npj Science of Learning最新文献

When memories are reactivated during sleep, they are potentially transformed and strengthened. However, disturbed sleep may make this process ineffective. In a prior study, memories formed shortly before sleep were weakened by auditory stimulation when that stimulation provoked memory reactivation while also disrupting sleep - a procedure known as targeted memory reactivation with sleep disruption (TMR-SD). Here we used TMR-SD to test whether memory weakening occurs for less-fragile memories. Participants first learned locations of 74 objects on a monitor. One week later, TMR-SD auditory cues linked with 50% of the previously learned object locations were presented during sleep. Even though the cues disturbed sleep, memories were not weakened when reactivated in this way, compared to when not reactivated. Whereas memory storage is vulnerable to disruption shortly after learning, this new evidence supports the notion that memory storage gradually gains resistance to the harm caused by reactivation combined with sleep disruption.

In many non-human species, learning retention decreases temporarily following training. This has led to the suggestion that these lapses reflect a fundamental component of memory formation. If so, transient memory lapses should also be prevalent in humans, and should occur for all types of learning. In line with these predictions, we report two cases of transient memory lapses in humans that occur 1-3 h after training on a perceptual-discrimination task. The results indicate that the occurrence of transient memory lapses extends to perceptual learning, a form of skill learning, and suggest that transient memory lapses may be a common but overlooked facet of memory formation in humans.

Motivational (i.e., appetitive or aversive) cues can bias value-based decisions by affecting either direction and intensity of instrumental actions. Despite several findings describing important interindividual differences in these biases, whether biological sex can also play a role is still up to debate. By comparing females and males in both appetitive and aversive Pavlovian-to-Instrumental Transfer paradigms we found that, while motivational cues similarly bias the direction of instrumental actions in both sexes, the intensity of such actions is increased by the cue in male participants only. The present results constitute compelling evidence that a crucial motivational bias of daily actions directed to obtaining rewards or avoiding punishments is modulated by biological sex. This evidence sheds new light on the role of sex in motivational processes that underlie decision-making, highlighting the importance of considering sex as a crucial factor in future research on this topic.

Observational practice is discussed as a substitute for physical practice for motor learning and adaptation. We systematically reviewed the literature on observational practice in reaching and aiming tasks. Our objectives were to identify (i) performance differences between observational and physical practice; (ii) factors that contribute to adaptation following observational practice; and (iii) the neural correlates of observational practice. We found 18 studies, all investigated adaptation of reaching in visuomotor rotations or force-field perturbations. Results of the studies showed that observational practice led to adaptation in both, visuomotor rotation and force-field paradigms (d = -2.16 as compared to no practice). However, direct effects were considerably smaller as compared to physical practice (d = 4.38) and aftereffects were absent, suggesting that observational practice informed inverse, but not forward modes. Contrarily, neurophysiological evidence in this review showed that observational and physical practice involved similar brain regions.

This study explores the use of Large Language Models (LLMs), specifically GPT-4, in analysing classroom dialogue-a key task for teaching diagnosis and quality improvement. Traditional qualitative methods are both knowledge- and labour-intensive. This research investigates the potential of LLMs to streamline and enhance this process. Using datasets from middle school mathematics and Chinese classes, classroom dialogues were manually coded by experts and then analysed with a customised GPT-4 model. The study compares manual annotations with GPT-4 outputs to evaluate efficacy. Metrics include time efficiency, inter-coder agreement, and reliability between human coders and GPT-4. Results show significant time savings and high coding consistency between the model and human coders, with minor discrepancies. These findings highlight the strong potential of LLMs in teaching evaluation and facilitation.

Gestures accent and illustrate our communication. Although previous studies have uncovered the positive effects of gestures on communication, little is known about the specific cognitive functions of different types of gestures, or the instantaneous multi-brain dynamics. Here we used the fNIRS-based hyperscanning technique to track the brain activity of two communicators, examining regions such as the PFC and rTPJ, which are part of the mirroring and mentalizing systems. When participants collaboratively solved open-ended realistic problems, we characterised the dynamic multi-brain states linked with specific social behaviours. Results demonstrated that gestures are associated with enhanced team performance, and different gestures serve distinct cognitive functions: interactive gestures are accompanied by better team originality and a more efficient inter-brain network, while fluid gestures correlate with individual cognitive fluency and efficient intra-brain states. These findings reveal a close association between social behaviours and multi-brain networks, providing a new way to explore the brain-behaviour relationship.

Arithmetic ability is critical for daily life, academic achievement, career development, and future economic success. Individual differences in arithmetic skills among children and adolescents are related to variations in brain structures. Most existing studies have used hypothesis-driven region of interest analysis. To identify distributed brain regions related to arithmetic ability, we used data-driven cross-validated predictive models to analyze cross-sectional behavioral and structural MRI data in children and adolescents. The gray matter volume (GMV) of widespread brain regions reliably predicted arithmetic abilities measured by the Comprehensive Mathematical Abilities Test. Furthermore, we applied neuroimaging-transcriptome association analysis to explore transcriptional signatures associated with structural patterns of arithmetic ability. Structural patterns of arithmetic ability primarily correlated with transcriptional profiles enriched for genes involved in transmembrane transport and synaptic signaling. Our findings enhance our understanding of the neural and genetic mechanisms underlying children's arithmetic ability and offer a practical predictor for arithmetic skills during development.

Grit, a non-cognitive skill that indicates perseverance and passion for long-term goals, has been shown to predict academic achievement. This paper provides evidence that grit also predicts student outcomes during the challenging period of the Covid-19 pandemic. We use a unique dataset from a digital learning platform in the United Arab Emirates to construct a behavioral measure of grit. We find that controlling for baseline achievement, students who were grittier according to this measure before the pandemic, register lower declines in math and science scores during the coronavirus period. Using machine learning, behavioral data obtained from the platform prior to the pandemic can explain 77% of the variance in academic resilience. A survey measure of grit coming from the same students, on the other hand, does not have significant predictive power over performance changes. Our findings have implications for interventions on non-cognitive skills, as well as how data from digital learning platforms can be used to predict student behavior and outcomes, which we expect will be increasingly relevant as AI-based learning technologies become more common.

This Perspective article expands on a working memory-dependent dual-process model, originally proposed by Zheng et al.¹, to elucidate individual differences in the testing effect. This model posits that the testing effect comprises two processes: retrieval-attempt and post-retrieval re-encoding. We substantiate this model with empirical evidence and propose future research. This model invites further studies on the trade-off between testing benefits and WM demands, facilitating the development of personalized educational practices.

We show that learners generalized more broadly around the learned stimulus when they expected more variability between the learning set and the generalization set, as well as within the generalization set. Experiments 1 and 3 used a predictive learning task and demonstrated border perceptual generalization both when expected variability was manipulated explicitly via instructions (Experiment 1), and implicitly by increasing temporal distance to the anticipated application of learning (Experiment 3). Experiment 2 showed that expecting to apply learning in the more distant future increases expected variability in the generalization set. We explain the relation between expected variability and generalization as an accuracy-applicability trade-off: when learners anticipate more variable generalization targets, they "cast a wider net" during learning, by attributing the outcome to a broader range of stimuli. The use of more abstract, broader categories when anticipating a more distant future application aligns with Construal Level Theory of psychological distance.