Chunking Versus Transitional Probabilities: Differentiating Between Theories of Statistical Learning.

There are two main approaches to how statistical patterns are extracted from sequences: The transitional probability approach proposes that statistical learning occurs through the computation of probabilities between items in a sequence. The chunking approach, including models such as PARSER and TRACX, proposes that units are extracted as chunks. Importantly, the chunking approach suggests that the extraction of full units weakens the processing of subunits while the transitional probability approach suggests that both units and subunits should strengthen. Previous findings using sequentially organized, auditory stimuli or spatially organized, visual stimuli support the chunking approach. However, one limitation of prior studies is that most assessed learning with the two-alternative forced-choice task. In contrast, this pre-registered experiment examined the two theoretical approaches in sequentially organized, visual stimuli using an online self-paced task-arguably providing a more sensitive index of learning as it occurs-and a secondary offline familiarity judgment task. During the self-paced task, abstract shapes were covertly organized into eight triplets (ABC) where one in every eight was altered (BCA) from the canonical structure in a way that disrupted the full unit while preserving a subunit (BC). Results from the offline familiarity judgment task revealed that the altered triplets were perceived as highly familiar, suggesting the learned representations were relatively flexible. More importantly, results from the online self-paced task demonstrated that processing for subunits, but not unit-initial stimuli, was impeded in the altered triplet. The pattern of results is in line with the chunking approach to statistical learning and, more specifically, the TRACX model.

Nonhuman primates learn adjacent dependencies but fail to learn nonadjacent dependencies in a statistical learning task with a salient cue.

There is ample evidence that humans and nonhuman animals can learn complex statistical regularities presented within various types of input. However, humans outperform their nonhuman primate counterparts when it comes to recognizing relationships that exist across one or several intervening stimuli (nonadjacent dependencies). This is especially true when the two elements in the dependency do not share any perceptual similarity (arbitrary associations). In the present study, we investigated whether manipulating the saliency of the predictive stimulus would enhance nonadjacent dependency learning in nonhuman primates. Rhesus macaques and tufted capuchins engaged in a computerized signal detection task that included sequences that were random in nature, included an adjacent dependency, or included a nonadjacent dependency. We manipulated the saliency of the predictive stimulus, such that the predictor jittered in place on the screen in some grammar blocks, as well as the transitional probability (the likelihood of the stimulus preceding the target to accurately predict the target's appearance) from block to block. Some monkeys evidenced learning of adjacent dependencies by faster response times to targets that followed a predictive stimulus compared to targets that were not preceded by a predictor. However, consistent with the body of evidence that indicates that nonhuman animals' statistical learning mechanisms are not at the same level of sophistication as humans', there was no evidence that monkeys learned nonadjacent dependencies of arbitrary associations, even when the salient cue was present.

Unraveling the Interconnections Between Statistical Learning and Dyslexia: A Review of Recent Empirical Studies.

One important aspect of human cognition involves the learning of structured information encountered in our environment, a phenomenon known as statistical learning. A growing body of research suggests that learning to read print is partially guided by learning the statistical contingencies existing between the letters within a word, and also between the letters and sounds to which the letters refer. Research also suggests that impairments to statistical learning ability may at least partially explain the difficulties experienced by individuals diagnosed with dyslexia. However, the findings regarding impaired learning are not consistent, perhaps partly due to the varied use of methodologies across studies - such as differences in the learning paradigms, stimuli used, and the way that learning is assessed - as well as differences in participant samples such as age and extent of the learning disorder. In this review, we attempt to examine the purported link between statistical learning and dyslexia by assessing a set of the most recent and relevant studies in both adults and children. Based on this review, we conclude that although there is some evidence for a statistical learning impairment in adults with dyslexia, the evidence for an impairment in children is much weaker. We discuss several suggestive trends that emerge from our examination of the research, such as issues related to task heterogeneity, possible age effects, the role of publication bias, and other suggestions for future research such as the use of neural measures and a need to better understand how statistical learning changes across typical development. We conclude that no current theoretical framework of dyslexia fully captures the extant research findings on statistical learning.

Visual Sequence Repetition Learning is Not Impaired in Signing DHH Children.

The auditory scaffolding hypothesis states that early experience with sound underpins the development of domain-general sequence processing abilities, supported by studies observing impaired sequence processing in deaf or hard-of-hearing (DHH) children. To test this hypothesis, we administered a sequence processing task to 77 DHH children who use American Sign Language (ASL) and 23 hearing monolingual children aged 7-12 years and found no performance difference between them after controlling for age and nonverbal intelligence. Additionally, neither spoken language comprehension scores nor hearing loss levels predicted sequence processing scores in the DHH group, whereas ASL comprehension scores did. Our results do not indicate sequence processing deficits in DHH children and do not support the auditory scaffolding hypothesis; instead, these findings suggest that factors related to experience with and/or proficiency in an accessible language during development may be more important determinants of sequence processing abilities.

How statistical learning interacts with the socioeconomic environment to shape children's language development.

Language is acquired in part through statistical learning abilities that encode environmental regularities. Language development is also heavily influenced by social environmental factors such as socioeconomic status. However, it is unknown to what extent statistical learning interacts with SES to affect language outcomes. We measured event-related potentials in 26 children aged 8-12 while they performed a visual statistical learning task. Regression analyses indicated that children's learning performance moderated the relationship between socioeconomic status and both syntactic and vocabulary language comprehension scores. For children demonstrating high learning, socioeconomic status had a weaker effect on language compared to children showing low learning. These results suggest that high statistical learning ability can provide a buffer against the disadvantages associated with being raised in a lower socioeconomic status household.

Semantic P600-but not N400-effects index crosslinguistic variability in speakers' expectancies for expression of motion.

The expression of motion shows strong crosslinguistic variability; however, less is known about speakers' expectancies for lexicalizations of motion at the neural level. We examined event-related brain potentials (ERPs) in native English or Spanish speakers while they read grammatical sentences describing animations involving manner and path components of motion that did or did not violate language-specific patterns of expression. ERPs demonstrated different expectancies between speakers: Spanish speakers showed higher expectancies for motion verbs to encode path and English speakers showed higher expectancies for motion verbs to encode manner followed by a secondary path expression. Interestingly, grammatical but infrequent motion expressions (manner verbs in Spanish, path verbs and secondary manner expressions in English) elicited semantic P600 rather than the expected N400 effects-with or without post-N400 positivities-that are typically associated with semantic processing. Overall, our findings provide the first empirical evidence for the effect of crosslinguistic variation in processing motion event descriptions at the neural level.

Double Dissociation of Auditory Attention Span and Visual Attention in Long-Term Survivors of Childhood Cerebellar Tumor: A Deterministic Tractography Study of the Cerebellar-Frontal and the Superior Longitudinal Fasciculus Pathways.

OBJECTIVE: Right cerebellar-left frontal (RC-LF) white matter integrity (WMI) has been associated with working memory. However, prior studies have employed measures of working memory that include processing speed and attention. We examined the relationships between the RC-LF WMI and processing speed, attention, and working memory to clarify the relationship of RC-LF WMI with a specific cognitive function. Right superior longitudinal fasciculus II (SLF II) WMI and visual attention were included as a negative control tract and task to demonstrate a double dissociation. METHODS: Adult survivors of childhood brain tumors [n = 29, age: M = 22 years (SD = 5), 45% female] and demographically matched controls were recruited (n = 29). Tests of auditory attention span, working memory, and visual attention served as cognitive measures. Participants completed a 3-T MRI diffusion-weighted imaging scan. Fractional anisotropy (FA) and radial diffusivity (RD) served as WMI measures. Partial correlations between WMI and cognitive scores included controlling for type of treatment. RESULTS: A correlational double dissociation was found. RC-LF WMI was associated with auditory attention (FA: r = .42, p = .03; RD: r = -.50, p = .01) and was not associated with visual attention (FA: r = -.11, p = .59; RD: r = -.11, p = .57). SLF II FA WMI was associated with visual attention (FA: r = .44, p = .02; RD: r = -.17, p = .40) and was not associated with auditory attention (FA: r = .24, p = .22; RD: r = -.10, p = .62). CONCLUSIONS: The results show that RC-LF WMI is associated with auditory attention span rather than working memory per se and provides evidence for a specificity based on the correlational double dissociation.

Distinct neural networks for detecting violations of adjacent versus nonadjacent sequential dependencies: An fMRI study.

The ability to learn and process sequential dependencies is essential for language acquisition and other cognitive domains. Recent studies suggest that the learning of adjacent (e.g., "A-B") versus nonadjacent (e.g., "A-X-B") dependencies have different cognitive demands, but the neural correlates accompanying such processing are currently underspecified. We developed a sequential learning task in which sequences of printed nonsense syllables containing both adjacent and nonadjacent dependencies were presented. After incidentally learning these grammatical sequences, twenty-one healthy adults (age M = 22.1, 12 females) made familiarity judgments about novel grammatical sequences and ungrammatical sequences containing violations of the adjacent or nonadjacent structure while in a 3T MRI scanner. Violations of adjacent dependencies were associated with increased BOLD activation in both posterior (lateral occipital and angular gyrus) as well as frontal regions (e.g., medial frontal gyrus, inferior frontal gyrus). Initial results indicated no regions showing significant BOLD activations for the violations of nonadjacent dependencies. However, when using a less stringent cluster threshold, exploratory analyses revealed that violations of nonadjacent dependencies were associated with increased activation in subcallosal cortex, paracingulate cortex, and anterior cingulate cortex (ACC). Finally, when directly comparing the adjacent condition to the nonadjacent condition, we found significantly greater levels of activation for the right superior lateral occipital cortex (BA 19) for the adjacent relative to nonadjacent condition. In sum, the detection of violations of adjacent and nonadjacent dependencies appear to involve distinct neural networks, with perceptual brain regions mediating the processing of adjacent but not nonadjacent dependencies. These results are consistent with recent proposals that statistical-sequential learning is not a unified construct but depends on the interaction of multiple neurocognitive mechanisms acting together.

How does the brain learn environmental structure? Ten core principles for understanding the neurocognitive mechanisms of statistical learning.

Despite a growing body of research devoted to the study of how humans encode environmental patterns, there is still no clear consensus about the nature of the neurocognitive mechanisms underpinning statistical learning nor what factors constrain or promote its emergence across individuals, species, and learning situations. Based on a review of research examining the roles of input modality and domain, input structure and complexity, attention, neuroanatomical bases, ontogeny, and phylogeny, ten core principles are proposed. Specifically, there exist two sets of neurocognitive mechanisms underlying statistical learning. First, a "suite" of associative-based, automatic, modality-specific learning mechanisms are mediated by the general principle of cortical plasticity, which results in improved processing and perceptual facilitation of encountered stimuli. Second, an attention-dependent system, mediated by the prefrontal cortex and related attentional and working memory networks, can modulate or gate learning and is necessary in order to learn nonadjacent dependencies and to integrate global patterns across time. This theoretical framework helps clarify conflicting research findings and provides the basis for future empirical and theoretical endeavors.

Concurrent Learning of Adjacent and Nonadjacent Dependencies in Visuo-Spatial and Visuo-Verbal Sequences.

Both adjacent and non-adjacent dependencies (AD and NAD) are present in natural language and other domains, yet the learning of non-adjacent sequential dependencies generally only occurs under favorable circumstances. It is currently unknown to what extent adults can learn AD and NAD, presented concurrently in spatial and verbal sequences during a single session, and whether a second session improves performance. In addition, the relationship between AD and NAD learning and other theoretically related cognitive and language processes has not yet been fully established. In this study, participants reproduced two types of sequences generated from an artificial grammar: visuo-spatial sequences with stimuli presented in four spatial locations, and visuo-verbal sequences with printed syllables. Participants were tested for incidental learning by reproducing novel sequences, half consistent with the grammar and half containing violations of either AD or NAD. The procedure was repeated on a second day. Results showed that both AD and NAD were learned in both visuo-spatial and visuo-verbal tasks, although AD learning was better than NAD and learning of NAD decreased over time. Furthermore, NAD learning for both spatial and verbal tasks was positively correlated with a language measure, whereas AD learning for both spatial and verbal tasks was negatively associated with working memory measures in the opposite domain. These results demonstrate that adults can learn both AD and NAD within a single session, but NAD learning is more easily disrupted than AD and both types of learning are sub-served by partially distinct cognitive processes. These findings increase our understanding of the processes governing the learning of AD and NAD in verbal and spatial domains.

Visual sequential processing and language ability in children who are deaf or hard of hearing.

This study investigated the role of sequential processing in spoken language outcomes for children who are deaf or hard of hearing (DHH), ages 5;3-11;4, by comparing them to children with typical hearing (TH), ages 6;3-9;7, on sequential learning and memory tasks involving easily nameable and difficult-to-name visual stimuli. Children who are DHH performed more poorly on easily nameable sequencing tasks, which positively predicted receptive vocabulary scores. Results suggest sequential learning and memory may underlie delayed language skills of many children who are DHH. Implications for language development in children who are DHH are discussed.

The Promise-and Challenge-of Statistical Learning for Elucidating Atypical Language Development.

Statistical learning plays an important role in the acquisition of spoken and written language. It has been proposed that impaired or atypical statistical learning may be linked with language difficulties in developmental disabilities. However, research on statistical learning in individuals with developmental disabilities such as autism spectrum disorder, dyslexia, and specific language impairment, and in individuals with cochlear implants, has produced divergent findings. It is unclear whether, and to what extent, statistical learning is impaired or atypical in each of these developmental disabilities. We suggest that these disparate findings point to several critical issues that must be addressed before we can evaluate the role of statistical learning in atypical child development. While the issues we outline are interrelated, we propose four key points relating to (a) the nature of statistical learning, (b) the myriad of ways in which statistical learning can be measured, (c) our lack of understanding regarding the developmental trajectory of statistical learning, and (d) the role of individual differences. We close by making suggestions that we believe will be helpful in moving the field forward and creating new synergies among researchers, clinicians, and educators to better support language learners.

Under What Conditions Can Recursion Be Learned? Effects of Starting Small in Artificial Grammar Learning of Center-Embedded Structure.

It has been suggested that external and/or internal limitations paradoxically may lead to superior learning, that is, the concepts of starting small and less is more (Elman, ; Newport, ). In this paper, we explore the type of incremental ordering during training that might help learning, and what mechanism explains this facilitation. We report four artificial grammar learning experiments with human participants. In Experiments 1a and 1b we found a beneficial effect of starting small using two types of simple recursive grammars: right-branching and center-embedding, with recursive embedded clauses in fixed positions and fixed length. This effect was replicated in Experiment 2 (N = 100). In Experiment 3 and 4, we used a more complex center-embedded grammar with recursive loops in variable positions, producing strings of variable length. When participants were presented an incremental ordering of training stimuli, as in natural language, they were better able to generalize their knowledge of simple units to more complex units when the training input "grew" according to structural complexity, compared to when it "grew" according to string length. Overall, the results suggest that starting small confers an advantage for learning complex center-embedded structures when the input is organized according to structural complexity.

The Role of Statistical Learning in Understanding and Treating Spoken Language Outcomes in Deaf Children With Cochlear Implants.

Purpose: Statistical learning-the ability to learn patterns in environmental input-is increasingly recognized as a foundational mechanism necessary for the successful acquisition of spoken language. Spoken language is a complex, serially presented signal that contains embedded statistical relations among linguistic units, such as phonemes, morphemes, and words, which represent the phonotactic and syntactic rules of language. In this review article, we first review recent work that demonstrates that, in typical language development, individuals who display better nonlinguistic statistical learning abilities also show better performance on different measures of language. We next review research findings that suggest that children who are deaf and use cochlear implants may have difficulties learning sequential input patterns, possibly due to auditory and/or linguistic deprivation early in development, and that the children who show better sequence learning abilities also display improved spoken language outcomes. Finally, we present recent findings suggesting that it may be possible to improve core statistical learning abilities with specialized training and interventions and that such improvements can potentially impact and facilitate the acquisition and processing of spoken language. Method: We conducted a literature search through various online databases including PsychINFO and PubMed, as well as including relevant review articles gleaned from the reference sections of other review articles used in this review. Search terms included various combinations of the following: sequential learning, sequence learning, statistical learning, sequence processing, procedural learning, procedural memory, implicit learning, language, computerized training, working memory training, statistical learning training, deaf, deafness, hearing impairment, hearing impaired, DHH, hard of hearing, cochlear implant(s), hearing aid(s), and auditory deprivation. To keep this review concise and clear, we limited inclusion to the foundational and most recent (2005-2018) relevant studies that explicitly included research or theoretical perspectives on statistical or sequential learning. We here summarize and synthesize the most recent and relevant literature to understanding and treating language delays in children using cochlear implants through the lens of statistical learning. Conclusions: We suggest that understanding how statistical learning contributes to spoken language development is important for understanding some of the difficulties that children who are deaf and use cochlear implants might face and argue that it may be beneficial to develop novel language interventions that focus specifically on improving core foundational statistical learning skills.

Atypical predictive processing during visual statistical learning in children with developmental dyslexia: an event-related potential study.

Previous research suggests that individuals with developmental dyslexia perform below typical readers on non-linguistic cognitive tasks involving the learning and encoding of statistical-sequential patterns. However, the neural mechanisms underlying such a deficit have not been well examined. The aim of the present study was to investigate the event-related potential (ERP) correlates of sequence processing in a sample of children diagnosed with dyslexia using a non-linguistic visual statistical learning paradigm. Whereas the response time data suggested that both typical and atypical readers learned the statistical patterns embedded in the task, the ERP data suggested otherwise. Specifically, ERPs of the typically developing children (n = 12) showed a P300-like response indicative of learning, whereas the children diagnosed with a reading disorder (n = 8) showed no such ERP effects. These results may be due to intact implicit motor learning in the children with dyslexia but delayed attention-dependent predictive processing. These findings are consistent with other evidence suggesting that differences in statistical learning ability might underlie some of the reading deficits observed in developmental dyslexia.

Lack of Cross-Modal Effects in Dual-Modality Implicit Statistical Learning.

A current controversy in the area of implicit statistical learning (ISL) is whether this process consists of a single, central mechanism or multiple modality-specific ones. To provide insight into this question, the current study involved three ISL experiments to explore whether multimodal input sources are processed separately in each modality or are integrated together across modalities. In Experiment 1, visual and auditory ISL were measured under unimodal conditions, with the results providing a baseline level of learning for subsequent experiments. Visual and auditory sequences were presented separately, and the underlying grammar used for both modalities was the same. In Experiment 2, visual and auditory sequences were presented simultaneously with each modality using the same artificial grammar to investigate whether redundant multisensory information would result in a facilitative effect (i.e., increased learning) compared to the baseline. In Experiment 3, visual and auditory sequences were again presented simultaneously but this time with each modality employing different artificial grammars to investigate whether an interference effect (i.e., decreased learning) would be observed compared to the baseline. Results showed that there was neither a facilitative learning effect in Experiment 2 nor an interference effect in Experiment 3. These findings suggest that participants were able to track simultaneously and independently two sets of sequential regularities under dual-modality conditions. These findings are consistent with the theories that posit the existence of multiple, modality-specific ISL mechanisms rather than a single central one.

Visual artificial grammar learning by rhesus macaques (Macaca mulatta): exploring the role of grammar complexity and sequence length.

Humans and nonhuman primates can learn about the organization of stimuli in the environment using implicit sequential pattern learning capabilities. However, most previous artificial grammar learning studies with nonhuman primates have involved relatively simple grammars and short input sequences. The goal in the current experiments was to assess the learning capabilities of monkeys on an artificial grammar-learning task that was more complex than most others previously used with nonhumans. Three experiments were conducted using a joystick-based, symmetrical-response serial reaction time task in which two monkeys were exposed to grammar-generated sequences at sequence lengths of four in Experiment 1, six in Experiment 2, and eight in Experiment 3. Over time, the monkeys came to respond faster to the sequences generated from the artificial grammar compared to random versions. In a subsequent generalization phase, subjects generalized their knowledge to novel sequences, responding significantly faster to novel instances of sequences produced using the familiar grammar compared to those constructed using an unfamiliar grammar. These results reveal that rhesus monkeys can learn and generalize the statistical structure inherent in an artificial grammar that is as complex as some used with humans, for sequences up to eight items long. These findings are discussed in relation to whether or not rhesus macaques and other primate species possess implicit sequence learning abilities that are similar to those that humans draw upon to learn natural language grammar.

Visual statistical learning is related to natural language ability in adults: An ERP study.

Statistical learning (SL) is believed to enable language acquisition by allowing individuals to learn regularities within linguistic input. However, neural evidence supporting a direct relationship between SL and language ability is scarce. We investigated whether there are associations between event-related potential (ERP) correlates of SL and language abilities while controlling for the general level of selective attention. Seventeen adults completed tests of visual SL, receptive vocabulary, grammatical ability, and sentence completion. Response times and ERPs showed that SL is related to receptive vocabulary and grammatical ability. ERPs indicated that the relationship between SL and grammatical ability was independent of attention while the association between SL and receptive vocabulary depended on attention. The implications of these dissociative relationships in terms of underlying mechanisms of SL and language are discussed. These results further elucidate the cognitive nature of the links between SL mechanisms and language abilities.