Statistical Regularities Research Articles

Sensitivity to Graphotactic Regularities in Elementary School: Development and Contributing Variables.

Graphotactic regularities are statistical regularities governing orthographic systems that children are sensitive to from the start of their literacy learning. The current study observed changes in children's sensitivity to a set of graphotactic patterns across different grades in elementary school and measured the contribution of skills such as expressive spelling, reading fluency, nonverbal reasoning, and receptive vocabulary to children's sensitivity of these graphotactic regularities. One thousand one hundred one French-speaking children in Grades 1-5 completed a writing under a dictation task, a text reading fluency task, and a pseudo-orthographic choice task involving different graphotactic regularities. These regularities fell into two categories: legal versus illegal, which defines the legality of letter strings in French, and frequent versus less frequent, which refers to acceptable letter strings that vary in frequency of occurrence either at the beginning or end of a word. The results of a repeated-measures analysis of variance showed a developmental difference between graphotactic regularity categories. The frequent versus infrequent patterns developed faster than the legal versus illegal patterns until reaching a point of equivalence in Grade 3. At Grades 4 and 5, legal versus illegal graphotactic regularities progressed more quickly while frequent versus less frequent regularities progressed more slowly. Furthermore, generalized linear mixed-model analyses for both types of graphotactic regularities revealed that they were affected by grade, expressive spelling, reading fluency level, and nonverbal reasoning. This study provides evidence of developmental differences in sensitivity to graphotactic regularities according to the type of regularity studied. Reading fluency and expressive spelling skills contribute to graphotactic regularity sensitivity. Higher scores in expressive spelling, reading fluency, nonverbal reasoning, and an older age were related to higher skills in identifying legal versus illegal graphotactic regularities. Such findings can contribute to educational, clinical, and research applications.

Pupil responds spontaneously to visuospatial regularity.

Beyond the light reflex, the pupil responds to various high-level cognitive processes. Multiple statistical regularities of stimuli have been found to modulate the pupillary response. However, most studies have used auditory or visual temporal sequences as stimuli, and it is unknown whether the pupil size is modulated by statistical regularity in the spatial arrangement of stimuli. In three experiments, we created perceived regular and irregular stimuli, matching physical regularity, to investigate the effect of spatial regularity on pupillary responses during passive viewing. Experiments using orientation (Experiments 1 and 2) and size (Experiment 3) as stimuli consistently showed that perceived irregular stimuli elicited more pupil constriction than regular stimuli. Furthermore, this effect was independent of the luminance of the stimuli. In conclusion, our study revealed that the pupil responds spontaneously to perceived visuospatial regularity, extending the stimulus regularity that influences the pupillary response into the visuospatial domain.

Statistical learning of artificial orthographic regularity arises from coordinated activity across distinct brain regions.

The human brain possesses the ability to automatically extract statistical regularities from environmental inputs, including visual-graphic symbols and printed units. However, the specific brain regions underlying the statistical learning of these visual-graphic symbols or artificial orthography remain unclear. This study utilized functional magnetic resonance imaging (fMRI) with an artificial orthography learning paradigm to measure brain activities associated with the statistical learning of radical positional regularities embedded in pseudocharacters containing high (100%), moderate (80%), and low (60%) levels of consistency, along with a series of random abstract figures. Thirty adults passively viewed a continuous stream of these pseudocharacters. fMRI data revealed that the left occipital area and the visual word form area (VWFA) exhibited greater responses at the low consistency level than at the high and moderate levels, suggesting implicit statistical learning of positional regularities. Functional connectivity analysis further revealed significant correlations between the occipital lobe, the VWFA, and other brain regions, such as the middle temporal gyrus (MTG), the superior occipital gyrus (SOG), and the cerebellum. Moreover, neural activity showed a tendency to correlate with behavioural recognition performance. These findings demonstrate that the incidental acquisition of statistical regularities in artificial orthography arises from the coordinated activation of multiple distinct neural circuits.

Distributed Representations for Cognitive Control in Frontal Medial Cortex.

In natural and artificial neural networks, modularity and distributed structure afford complementary but competing benefits. The former allows for hierarchical representations that can flexibly recombine modules to address novel problems, whereas the latter can benefit from less constrained training, potentially uncovering fruitful statistical regularities. Here, we investigate these competing demands in the context of human sequential behavior. First, we explore this setting by comparing the properties of several recurrent neural network models. We find that explicit hierarchical structure by itself fails to provide a critical performance advantage when compared with a "flat" model that does not incorporate hierarchical structure. However, hierarchy appears to facilitate cognitive control processes that support nonroutine behaviors and behaviors that are carried out under computational stress. Second, we compare these models against fMRI data using representational similarity analysis. We find that a model that incorporates so-called wiring costs in the cost function, which produces a hierarchically organized gradient of representational structure across the hidden layer of the neural network, best accounts for fMRI data collected from human participants in a previous study [Holroyd, C. B., Ribas-Fernandes, J. J. F., Shahnazian, D., Silvetti, M., & Verguts, T., Human midcingulate cortex encodes distributed representations of task progress. Proceedings of the National Academy of Sciences, U.S.A., 115, 6398-6403, 2018]. The results reveal that the ACC encodes distributed representations of sequential task context along a rostro-caudal gradient of abstraction: Rostral ACC encodes relatively abstract and temporally extended patterns of activity compared with those encoded by caudal ACC. These results provide insight into the role of ACC in motivation and cognitive control.

Synthetic graphs for link prediction benchmarking

Abstract Predicting missing links in complex networks requires algorithms that are able to explore statistical regularities in the existing data. Here we investigate the interplay between algorithm efficiency and network structures through the introduction of suitably-designed synthetic graphs. We propose a family of random graphs that incorporates both micro-scale motifs and meso-scale communities, two ubiquitous
structures in complex networks. A key contribution is the derivation of theoretical upper bounds for link prediction performance in our synthetic graphs, allowing us to estimate the predictability of the task and obtain an improved assessment of the performance of any method. Our results on the performance of classical methods (e.g., Stochastic Block Models, Node2Vec, GraphSage) show that the performance of all methods correlate with the theoretical predictability, that no single method is universally superior, and that each of the methods exploit different characteristics known to exist in large classes of networks. Our findings underline the need for careful consideration of graph structure when selecting a link prediction method and emphasize the value of comparing performance against synthetic benchmarks. We provide open-source code for generating these synthetic graphs, enabling further research on link
prediction methods.

How long-term memory facilitates working memory: Evidence from flexible responses and neural oscillations.

Historically, working memory (WM) and long-term memory (LTM) were viewed as distinct systems, operating independently. Recent research, however, has uncovered intricate interactions between these memory systems, revealing that LTM information can enhance the WM performance. This study investigates the mechanisms underlying such facilitation through a delayed color-recall task, adapted from Brady et al. (2009). Across three experiments, we systematically manipulated stimulus pairings with temporarily invalidating (Experiment 2) or changing (Experiment 3) statistical pairing regularities. The results demonstrated subjects' flexible utilization of recent input regularities, showcasing their control over this information. Furthermore, regularity learning selectively modulated neural oscillations during the encoding phase, indicating reduced information storage and increased mental resource deployment when leveraging regularities to enhance the WM performances. In sum, this study shed new light on how and when LTM facilitates WM performances through rapid learning and flexible implementation of regularities between stimuli. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Exploration of auditory statistical learning, socioeconomic status, and language outcomes in Bangladeshi children: A functional near-infrared spectroscopy study.

Auditory statistical learning, or the ability to detect statistical regularities in continuously presented stimuli, is thought to be one element that underlies language acquisition. Prior studies have uncovered behavioral and neural correlates of statistical learning, yet additional work is needed from low- and middle-income countries to explore whether statistical learning varies across cultures or underlies associations often found between socioeconomic status (SES) and language outcomes. In the present study, we explored the feasibility of using functional near-infrared spectroscopy (fNIRS) to explore auditory statistical learning in Bangladesh, a lower-middle-income country. Participants were 102 2-year-old (M = 25.72 months, SD = 2.07 months) and 125 5-year-old children (M = 62.35 months, SD = 2.46 months) living in a low-income urban neighborhood of Dhaka (average family income of 28,145.13 Bangladeshi Takas or 260.06 U.S. dollars per month). We also collected measures of SES and language outcomes. Brain responses during the statistical learning paradigm could be detected with fNIRS in both two- and 5-year-olds, with 2-year-olds exhibiting a higher response to predictable sequences and 5-year-olds exhibiting higher responses to unpredictable sequences. fNIRS correlates of statistical learning were not related to language outcomes but were associated with SES in the 5-year-old cohort. This study demonstrates the utility of employing fNIRS to study the neural correlates of statistical learning in low- and middle-income countries and the feasibility of expanding the representativeness of the existing literature. These findings also highlight potential areas for inquiry into how SES may relate to individual differences in statistical learning responses. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Interbrain neural correlates of self and other integration in joint statistical learning

While statistical learning is often studied individually, its collective representation through self-other integration remains unclear. This study examines dynamic self-other integration and its multi-brain mechanism using simultaneous recordings from dyads. Participants (N = 112) each repeatedly responded to half of a fixed stimulus sequence with either an active partner (joint context) or a passive observer (baseline context). Significant individual statistical learning was evident in the joint context, characterized by decreased reaction time (RT) and intra-brain neural responses, followed by a quadratic trend (i.e., first increasing and then decreasing) upon insertion of an interference sequence. More importantly, Brain-to-Brain Coupling (BtBC) in the theta band also showed learning and modulation-related trends, with its slope negatively and positively correlating with the slopes of RT and intra-brain functional connectivity, respectively. These results highlight the dynamic nature of self-other integration in joint statistical learning, with statistical regularities implicitly and spontaneously modulating this process. Notably, the BtBC serves as a key neural correlate underlying the dynamics of self-other integration.

Mechanical properties variation of samples fabricated by fused deposition additive manufacturing as a function of filler percentage and structure for different plastics

One of the key factors in manufacturing products by fused deposition molding (FDM) or layer-by-layer printing technology is the material intensity of the product. The task of reducing the amount of material required to manufacture a product without significant loss of mechanical properties is one of the most practically important technological tasks. Material saving in FDM printing of products allows to reduce financial costs and increase the speed of manufacturing of the final product without reducing (or not significantly reducing) the quality properties of the product. In our work it is demonstrated that using Combs filling type and materials of poly lactic acid (PLA) and polyethylene terephthalate glycol (PETG) it is possible to achieve material savings of up to 23% at 50% filling for PLA and 17% at 75% filling for PETG without significant reduction of product strength in comparison with other filling types. Exceptions are PLA samples with 100% fill and Lateral fill. Application of Kruskal-Wallis criterion and Dunn’s criterion with Bonferroni multiple comparison correction showed that there were no statistically significant differences within the strength limits of samples made by FDM printing technology from PLA and PETG plastics (p-value = 0.0514), as well as samples with Triangle and Grid filling type (p-value = 1). Based on this result, three groups of samples statistically significantly differing in ultimate strength were identified by methods of hierarchical cluster analysis; in each group (except for group 1, which included samples made of PLA plastic with Lateral filling type and 100% filling), correlation analysis was performed (Spearman correlation was used). The results of the correlation analysis showed a stable average correlation between the percentage of filling, modulus along the secant 0.05–0.2% strain, ultimate strength and strain corresponding to the yield stress. Analysis of the correlation graph showed that the main parameter correlating with all mechanical properties of the specimen is the 0.05–0.2% strain modulus. Based on this conclusion, robust regression equations predicting the 0.05–0.2% strain modulus as a function of the percentage of specimen filling were constructed for the two selected groups. Analysis of the equations showed that in the third group of specimens, the average modulus of 0.05–0.2% strain is more than twice the modulus of 0.05–0.2% strain in the second group. The detected statistical regularities can be explained by the mechanism of strain hardening, the actual value of which depends on the structure of the macrodefect (type of filling), properties and volume of the material (percentage of filling) used in the fabrication of samples using FDM printing technology.

The algorithmic nature of song-sequencing: statistical regularities in music albums

Based on a review of anecdotal beliefs, we explored statistical patterns of track-sequencing within a large set of released music albums. We found that songs with high levels of valence, energy and loudness are more likely to be positioned at the beginning of each album. We also found that transitions between consecutive tracks tend to alternate between increases and decreases of valence and energy. These findings were used to build a system which automates the process of album-sequencing. Our results and hypothesis have both practical and theoretical applications. Practically, sequencing regularities can be used to inform playlist generation systems. Theoretically, we show that professional musicians and music producers have significant levels of agreement about how to determine the order of tracks in their albums.

Predictive learning shapes the representational geometry of the human brain

Predictive coding theories propose that the brain constantly updates internal models to minimize prediction errors and optimize sensory processing. However, the neural mechanisms that link prediction error encoding and optimization of sensory representations remain unclear. Here, we provide evidence how predictive learning shapes the representational geometry of the human brain. We recorded magnetoencephalography (MEG) in humans listening to acoustic sequences with different levels of regularity. We found that the brain aligns its representational geometry to match the statistical structure of the sensory inputs, by clustering temporally contiguous and predictable stimuli. Crucially, the magnitude of this representational shift correlates with the synergistic encoding of prediction errors in a network of high-level and sensory areas. Our findings suggest that, in response to the statistical regularities of the environment, large-scale neural interactions engaged in predictive processing modulate the representational content of sensory areas to enhance sensory processing.

Can Language Models Trained on Written Monologue Learn to Predict Spoken Dialogue?

Transformer-based Large Language Models (LLMs) have recently increased in popularity, in part due to their impressive performance on a number of language tasks. While LLMs can produce human-like writing, the extent to which these models can learn to predict spoken language in natural interaction remains unclear. This is a nontrivial question, as spoken and written language differ in syntax, pragmatics, and norms that interlocutors follow. Previous work suggests that while LLMs may develop an understanding of linguistic rules based on statistical regularities, they fail to acquire the knowledge required for language use. This implies that LLMs may not learn the normative structure underlying interactive spoken language, but may instead only model superficial regularities in speech. In this paper, we aim to evaluate LLMs as models of spoken dialogue. Specifically, we investigate whether LLMs can learn that the identity of a speaker in spoken dialogue influences what is likely to be said. To answer this question, we first fine-tuned two variants of a specific LLM (GPT-2) on transcripts of natural spoken dialogue in English. Then, we used these models to compute surprisal values for two-turn sequences with the same first-turn but different second-turn speakers and compared the output to human behavioral data. While the predictability of words in all fine-tuned models was influenced by speaker identity information, the models did not replicate humans' use of this information. Our findings suggest that although LLMs may learn to generate text conforming to normative linguistic structure, they do not (yet) faithfully replicate human behavior in naturalconversation.

The one exception: The impact of statistical regularities on explicit sense of agency.

Establishing causal beliefs by observing regularities between actions and events in the environment is a crucial part of goal-directed behavior. Sense of agency (SoA) describes the corresponding experience of generating and controlling actions and subsequent events. Investigating how SoA adapts to situational changes in action-effect contingency, we observed even singular disturbances of perfect action-effect contingencies to yield a striking impact on SoA formation. Moreover, we additionally included disturbances of regularity that are not directly linked to one's own actions. Doing so allowed us to investigate how SoA might be a concept that goes beyond own actions toward a more generalized, subjective representation of control regarding environmental events. Indeed, the present experiments establish that, while SoA is highly tuned toward action-effect relations, it is also sensitive to events that occur without one's own action contribution. SoA thus appears to be exceptionally sensitive to singular breakpoints of perfect control with agents disproportionally incorporating such events during SoA formation while at the same time building on a rich situation model. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Measuring Multi-Dimensional Mobile Behavior Effect on Inclusive Finance: Evidence from China

Credit Invisible is one key area that many countries put much effort to solve in decades. According to the 2020 World Bank statistics, for example, there are over 500 million Chinese and 45 million American, classified as credit invisible who don’t have banking and finance history in bank or credit bureau, making them difficult to borrow money from financial institution. Previous studies adopted different non-financial information to evaluate one’s credit worthiness and status to address this issue. However, they provide little information about how real mobile user interactions can be used to solve this issue in inclusive finance. This paper proposes a novel data generative framework to fusion APP data, call detail record data and SMS data with a total of 4,689 attributes derived from a large-scale mobile dataset. We then construct a unique set of mobile behavior-driven credit risk factors based on statistical diversity, intensity, consistency, and regularity of mobile user behavior characterizing user preferences, attitudes, geolocation, and temporal patterns. Empirical analysis demonstrates that the newly discovered mobile behavior factors are useful as new inputs for credit scoring and proves the factors representing new source of positive and negative credit information. Decision tree analysis and Quantile regression are conducted to validate effect of these factors to credit default. It facilitates credit assessment based on non-financial data for the credit invisible people, which promoting inclusive finance to larger community in society. We also analyze implications of mobile user characterization findings in relation to credit default which helps decision makers to optimize credit policy and product design.

The Domain-Specific Neural Basis of Auditory Statistical Learning in 5-7-Year-Old Children.

Statistical learning (SL) is the ability to rapidly track statistical regularities and learn patterns in the environment. Recent studies show that SL is constrained by domain-specific features, rather than being a uniform learning mechanism across domains and modalities. This domain-specificity has been reflected at the neural level, as SL occurs in regions primarily involved in processing of specific modalities or domains of input. However, our understanding of how SL is constrained by domain-specific features in the developing brain is severely lacking. The present study aims to identify the functional neural profiles of auditory SL of linguistic and nonlinguistic regularities among children. Thirty children between 5 and 7 years old completed an auditory fMRI SL task containing interwoven sequences of structured and random syllable/tone sequences. Using traditional group univariate analyses and a group-constrained subject-specific analysis, frontal and temporal cortices showed significant activation when processing structured versus random sequences across both linguistic and nonlinguistic domains. However, conjunction analyses failed to identify overlapping neural indices across domains. These findings are the first to compare brain regions supporting SL of linguistic and nonlinguistic regularities in the developing brain and indicate that auditory SL among developing children may be constrained by domain-specific features.

Are Late Talkers Just Late? Neighborhood Density and Word Frequency Properties of Late Talkers' Spoken Vocabularies.

Typically developing toddlers extract patterns from their input to add words to their spoken lexicons, yet some evidence suggests that late talkers leverage the statistical regularities of the ambient language differently than do peers. Using the extended statistical learning account, we sought to compare lexical-level statistical features of spoken vocabularies between late talkers and two typically developing comparison groups. MacArthur-Bates Communicative Developmental Inventories American English Words and Sentences (N = 1,636) were extracted from Wordbank, a database of CDIs. Inventories were divided into three groups: (a) a late talker group (n = 202); (b) a typically developing age-matched group (n = 1,238); and (c) a younger, typically developing group (n = 196) matched to the late talkers on expressive language. Neighborhood density and word frequency were calculated for each word produced by each participant and standardized to z scores. Mixed-effects models were used to evaluate group differences. The late talker and younger, language-matched groups' spoken vocabularies consist, on standardized average, of words from denser phonological neighborhoods and words higher in frequency of occurrence in parent-child speech, compared to older, typically developing toddlers. These findings provide support for the extended statistical learning account. Late talkers appear to generally be extracting and using similar patterns from their language input as do younger toddlers with similar levels of expressive vocabulary. This suggests that late talkers may be following a delayed, not deviant, trajectory of expressive language growth.

Wellbeing public policy needs more theory

Advocacy for ‘wellbeing public policy’ (WPP) requires suitable evidence. Endorsing the ambition to focus policy on wellbeing outcomes, we nevertheless argue that the current evidence base is deficient due to a lack of theory. For the purposes of our analysis, we identify theory narrowly with conceptual clarity and themodelling of causal mechanisms underlying statistical regularities. The prevailing focus on identifying ‘drivers’ of wellbeing and their effect sizes is not well suited for such theorising. We show that this status quo creates potential for conceptual confusion, incorrect aggregation, poor robustness and external validity of policy evaluations, inept interventions, and raises the prospect of a ‘Lucas critique’ in wellbeing economics. We discuss what sort of theory addresses these pitfalls, and where WPP could proceed fruitfully even in the absence of such theory. Ultimately, we call upon wellbeing experts to invest in developing theory as this would improve the basis for WPP and outcomes for those affected by it. Moreover, such theoretical contributions from the field of WPP could spillover to other disciplines, extending the reach and influence of wellbeing research.

Chunking in the Second Language: Implications for Language Learning and Teaching

Among the various challenges that adult and other late language learners face on their journey to achieving nativelike proficiency, chunking has been identified as one of the most difficult tasks to master. Language users are able to derive and utilize chunks during language processing—both in the first (L1) and the second language (L2)—yet the extent to which the L2 learners utilize and benefit from chunking is not on a par with L1 speakers. L2 learners are generally less sensitive to the statistical regularities in the linguistic input and possess a smaller repertoire of multiword expressions, leaving them susceptible to slower real-time language processing, hampered comprehension during conversation, and distinct production errors. Drawing on insights from Brian MacWhinney's Unified Competition Model of L1 and L2 acquisition, this review examines these unique challenges in L2 chunking as a function of differences between L1 and L2 learning. According to the Unified Competition Model, the existence of deeply entrenched L1 linguistic representations may hinder effective L2 chunking by encouraging over-segmentation in favor of L1 transference at the lexical level and diverting the necessary attention away from the grammatical elements in prefabricated multiword units. Based on these observations, we offer practical suggestions for educators to facilitate chunking in L2 learners and bring them closer to nativelike fluency.

Spatial Predictive Context Speeds Up Visual Search by Biasing Local Attentional Competition.

The human visual system is equipped to rapidly and implicitly learn and exploit the statistical regularities in our environment. Within visual search, contextual cueing demonstrates how implicit knowledge of scenes can improve search performance. This is commonly interpreted as spatial context in the scenes becoming predictive of the target location, which leads to a more efficient guidance of attention during search. However, what drives this enhanced guidance is unknown. First, it is under debate whether the entire scene (global context) or more local context drives this phenomenon. Second, it is unclear how exactly improved attentional guidance is enabled by target enhancement and distractor suppression. In the present magnetoencephalography experiment, we leveraged rapid invisible frequency tagging to answer these two outstanding questions. We found that the improved performance when searching implicitly familiar scenes was accompanied by a stronger neural representation of the target stimulus, at the cost specifically of those distractors directly surrounding the target. Crucially, this biasing of local attentional competition was behaviorally relevant when searching familiar scenes. Taken together, we conclude that implicitly learned spatial predictive context improves how we search our environment by sharpening the attentional field.

Intrinsic statistical regularity of topological charges revealed in dynamical disk model.

Identifying ordered structures hidden in the packings of particles is a common scientific question in multiple fields. In this work, we investigate the dynamical organizations of a large number of initially randomly packed repulsive particles confined on a disk under the Hamiltonian dynamics by the recently developed algorithm called the random batch method. This algorithm is specifically designed for reducing the computational complexity of long-range interacting particle systems. We highlight the revealed intrinsic statistical regularity of topological charges that is otherwise unattainable by the continuum analysis of particle density. We also identify distinct collective dynamics of the interacting particles under short- and long-range repulsive forces. This work shows the robustness and effectiveness of the concept of topological charge for characterizing the convoluted particle dynamics, and itdemonstrates the promising potential of the random batch method for exploring fundamental scientific questions arising in a variety of long-range interacting particle systems in soft matter physics and other relevant fields.