. A mixed-model MANOVA indicated that children with RD had a slower learning curve and recalled fewer words than TDC across experimental modalities. Both groups of participants benefited from the visual presentation of objects; however, children with RD showed the greatest gains during this condition. In conclusion, working memory for common verbal items is impaired in children with RD; however, performance can be facilitated, and learning efficiency maximized, when information is presented visually. The results provide further evidence for the pictorial superiority hypothesis and the theory that pictorial presentation of verbal stimuli is adequate for dual coding.     

Rapid learning of syllable classes from a perceptually continuous speech stream

To learn a language, speakers must learn its words and rules from fluent speech; in particular, they must learn dependencies among linguistic classes. We show that when familiarized with a short artificial, subliminally bracketed stream, participants can learn relations about the structure of its words, which specify the classes of syllables occurring in first and last word positions. By studying the effect of familiarization length, we compared the general predictions of associative theories of learning and those of models postulating separate mechanisms for quickly extracting the word structure and for tracking the syllable distribution in the stream. As predicted by the dual-mechanism model, the preference for structurally correct items was negatively correlated with the familiarization length. This result is difficult to explain by purely associative schemes; an extensive set of neural network simulations confirmed this difficulty. Still, we show that powerful statistical computations operating on the stream are available to our participants, as they are sensitive to co-occurrence statistics among non-adjacent syllables. We suggest that different learning mechanisms analyze speech on-line: A rapid mechanism extracting structural information about the stream, and a slower mechanism detecting statistical regularities among the items occurring in it.     

The contribution of phonological short-term memory to artificial grammar learning

Three experiments investigated the contribution of phonological short-term memory (STM) to grammar learning by manipulating rehearsal during study of an auditory artificial grammar made up from a vocabulary of spoken Mandarin syllables. Experiment 1 showed that concurrent, irrelevant articulation impaired grammar learning compared with a nonverbal control task. Experiment 2 replicated and extended this finding, showing that repeating the grammatical strings at study improved grammar learning compared with suppressing rehearsal or remaining silent during learning. Experiment 3 found no effects of rehearsal on grammar learning once participants had learned the component syllables. The findings suggest that phonological STM aids artificial grammar learning via effects on vocabulary learning.     

Domain-specific and domain-general constraints on word and sequence learning

The relative influences of language-related and memory-related constraints on the learning of novel words and sequences were examined by comparing individual differences in performance of children with and without specific deficits in either language or working memory. Children recalled lists of words in a Hebbian learning protocol in which occasional lists repeated, yielding improved recall over the course of the task on the repeated lists. The task involved presentation of pictures of common nouns followed immediately by equivalent presentations of the spoken names. The same participants also completed a paired-associate learning task involving word–picture and nonword–picture pairs. Hebbian learning was observed for all groups. Domain-general working memory constrained immediate recall, whereas language abilities impacted recall in the auditory modality only. In addition, working memory constrained paired-associate learning generally, whereas language abilities disproportionately impacted novel word learning. Overall, all of the learning tasks were highly correlated with domain-general working memory. The learning of nonwords was additionally related to general intelligence, phonological short-term memory, language abilities, and implicit learning. The results suggest that distinct associations between language- and memory-related mechanisms support learning of familiar and unfamiliar phonological forms and sequences.     

Chunking or not chunking? How do we find words in artificial language learning?

What is the nature of the representations acquired in implicit statistical learning? Recent results in the field of language learning have shown that adults and infants are able to find the words of an artificial language when exposed to a continuous auditory sequence consisting in a random ordering of these words. Such performance can only be based on processing the transitional probabilities between sequence elements. Two different kinds of mechanisms may account for these data: Participants may either parse the sequence into smaller chunks corresponding to the words of the artificial language, or they may become progressively sensitive to the actual values of the transitional probabilities between syllables. The two accounts are difficult to differentiate because they make similar predictions in comparable experimental settings. In this study, we present two experiments that aimed at contrasting these two theories. In these experiments, participants had to learn 2 sets of pseudo-linguistic regularities: Language 1 (L1) and Language 2 (L2) presented in the context of a serial reaction time task. L1 and L2 were either unrelated (none of the syllabic transitions of L1 were present in L2), or partly related (some of the intra-words transitions of L1 were used as inter-words transitions of L2). The two accounts make opposite predictions in these two settings. Our results indicate that the nature of the representations depends on the learning condition. When cues were presented to facilitate parsing of the sequence, participants learned the words of the artificial language. However, when no cues were provided, performance was strongly influenced by the employed transitional probabilities.     

Primitive computations in speech processing

Previous research suggests that artificial-language learners exposed to quasi-continuous speech can learn that the first and the last syllables of words have to belong to distinct classes (e.g., Endress & Bonatti, 2007; Peña, Bonatti, Nespor, & Mehler, 2002). The mechanisms of these generalizations, however, are debated. Here we show that participants learn such generalizations only when the crucial syllables are in edge positions (i.e., the first and the last), but not when they are in medial positions (i.e., the second and the fourth in pentasyllabic items). In contrast to the generalizations, participants readily perform statistical analyses also in word middles. In analogy to sequential memory, we suggest that participants extract the generalizations using a simple but specific mechanism that encodes the positions of syllables that occur in edges. Simultaneously, they use another mechanism to track the syllable distribution in the speech streams. In contrast to previous accounts, this model explains why the generalizations are faster than the statistical computations, require additional cues, and break down under different conditions, and why they can be performed at all. We also show that that similar edge-based mechanisms may explain many results in artificial-grammar learning and also various linguistic observations.     

The Hebb repetition effect as a laboratory analogue of novel word learning

The present study tests the hypothesis that a common ordering mechanism underlies both short-term serial recall of verbal materials and the acquisition of novel long-term lexical representations, using the Hebb repetition effect. In the first experiment, participants recalled visually presented nonsense syllables following a typical Hebb effect learning protocol. Replicating the Hebb repetition effect, we observed improved recall for repeated sequences of syllables. In the second experiment, the same participants performed an auditory lexical decision task, which included nonwords that were constructed from the syllables used in the first experiment. We observed inhibited rejection of nonwords that were composed of the repeated Hebb sequences, compared to nonwords that were built from nonrepeated filler sequences. This suggests that a long-term phonological lexical representation developed during Hebb learning. Accordingly, the relation between immediate serial recall and word learning is made explicit by arguing that the Hebb repetition effect is a laboratory analogue of naturalistic vocabulary acquisition.     

How verbal memory loads consume attention

According to a traditional assumption about working memory, participants retain a series of verbal items for immediate recall using covert verbal rehearsal, without much need for attention. We reassessed this assumption by imposing a speeded, nonverbal choice reaction time (CRT) task following the presentation of each digit in a list to be recalled. When the memory load surpassed a few items, performance on the speeded CRT task became increasingly impaired. This CRT task impairment depended only on attention-related components of working memory; it was not alleviated by the presence of an auditory memory trace that automatically helped the recall of items at the ends of spoken lists. We suggest that attention-demanding refreshing of verbal stimuli occurs along with any covert rehearsal.     

The Hebb repetition effect as a laboratory analogue of novel word learning

The present study tests the hypothesis that a common ordering mechanism underlies both short-term serial recall of verbal materials and the acquisition of novel long-term lexical representations, using the Hebb repetition effect. In the first experiment, participants recalled visually presented nonsense syllables following a typical Hebb effect learning protocol. Replicating the Hebb repetition effect, we observed improved recall for repeated sequences of syllables. In the second experiment, the same participants performed an auditory lexical decision task, which included nonwords that were constructed from the syllables used in the first experiment. We observed inhibited rejection of nonwords that were composed of the repeated Hebb sequences, compared to nonwords that were built from nonrepeated filler sequences. This suggests that a long-term phonological lexical representation developed during Hebb learning. Accordingly, the relation between immediate serial recall and word learning is made explicit by arguing that the Hebb repetition effect is a laboratory analogue of naturalistic vocabulary acquisition.     

Peer Relationships and Emotional Well-Being of Children with Sickle Cell Disease: A Controlled Replication

This study investigated the effects of stimulus presentation modality on working memory performance in elementary school-age children ages 7–13. The experimental paradigm implemented a multitrial learning task incorporating three presentation modalities: Auditory, Visual, and simultaneous Auditory plus Visual. The first experiment compared the learning and memory performance of older and younger elementary school children. The second experiment compared verbal learning and memory performance in elementary school children with major depressive disorder (MDD) to the performance of nondepressed children. All participants benefited from the pictorial presentation of information during learning and recall of information as compared to the auditory presentation alone. Both age and socioeconomic status affected working memory performance in typically developing children. Children with depression demonstrated a more passive learning style during the auditory list acquisition. The present study supports the pictorial superiority hypothesis in verbal learning tasks and the theory that working memory matures during elementary school years. Furthermore, current results indicate that complex working memory measures are not entirely independent of previous experience.     

Speech segmentation by statistical learning depends on attention

We addressed the hypothesis that word segmentation based on statistical regularities occurs without the need of attention. Participants were presented with a stream of artificial speech in which the only cue to extract the words was the presence of statistical regularities between syllables. Half of the participants were asked to passively listen to the speech stream, while the other half were asked to perform a concurrent task. In Experiment 1, the concurrent task was performed on a separate auditory stream (noises), in Experiment 2 it was performed on a visual stream (pictures), and in Experiment 3 it was performed on pitch changes in the speech stream itself. Invariably, passive listening to the speech stream led to successful word extraction (as measured by a recognition test presented after the exposure phase), whereas diverted attention led to a dramatic impairment in word segmentation performance. These findings demonstrate that when attentional resources are depleted, word segmentation based on statistical regularities is seriously compromised.     

Recency effects in memory,as a function of modality of intervening events

Summary Word lists of fifteen items were presented to eye or to ear, with recall either immediately, or after a visual task, or after an auditory one. Instructions were to recall the last items first. An intervening task using the same modality greatly reduced recall of the last items presented; whereas a visual task did not do so for acoustically presented items. An auditory task reduced visual memory. These results suggest a specific auditory memory for recent events, over-written by subsequent auditory events.This research was supported by the Medical Research Council. The experimental work was performed at the Applied Psychology Unit, Cambridge.     

The role of overt language production in the Hebb repetition effect

The Hebb repetition effect (Hebb, 1961) occurs when recall performance improves for a list that is repeated during a serial-recall task. This effect is considered a good experimental analogue to language learning. Our objective was to evaluate the role of overt language production in language learning by manipulating recall direction during a Hebb repetition paradigm. In each trial, seven nonsense syllables were presented auditorily. Participants had to orally recall the items either in the presentation order or in reverse order. One sequence was repeated every third trial. In Experiment 1, we compared learning from a group that had recalled the items in their presentation order to learning from a group that had recalled the items in the reverse order. The two groups yielded similar learning rates. In Experiment 2, recall direction was varied between trials. The learning rate was not affected when recall direction varied between trials, suggesting a limited role of overt language production in language learning.     

Activation of midbrain structures by associative novelty and the formation of explicit memory in humans

Recent evidence suggests a close functional relationship between memory formation in the hippocampus and dopaminergic neuromodulation originating in the ventral tegmental area and medial substantia nigra of the midbrain. Here we report midbrain activation in two functional MRI studies of visual memory in healthy young adults. In the first study, participants distinguished between familiar and novel configurations of pairs of items which had been studied together by either learning the location or the identity of the items. In the second study, participants studied words by either rating the words' pleasantness or counting syllables. The ventral tegmental area and medial substantia nigra showed increased activation by associative novelty (first study) and subsequent free recall performance (second study). In both studies, this activation accompanied hippocampal activation, but was unaffected by the study task. Thus midbrain regions seem to participate selectively in hippocampus-dependent processes of associative novelty and explicit memory formation, but appear to be unaffected by other task-relevant aspects.     

Chunking by colors: assessing discrete learning in a continuous serial reaction-time task

Chunk learning (the process by which a sequence is learned and retrieved from memory in smaller, decomposed units of information) has been postulated as the main learning mechanism underlying sequence learning (Perruchet & Pacton, 2006). However, the evidence for chunk formation has been elusive in the continuous serial reaction-time task, whereas other continuous, statistical processes of learning account well for the results observed in this task. This article proposes a new index to capture segmentation in learning, based on the variance of responding to different parts of a sequence. We assess the validity of this measure by comparing performance in a control group with that of another group in which color codes were used to induce a uniform segmentation. Results showed that evidence of chunking was obtained when the color codes were consistently coupled to responses, but that chunking was not maintained after the colors were removed.     

Overt language production plays a key role in the Hebb repetition effect

When asked to recall verbatim a short list of items, performance is very limited. However, if the list of items is repeated across trials, recall performance improves. This phenomenon, known as the Hebb repetition effect (Hebb, 1961; Brain Mechanisms and Learning: A Symposium, pp. 37–51), is considered a laboratory analogue of language learning. In effect, learning a new word implies the maintenance of a series of smaller units, such as phonemes or syllables, in the correct order for a short amount of time before producing them. The sequence of smaller units is typically presented more than once. In the present study, we investigated the role of overt language production in language learning by manipulating recall direction. If the learning of a repeated list of items relies on overt language production processes, changing list production order by manipulating recall direction should impact the learning of the list. In Experiment 1, one list was repeated every third trial, and recall direction of the repeated list changed on the ninth repetition. In Experiment 1a, the repeated list changed from a forward to a backward order recall, where participants had to recall the items in reverse presentation order. In Experiment 1b, the repeated list changed from a backward to a forward order recall. Results showed a cost in recall performance for the repeated list when recall direction switched from forward to backward recall, whereas it was unaffected by the change from backward to forward recall. In Experiment 2, we increased the number of trials before introducing the change from a backward to a forward order recall. Results showed a decrement in recall performance for the repeated list following the change in recall direction, suggesting that language production processes play a role in the Hebb repetition effect.     

On-line statistical segmentation of a non-speech auditory stream in neonates as demonstrated by event-related brain potentials

The ability to statistically segment a continuous auditory stream is one of the most important preparations for initiating language learning. Such ability is available to human infants at 8 months of age, as shown by a behavioral measurement. However, behavioral study alone cannot determine how early this ability is available. A recent study using measurements of event-related potential (ERP) revealed that neonates are able to detect statistical boundaries within auditory streams of speech syllables. Extending this line of research will allow us to better understand the cognitive preparation for language acquisition that is available to neonates. The aim of the present study was to examine the domain-generality of such statistical segmentation. Neonates were presented with nonlinguistic tone sequences composed of four tritone units, each consisting of three semitones extracted from one octave, for two 5-minute sessions. Only the first tone of each unit evoked a significant positivity in the frontal area during the second session, but not in the first session. This result suggests that the general ability to distinguish units in an auditory stream by statistical information is activated at birth and is probably innately prepared in humans.     

Irrelevant speech effects and statistical learning

Immediate serial recall of visually presented verbal stimuli is impaired by the presence of irrelevant auditory background speech, the so-called irrelevant speech effect. Two of the three main accounts of this effect place restrictions on when it will be observed, limiting its occurrence either to items processed by the phonological loop (the phonological loop hypothesis) or to items that are not too dissimilar from the irrelevant speech (the feature model). A third, the object-oriented episodic record (O-OER) model, requires only that the memory task involves seriation. The present studies test these three accounts by examining whether irrelevant auditory speech will interfere with a task that does not involve the phonological loop, does not use stimuli that are compatible with those to be remembered, but does require seriation. Two experiments found that irrelevant speech led to lower levels of performance in a visual statistical learning task, offering more support for the O-OER model and posing a challenge for the other two accounts.