Underlying Learning Mechanisms Research Articles

Intensive Modes of Study and the Need to Focus on the Process of Learning in Higher Education

In the context of a constantly evolving international higher education sector, this commentary emphasises the need for consilience between basic research on learning processes and observations from intensive modes of study. Following a discussion of conflicting evidence on optimal learning time frames, we advocate for seeking alignment between classroom practices with underlying learning mechanisms. We argue for a unified understanding of effective learning beyond notions of the credit point hour or volume of learning, focusing on processes rather than mere inputs and outputs. A collaborative approach between researchers, educators, and policymakers aiming for consilience has the potential to provide practical insights and strategies to enhance student learning and success. Understanding the mechanisms beneath the impact of intensive modes of study, as outlined in this special issue, has the potential to advance the conversation about quality higher education for the 21st century.

Exploring brain plasticity in developmental dyslexia through implicit sequence learning.

Developmental dyslexia (DD) is defined as difficulties in learning to read even with normal intelligence and adequate educational guidance. Deficits in implicit sequence learning (ISL) abilities have been reported in children with DD. We investigated brain plasticity in a group of 17 children with DD, compared with 18 typically developing (TD) children, after two sessions of training on a serial reaction time (SRT) task with a 24-h interval. Our outcome measures for the task were: a sequence-specific implicit learning measure (ISL), entailing implicit recognition and learning of sequential associations; and a general visuomotor skill learning measure (GSL). Gray matter volume (GMV) increased, and white matter volume (WMV) decreased from day 1 to day 2 in cerebellar areas regardless of group. A moderating effect of group was found on the correlation between WMV underlying the left precentral gyrus at day 2 and the change in ISL performance, suggesting the use of different underlying learning mechanisms in DD and TD children during the ISL task. Moreover, DD had larger WMV in the posterior thalamic radiation compared with TD, supporting previous reports of atypical development of this structure in DD. Further studies with larger sample sizes are warranted to validate these results.

Intensive modes of study and the need to focus on the process of learning in higher education

In the context of a constantly evolving international higher education sector, this commentary emphasises the need for consilience between basic research on learning processes and observations from intensive modes of study. Following a discussion of conflicting evidence on optimal learning time frames, we advocate for seeking alignment between classroom practices with underlying learning mechanisms. We argue for a unified understanding of effective learning beyond notions of the credit point hour or volume of learning, focusing on processes rather than mere inputs and outputs. A collaborative approach between researchers, educators, and policymakers aiming for consilience has the potential to provide practical insights and strategies to enhance student learning and success. Understanding the mechanisms beneath the impact of intensive modes of study, as outlined in this special issue, has the potential to advance the conversation about quality higher education for the 21st century.

Exploring real-time functional magnetic resonance imaging neurofeedback in adolescents with disruptive behavior disorder and callous unemotional traits

IntroductionAdolescents with increased callous unemotional traits (CU traits) in the context of disruptive behavior disorder (DBD) show a persistent pattern of antisocial behavior with shallow affect and a lack of empathy or remorse. The amygdala and insula as regions commonly associated with emotion processing, empathy and arousal are implicated in DBD with high CU traits. While behavioral therapies for DBD provide significant but small effects, individualized treatments targeting the implicated brain regions are missing. MethodsIn this explorative randomized controlled trial we randomly assigned twenty-seven adolescents with DBD to individualized real-time functional magnetic resonance neurofeedback (rtfMRI-NF) or behavioral treatment as usual (TAU). Visual feedback of either amygdala or insula activity was provided during rtfMRI-NF by gauges and included a simple and concurrent video run plus a transfer run. A linear mixed model (LMM) was applied to determine improvement of self-regulation. Specificity was assessed by correlating individual self-regulation improvement with clinical outcomes. ResultsThe rtfMRI-NF (n = 11) and TAU (n = 10) completers showed comparable and significant clinical improvement indicating neither superiority nor inferiority of rtfMRI-NF. The exploratory LMM revealed successful learning of self-regulation along the course of training for participants who received feedback from the amygdala. A significant exploratory correlation between individual target region activity in the simple run and clinical improvement was found for one dimension of DBD. ConclusionsThis exploratory study demonstrated feasibility and suggests clinical efficacy of individualized rtfMRI-NF comparable to active TAU for adolescents with DBD and increased CU traits. Further studies are needed to confirm efficacy, specificity and to clarify underlying learning mechanisms.

Virtual reality and augmented reality smartphone applications for upskilling care home workers in hand hygiene: a realist multi-site feasibility, usability, acceptability, and efficacy study.

To assess the feasibility and implementation, usability, acceptability and efficacy of virtual reality (VR), and augmented reality (AR) smartphone applications for upskilling care home workers in hand hygiene and to explore underlying learning mechanisms. Care homes in Northwest England were recruited. We took a mixed-methods and pre-test and post-test approach by analyzing uptake and completion rates of AR, immersive VR or non-immersive VR training, validated and bespoke questionnaires, observations, videos, and interviews. Quantitative data were analyzed descriptively. Qualitative data were analyzed using a combined inductive and deductive approach. Forty-eight care staff completed AR training (n = 19), immersive VR training (n = 21), or non-immersive VR training (n = 8). The immersive VR and AR training had good usability with System Usability Scale scores of 84.40 and 77.89 (of 100), respectively. They had high acceptability, with 95% of staff supporting further use. The non-immersive VR training had borderline poor usability, scoring 67.19 and only 63% would support further use. There was minimal improved knowledge, with an average of 6% increase to the knowledge questionnaire. Average hand hygiene technique scores increased from 4.77 (of 11) to 7.23 after the training. Repeated practice, task realism, feedback and reminding, and interactivity were important learning mechanisms triggered by AR/VR. Feasibility and implementation considerations included managerial support, physical space, providing support, screen size, lagging Internet, and fitting the headset. AR and immersive VR apps are feasible, usable, and acceptable for delivering training. Future work should explore whether they are more effective than previous training and ensure equity in training opportunities.

Active before passive tasks improve long-term visual learning in difficult-to-classify skin lesions

BackgroundVisual perceptual learning is essential in many domains, including medicine. Prior research has shown that combining active and passive tasks improves learning efficiency. However, previous studies have only implemented passive before active tasks, contradicting findings on productive failure learning designs. AimsWe aimed to replicate and extend earlier results by comparing 1) combined versus uniform active and passive tasks and 2) active before passive versus passive before active tasks in visual learning. The to-be-learned skill was the detection of pigmented skin cancer. SampleThe sample consisted of 161 university students without professional knowledge about skin lesion classification. MethodsWe randomly assigned the participants to four groups: 1) active before passive, 2) passive before active, 3) uniform active, and 4) uniform passive tasks. The students completed the learning intervention, an intermediate and three post-tests (immediate, two days delayed and two weeks delayed) online. ResultsIn line with our hypotheses, we found that learning with combined active and passive tasks led to higher diagnostic accuracy in difficult skin lesion classification tasks in the two weeks delayed post-test than learning with only one of the two task types. Furthermore, we found that active before passive tasks resulted in higher diagnostic accuracy than passive before active tasks. ConclusionOur findings suggest that initial active tasks improve long-term visual learning outcomes in difficult melanoma detection tasks, but future research needs to confirm this result and explore the underlying learning mechanisms further.

The relationship between semantic and episodic memory: evidence from a case of severe anterograde amnesia

ABSTRACT It is increasingly being recognized that new declarative, consciously accessible information can be learned in anterograde amnesia, but it is not clear whether this learning is supported by episodic or semantic memory. We report a case of a 55-year-old man who experienced severe amnesia after limited damage to the medial temporal lobe following neurosurgical complications. His general cognitive performance and knowledge of new French words and public events that occurred before and after the onset of amnesia were assessed. Performance remained satisfactory on post-morbid vocabulary and public events, with a drop in performance observed for very recent public events only, while knowledge of very recent vocabulary was comparable to that of the control subjects. The implications of these findings for our understanding of the underlying learning mechanisms are discussed. This is the first report of acquisition of consciously accessible postmorbid knowledge of public events in a patient with severe amnesia.

Morphemes as letter chunks: Linguistic information enhances the learning of visual regularities

We have previously shown that readers use co-occurrence statistics to learn about the presence and position of affix-like chunks in strings of pseudo-letters (Lelonkiewicz, Ktori & Crepaldi, 2020). These findings were taken as evidence that visual statistical learning might be implicated in morphological processing during visual word recognition. The present study seeks to specify this claim by (a) establishing the visual, language-agnostic nature of the underlying learning mechanism and (b) examining it in the presence of higher-order linguistic information. In Experiments 1a and 1b, readers were familiarized with strings of abstract shapes that involved affix-like chunks of frequently co-occurring shapes. We found that readers were sensitive to the presence and position of chunks. Further experiments revealed that presence and position effects were stronger when readers were exposed to letter strings which allowed access to orthographic and phonological representations (Experiments 2a and 2b), and were enhanced by access to semantics (Experiment 3). Our study demonstrates that the learning of visual regularities supports chunk identification both in purely visual and language-like materials, and that the availability of linguistic information enhances this learning.

Advancements on action learning and lean complementarity: a case of developing leaders as lean learning facilitators

ABSTRACT This paper seeks to advance the understanding of the complementarity between action learning and lean. Today, this is an underexplored research area, despite the high degree of similarities and syngeneic possibilities between these two research streams. The paper describes an action learning intervention at VELUX, a Danish rooftop manufacturer designed to develop its leaders as lean learning facilitators to cope with the increasing velocity of change stemming from growth, sustainability, and digitalisation agendas. The paper locates the complementary between action learning and lean in the extant literature and presents an account of practice from VELUX for extrapolating five promoting factors for developing leaders as lean learning facilitators. The paper concludes that lean complements action learning with a suite of concepts, systems, practices, and methods for institutionalising ongoing action learning and concepts on how to think and act as a leader to foster a lean learning system consisting of empowered and proficient problem-solvers. Furthermore, action learning complements lean with the underlying learning mechanisms of facilitating and sustaining the change towards instituting leaders as lean learning facilitators and adopting a lean learning system.

Placebo effects on cutaneous pain and itch: a systematic review and meta-analysis of experimental results and methodology.

Placebo effects, positive treatment outcomes that go beyond treatment processes, can alter sensations through learning mechanisms. Understanding how methodological factors contribute to the magnitude of placebo effects will help define the mechanisms by which these effects occur. We conducted a systematic review and meta-analysis of experimental placebo studies in cutaneous pain and itch in healthy samples, focused on how differences in methodology contribute to the resulting placebo effect magnitude. We conducted meta-analyses by learning mechanism and sensation, namely, for classical conditioning with verbal suggestion, verbal suggestion alone, and observational learning, separately for pain and itch. We conducted subgroup analyses and meta-regression on the type of sensory stimuli, placebo treatment, number of acquisition and evocation trials, differences in calibrated intensities for placebo and control stimuli during acquisition, age, and sex. We replicated findings showing that a combination of classical conditioning with verbal suggestion induced larger placebo effects on pain ( k = 68, g = 0 . 59) than verbal suggestion alone ( k = 39, g = 0.38) and found a smaller effect for itch with verbal suggestion alone ( k = 7, g = 0.14). Using sham electrodes as placebo treatments corresponded with larger placebo effects on pain than when topical gels were used. Other methodological and demographic factors did not significantly affect placebo magnitudes. Placebo effects on pain and itch reliably occur in experimental settings with varied methods, and conditioning with verbal suggestion produced the strongest effects. Although methods may shape the placebo effect to some extent, these effects appear robust overall, and their underlying learning mechanisms may be harnessed for applications outside the laboratory.

Female fruit flies copy the acceptance, but not the rejection, of a mate

AbstractAcceptance and avoidance can be socially transmitted, especially in the case of mate choice. When a Drosophila melanogaster female observes a conspecific female (called demonstrator female) choosing to mate with one of two males, the former female (called observer female) can memorize and copy the latter female’s choice. Traditionally in mate-copying experiments, demonstrations provide two types of information to observer females, namely, the acceptance (positive) of one male and the rejection of the other male (negative). To disentangle the respective roles of positive and negative information in Drosophila mate copying, we performed experiments in which demonstrations provided only one type of information at a time. We found that positive information alone is sufficient to trigger mate copying. Observer females preferred males of phenotype A after watching a female mating with a male of phenotype A in the absence of any other male. Contrastingly, negative information alone (provided by a demonstrator female actively rejecting a male of phenotype B) did not affect future observer females’ mate choice. These results suggest that the informative part of demonstrations in Drosophila mate-copying experiments lies mainly, if not exclusively, in the positive information provided by the copulation with a given male. We discuss the reasons for such a result and suggest that Drosophila females learn to prefer the successful males, implying that the underlying learning mechanisms may be shared with those of appetitive memory in non-social associative learning.

Artificial intelligence in renewable systems for transformation towards intelligent buildings

Carbon-neutrality transition in building sectors requires combinations of renewable systems and artificial intelligence (AI) for robustness, reliability, automation, and flexibility. In this study, a comprehensive review on AI applications in renewable systems, have been conducted, to report the current progress, tendency and challenges. Underlying learning mechanisms of AI-based various applications have been studied, in terms of modelling techniques on solar power forecasting, multi-level stochastic uncertainty analysis, smart controls, fault detection and diagnosis, single and multi-objective optimisations and intelligent buildings. Furthermore, AI techniques in renewable energy utilisations have been clearly reviewed, with respect to solar potential evaluation, multi-level stochastic uncertainty analysis, smart controls, fault detection and diagnosis, single and multi-objective optimisations. Results showed that, compared to physical models, the data-driven models show superiority in modelling simplification and modification, high prediction accuracy, and computational efficiency. In order to improve the system operational robustness under multi-level scenario uncertainty, a generic approach with artificial intelligence was proposed, including nonlinear data-driven model development, uncertainty-based energy performance prediction, stochastic uncertainty and statistical analysis. Furthermore, artificial intelligence can also be applied in renewable systems for security, reliability and stability. This study provides a clear roadmap on historical development, recent advances, cutting-edge techniques and progress, and future challenges of AI applications in renewable energy systems, paving path for developing smart and resilient renewable energy systems for decarbonization in buildings.

Recent Advances in Representation Learning for Electronic Health Records: A Systematic Review

Representation Learning (RL) aims to convert data into low-dimensional and dense real-valued vectors, so as to realize reasoning in vector space. RL is one of the important research contents in the analysis of health data. This paper systematically reviews the latest research on Electronic Health Records (EHR) RL. We searched the Web of Science, Google Scholar, and Association for Computing Machinery Digital Library for papers involving EHR RL. On the basis of literature review, we propose a new taxonomy to categorize the state-of-the-art EHR RL methods into three categories: statistics learning-based RL methods, knowledge RL methods and graph RL methods. We analyze and summarize their characteristics according to the input data form and underlying learning mechanisms. In addition, we provide evaluation strategies to verify the quality of EHR representations from both intrinsic and extrinsic perspectives. Finally, we put forward three promising research directions to promote future research. Overall, this survey aims to provide a profound overview of state-of-the-art developments in the field of EHR RL and to help researchers find the most appropriate methods.

Context-Specific Habituation: A Review.

Simple SummaryHabituation reflects the ability to learn to ignore irrelevant stimuli, which form the vast majority of the sensory input impinging on any organism’s sensory systems at a given moment. However, although habituation is often described as one of the simplest forms of learning affected only by the stimulus features, such as frequency or intensity, in fact evidence exists showing that habituation can be specific for the context in which it takes place. This shows that habituation, in some cases, exhibits an associative nature, and that the underlying learning mechanism is more sophisticated than previously believed.Habituation consists of the progressive response decrement to a repeated stimulation, a response decline that is not accounted for by sensory or motor fatigue. Together with sensitization, habituation has been traditionally considered to be a prototypical example of non-associative learning, being affected only by the features of the stimulation, as for instance its intensity or frequency. However, despite this widespread belief, evidence exists showing that habituation can be specific to the context of the stimulation, thus suggesting that habituation can have an associative nature. Such an unexpected characteristic of habituation was in fact predicted by a theoretical model of associative learning proposed by Wagner in a series of works that appeared in the late 1970s. Here, we critically review the experimental data that since then have been accumulated in support of this hypothesis. What emerges from the literature is that context-specific habituation is common to several animal species and that the ability to form an association between the habituating stimulus and its context is independent of the complexity of the animal’s nervous system. Finally, context-specific habituation is observed for a variety of organism’s responses, ranging from visceral to motor and mental activities.

Learning Processes and Mechanisms for Interorganizational Projects: Insights From the Islamabad–Rawalpindi Metro Bus Project

The purpose of this article is to understand how learning is disseminated between individual, project, organizational, and interorganizational levels—A phenomenon that remains underresearched, particularly in the context of interorganizational projects. This article unpacks the learning processes and underlying learning mechanisms that disseminate learning at different levels. For this purpose, we examine the Islamabad–Rawalpindi metro bus project in Pakistan. Using an abductive research approach, data for this article are collected through interviews, archival records, and illustrative materials. The findings reveal 11 learning mechanisms: 1) learning by observing; 2) learning from past experiences; 3) learning from an expert; 4) learning from the experiences of others; 5) learning by joint problem-solving; 6) learning by working together; 7) learning from mistakes; 8) learning by observing other projects; 9) learning by building organizational memory; 10) learning from existing organizational memory; and 11) learning by coordinating and consulting. These learning mechanisms trigger learning processes of intuiting, interpreting, integrating, institutionalizing, and intertwining at the individual, team, organizational, and interorganizational levels. This article enriches the understanding of project-based learning, focusing on different levels and exploring the interplay between learning processes and mechanisms.

Use of explicit processes during a visually guided locomotor learning task predicts 24-h retention after stroke.

Implicit and explicit processes can occur within a single locomotor learning task. The combination of these learning processes may impact how individuals acquire/retain the task. Because these learning processes rely on distinct neural pathways, neurological conditions may selectively impact the processes that occur, thus, impacting learning and retention. Thus, our purpose was to examine the contribution of implicit and explicit processes during a visually guided walking task and characterize the relationship between explicit processes and performance/retention in stroke survivors and age-matched healthy adults. Twenty chronic stroke survivors and twenty healthy adults participated in a 2-day treadmill study. Day 1 included baseline, acquisition1, catch, acquisition2, and immediate retention phases, and day 2 included 24-h retention. During acquisition phases, subjects learned to take a longer step with one leg through distorted visual feedback. During catch and retention phases, visual feedback was removed and subjects were instructed to walk normally (catch) or how they walked during the acquisition phases (retention). Change in step length from baseline to catch represented implicit processes. Change in step length from catch to the end of acquisition2 represented explicit processes. A mixed ANOVA found no difference in the type of learning between groups (P = 0.74). There was a significant relationship between explicit processes and 24-h retention in stroke survivors (r = 0.47, P = 0.04) but not in healthy adults (r = 0.34, P = 0.15). These results suggest that stroke may not affect the underlying learning mechanisms used during locomotor learning, but that these mechanisms impact how well stroke survivors retain the new walking pattern.NEW & NOTEWORTHY This study found that stroke survivors used implicit and explicit processes similar to age-matched healthy adults during a visually guided locomotion learning task. The amount of explicit processes was related to how well stroke survivors retained the new walking pattern but not to how well they performed during the task. This work illustrates the importance of understanding the underlying learning mechanisms to maximize retention of a newly learned motor behavior.

The ERP, frequency, and time-frequency correlates of feedback processing: Insights from a large sample study.

Human learning, at least in part, appears to be dependent on the evaluation of how outcomes of our actions align with our expectations. Over the past 23years, electroencephalography (EEG) has been used to probe the neural signatures of feedback processing. Seminal work demonstrated a difference in the human event-related potential (ERP) dependent on whether people were processing correct or incorrect feedback. Since then, these feedback evoked ERPs have been associated with reinforcement learning and conflict monitoring, tied to subsequent behavioral adaptations, and shown to be sensitive to a wide range of factors (e.g., Parkinson's disease). Recently, research has turned to frequency decomposition techniques to examine how changes in the EEG power spectra are related to underlying learning mechanisms. Although the literature on the neural correlates of feedback processing is vast, there are still methodological discrepancies and differences in results across studies. Here, we provide reference results and an investigation of methodological considerations for the ERP (reward positivity) and frequency (delta and theta power) correlates of feedback evaluation with a large sample size. Specifically, participants (n=500) performed a two-armed bandit task while we recorded EEG. Our findings provide key information about the data characteristics and relationships that exist between the neural signatures of feedback evaluation. Additionally, we conclude with selected methodological recommendations for standardization of future research. All data and scripts are freely provided to facilitate open science.

Task Aspects Triggering Observational Learning in Jackdaws (Corvus monedula)

Complex novel tasks are often used in animal cognition research to allow discrimination between various learning mechanisms. Successful performance relies on the capacity to identify informational cues from features in the environment. Additionally, observational learning is often considered more beneficial for survival than individual learning. Despite the importance of controlling task complexity, it can often be challenging to operationalize. This study investigated whether jackdaws, a highly social corvid species, can learn to drop stones inside a tube to release a reward after observing a trained conspecific. Additionally, it aimed to identify the underlying learning mechanisms and to detect the informational cues triggering learning. A research design was developed to highlight different aspects of the required action sequence. Experimental conditions included a conspecific model demonstrating the full sequence, parts of the sequence, consuming the reward without solving the task, and consuming the reward after the solution was demonstrated without a visible model. None of the 12 naïve jackdaws solved the task in pretests. Two subjects started solving in test sessions and they developed modified versions of the demonstrated action. Observing the full action sequence performed by a conspecific seemed to trigger learning. The majority of the subjects exhibited changes in their stone-oriented behavior, most likely due to stimulus and/or local enhancement. As predicted, jackdaws were influenced by conspecific model demonstrations when manipulating a complex novel foraging task. Factors contributing to the apparent task difficulty and directions for future studies are discussed within a tri-dimensional framework including the task, setup, and individuals.

Referent product information from a credible source: How front line employees can stimulate acceptance of incrementally new food products

Generating consumer acceptance for incrementally new food products remains a challenge. Drawing on analogical learning theory, we explore the combined effect of providing information about the new product’s closest referent (‘the product is like…’) and the instrumental value of store’s frontline employees (FLEs) as credible source of this information on a new food product’s acceptance. Using the context of aquaculture, two consecutive experimental studies with a joint data collection have been conducted among UK consumers. Study 1 focuses on the adoption of fillets of an unfamiliar fish species with the FLE communicating newness, or newness including closest referent information. Results confirm that information provided by the FLE has a positive effect on consumer attitude and willingness to buy the new product, but a similar rather than stronger effect is found for the condition including closest referent information. Using a subsample of the data of Study 1, Study 2 explores the underlying learning mechanisms involved. The findings show that the influence of the FLE’s information on the consumer’s attitude towards the new product is positively mediated by the consumer’s attitude towards FLE’s advice. Additionally, the positive indirect effect of the FLE’s information regarding closest referent on willingness to buy is conditional on the consumer’s (dis)liking of the referent product and his/her need to taste the new product first. These findings confirm that offering information on closest referent by FLEs is powerful and can be a useful variable in consumer segmentation.

Reinforcement learning as an intermediate phenotype in psychosis? Deficits sensitive to illness stage but not associated with polygenic risk of schizophrenia in the general population

BackgroundSchizophrenia is a complex disorder in which the causal relations between risk genes and observed clinical symptoms are not well understood and the explanatory gap is too wide to be clarified without considering an intermediary level. Thus, we aimed to test the hypothesis of a pathway from molecular polygenic influence to clinical presentation occurring via deficits in reinforcement learning. MethodsWe administered a reinforcement learning task (Go/NoGo) that measures reinforcement learning and the effect of Pavlovian bias on decision making. We modelled the behavioural data with a hierarchical Bayesian approach (hBayesDM) to decompose task performance into its underlying learning mechanisms. Study 1 included controls (n = 29, F|M = 0.81), At Risk Mental State for psychosis (ARMS, n = 23, F|M = 0.35) and FEP (First-episode psychosis, n = 26, F|M = 0.18). Study 2 included healthy adolescents (n = 735, F|M = 1.06), 390 of whom had their polygenic risk scores for schizophrenia (PRSs) calculated. ResultsPatients with FEP showed significant impairments in overriding Pavlovian conflict, a lower learning rate and a lower sensitivity to both reward and punishment. Less widespread deficits were observed in ARMS. PRSs did not significantly predict performance on the task in the general population, which only partially correlated with measures of psychopathology. ConclusionsReinforcement learning deficits are observed in first episode psychosis and, to some extent, in those at clinical risk for psychosis, and were not predicted by molecular genetic risk for schizophrenia in healthy individuals. The study does not support the role of reinforcement learning as an intermediate phenotype in psychosis.