npj Science of Learning最新文献

The effect of mindfulness training on working memory is unclear. The current study sought to confirm the impact of mindfulness training on working memory for facial stimuli and to reveal the cognitive mechanisms underlying this effect by using drift-diffusion modeling (DDM). Using a delayed match-to-sample task with facial stimuli, we measured memory performance across five emotional categories (happy, sad, fearful, angry, neutral). Sixty participants received five-week emotion-targeted mindfulness training versus 60 waitlist controls. Assessments pre-training, post-training, and at one-month follow-up revealed significantly improved memory accuracy for all emotions except fear, with the effects persisting for one month. More importantly, drift-diffusion modeling showed increased drift rates across emotional categories post-training. Furthermore, accuracy improvements strongly correlated with drift rate enhancements within each emotion category. These findings demonstrate that mindfulness training induces lasting improvements in both accuracy and processing efficiency of visual working memory, independent of facial emotions, clarifying its cognitive mechanisms.

Visual-to-auditory sensory substitution devices (SSDs) translate images to sounds. One SSD, The vOICe, translates a pixel's vertical position into pitch and horizontal position into time. This mapping is primarily based on technical considerations for preserving image content in human-audible sounds without presupposing intuitiveness, although some literature also invokes crossmodal correspondences in perception, such as pitch for elevation. We investigated these presuppositions and the efficacy of learning a traditional algorithm (i.e., pitch indicating elevation and time indicating azimuth) versus a reversed algorithm (i.e., pitch indicating azimuth and time indicating elevation), or an arbitrary single-tone control mapping (i.e., each visual stimulus was represented by a single non-systematic pitch-time pairing without structured spatial correspondences). Sixty sighted adults participated with random assignment to the Traditional, Reversed, or Control groups. They completed learning and evaluation sessions using simplified black-and-white visual stimuli. Both the Traditional and Reversed groups learned mappings within 30 minutes and demonstrated successful recognition of novel stimuli, outperforming the Control group but not differing between them. Structured mappings facilitate SSD learning. Mapping pixel position onto spectral-temporal acoustic axes appears flexible, rather than anchored to cross-modal correspondences. These findings reveal how SSDs may be rendered bespoke across user, stimuli, and functionality levels.

Curriculum design in programs to build life skills often focuses on explicit learning methods that aim to increase declarative knowledge. However, this type of education has been shown to have minimal impact on behavior. We introduce a continuum of knowledge and argue that more flexible forms of knowledge are required to impact behavior, especially for novices. Using a randomized controlled trial conducted over several sessions, this study explores the ability of semi-flexible and flexible knowledge to promote optimal behaviors in the context of personal finances. We found that as knowledge became more flexible, desired changes in actual behavior became more likely. Our results provide evidence that life skills education programs, such as collegiate financial education, may be "barking up the wrong tree" with the focus on explicit learning. Expanding program design to incorporate a focus on flexible knowledge may improve the impact of such programs on desired behavior.

The proliferation of short-video platforms prompts critical investigation of their effects on human cognitive functions. We hypothesized that the frequent, user-driven content shifts inherent to short-video watching impair event segmentation, a cognitive process critical for continuous memory encoding. Combining behavioral, eye-tracking, and self-report data, we revealed that acute exposure to randomly selected short videos was associated with poorer memory for continuous movies, particularly in participants with more frequent daily short-video viewing. This effect was absent after viewing personalized short videos and did not apply to static image encoding tasks. Intersubject correlation analysis of eye movements revealed that random short video watching attenuated eye synchronization at event boundaries. Furthermore, Hidden Markov Model analysis indicated that personalized and random short videos induced qualitatively different latent event structures. These findings indicate that the algorithmic curation of content, not merely the short-video format, is a crucial factor shaping event segmentation and subsequent memory.

The visual system continuously generates predictions to guide behavior, yet how visuomotor adaptation relates to sensory detection and motor variability remains unclear. We addressed this question using joystick-based tasks: a visuomotor interception task with angular or speed perturbations, a sensory detection task, and a no-feedback motor variability task. Participants showed robust within-task responses, with angular discrepancies engaging both external (target-based) and self-referential control, while speed discrepancies primarily involved self-referential strategies. Gaze behavior reflected distinct tracking modes depending on perturbation type. However, cross-task regression analyses revealed weak associations between detection, variability, and adaptation. These dissociations were not due to noise or low power but reflected consistent performance patterns. Notably, within-subject variability exceeded between-subject variability across all tasks, highlighting trial-to-trial fluctuations as key drivers of behavior. Together, these findings support the view that predictive control relies on specialized, context-dependent mechanisms, in which task-specific computations adaptively integrate context and internal state dynamics.

Simulation-based learning (SBL) has been mostly studied in medical and STEM settings, where its specific role in learning is difficult to isolate. We ran a randomized, counterbalanced mixed design (N = 88) in which naïve adults learned two fictitious clinical cases via a live standardized-patient simulation or a structured text format; content and exposure time were matched. One month later, video- and text-based evaluations assessed accuracy, reaction time, and confidence. SBL yielded faster responses and higher accuracy than text. An order effect emerged: learning text first boosted subsequent SBL accuracy, while the opposite order did not. Individual imagery ability predicted better accuracy overall and more strongly in the text condition. Participants were more likely to report absolute confidence after SBL, with confidence more closely aligned to accuracy. These findings suggest that both SBL and text support durable learning, but their effects depend on instructional order and imagery ability, and SBL additionally enhances confidence-accuracy calibration.

The development of brain oscillation patterns during knowledge acquisition has gained attention, yet studies in realistic learning contexts remain limited. This study investigated dynamic brain activity across an 11-lesson biology course simulating a MOOC environment. Twenty undergraduates wore 14-channel Electroencephalography (EEG) headsets while completing lecture, virtual lab, and quiz tasks across three progressive stages. EEG signals from six participants (after quality screening) were analyzed using amplitude, power spectral density (PSD), and phase-locking index (PLI). Wilcoxon rank sum tests revealed significant stage- and task-related differences despite the small sample size, including increased frontal theta during quizzes, parietal alpha suppression during lectures, and high-beta enhancements in later stages of labs and quizzes. Machine learning models trained on EEG features achieved a classification accuracy of 83% for three learning stage discrimination, validating that the brain presents nonidentical functional patterns during cognitive learning. These results underscore the potential for real-time EEG-based personalized educational interventions.

Extrinsic reward and intrinsic motivation jointly influence memory. While studies show that rewards diminish the effect of intrinsic motivation on memory, rewards usually precede intrinsic motivation, leaving a gap in understanding their interaction when rewards are presented afterward. So, this study investigated how the timing of reward cues moderated the interaction between reward and intrinsic motivation on memory over time. Participants viewed film clips, rated interest, and completed memory tests at different delays. Reward cues were presented either before or after each clip and rating. Results showed that when cued before clips, rewards improved memory and attenuated forgetting for low-interest clips, thereby decreasing the interest effect on memory. Conversely, when cued after clips, rewards improved memory and attenuated forgetting for high-interest clips, thereby increasing the interest effect. Both patterns emerged through time-dependent consolidation. These findings clarify how extrinsic reward and intrinsic motivation interact across temporal dynamics to influence memory.

Early readers encounter thousands of printed words in children's books. The frequency with which they see each word shapes both neural and behavioral responses. Teachers also introduce novel written words through short, intensive learning experiences. Here we combined steady-state visual evoked potentials (SSVEP), corpus-based word frequency counts, and a novel two-week classroom "learning sprint" to examine and compare these two forms of experience-dependent plasticity. Cortical responses at 4 Hz to contrasts between real words of varying frequency (high: on average 1000 per million; medium: on average 200 per million) and pseudowords were sensitive to corpus-based frequency estimates-marking the first such finding using SSVEP. Strikingly, newly acquired low-frequency words (<1 per million)-taught in a child's own classroom versus counterbalanced words taught in two other classrooms-elicited cortical responses nearly identical to those evoked by high-frequency words versus pseudowords. Furthermore, 1 Hz responses to new vocabulary learning was linked to individual differences in reading skills, including word decoding and rapid automatic naming. Together, these findings highlight the causal impact of authentic instruction and the value of neuroscience-informed methods in education research.

Growth mindset is widely perceived to be a powerful lever for reducing inequalities in learning outcomes. This study investigates whether a growth mindset moderates or mediates the effect of socio-economic status (SES) on academic achievement across 73 countries, using data from the 2022 Programme for International Student Assessment (PISA). To do this, we employ a four-way decomposition approach to separate the total effect of SES on standardized test scores in math, reading, and science into direct, indirect, and interaction effects. The results show that growth mindset mediates only a small portion of the effect of SES on student achievements, accounting for no more than 2.9% to 3.2% of the total effect, depending on the subject. These findings challenge the influential idea that growth mindset can 'temper' the effect of poverty on academic achievement.