Where it is used
Each entry identifies the volume, edition, chapter or appendix in which the work appears.
The evidence behind the series
Browse the scientific research and professional literature used across all seven volumes of The Science of Learning for Educators.
Search the collection
Choose an author range, search a title or concept, or narrow the collection by source type. Open any result to see where it appears in the books, read its evidence summary and follow available source links.
58 references
King tests whether teaching children how to ask questions and generate explanations can improve peer learning after teacher-presented science lessons. Fourth- and fifth-grade students studied in pairs using self-generated questions.
One condition used prompts connecting ideas within the lesson; another added questions designed to activate prior knowledge and link it with new material; controls received less generative support. All treatment students learned to explain answers rather than exchange short responses. Analyses of dialogue, knowledge maps, and comprehension showed more complex knowledge construction when questioning explicitly connected new ideas with prior experience, with lesson-based generative questioning also offering benefits. The mechanism is not questioning frequency alone. Productive prompts invite comparison, causal reasoning, prediction, justification, and integration, while explaining makes understanding visible and gives partners material to examine. Teachers must model the discourse, provide stems, and establish norms for listening and elaborating.
The study involved particular ages, science lessons, and structured pair work, so transfer requires care. Its practical contribution is a teachable method for turning peer discussion from answer trading into collaborative knowledge construction.
Kirschner, Paas, and Kirschner use cognitive load theory to explain when collaborative learning may be more efficient than individual learning. Working memory is limited, and complex tasks can exceed one learner’s processing capacity.
A group can distribute interacting information across several working memories, combine complementary knowledge, and build a shared representation. Collaboration also creates transaction costs: communicating, coordinating, resolving different interpretations, and maintaining shared awareness consume resources. For simple tasks these costs may outweigh the benefit, making individual work more efficient. As intrinsic task complexity rises, distribution can become advantageous, especially when members possess relevant knowledge and coordinate effectively. The authors review an inconsistent literature and propose task complexity as a key moderator rather than treating group learning as uniformly superior. They distinguish performance, learning, mental effort, and efficiency, since a group may solve a task without every member acquiring transferable schemas.
Instructional design should therefore select complex tasks, ensure individual accountability and participation, support coordination, and assess individual transfer. The article offers a theoretical framework and research agenda rather than a universal rule to put learners into groups.
Kirschner and De Bruyckere challenge two popular claims about contemporary students. The “digital native” story assumes that growing up around technology automatically produces sophisticated information literacy, strategic learning, and a need for fundamentally different education.
Evidence instead shows wide variation in access and skill, frequent concentration on a limited set of everyday applications, and no generational transformation of cognitive architecture. The multitasking claim confuses rapid task switching with simultaneous cognitive processing. When two activities require attention, switching imposes time, error, and memory costs; media multitasking can fragment concentration and hinder comprehension and learning. Familiarity with devices therefore does not remove the need to teach search, source evaluation, privacy, knowledge organisation, and deliberate tool use. The authors do not argue that technology is educationally irrelevant or that learners never change.
They reject unsupported generational labels and designs built on presumed innate competence. Teachers and policymakers should assess actual knowledge and needs, choose technology for pedagogical purposes, reduce unnecessary competing demands, and explicitly develop digital and information skills rather than abandoning guidance because students appear fluent with consumer media.
Kirschner, Hendrick, and Heal translate 30 influential studies and scholarly works on teaching and teacher effectiveness into accessible guidance for educators. Each chapter introduces a seminal contribution, explains its research context and central findings, examines cautions or limitations, and draws practical implications.
The selections span instructional objectives, explicit and direct instruction, modelling, guided practice, feedback, assessment, classroom management, teacher knowledge, expertise, reflection, and the relationship between teaching methods and student learning. Rather than offering a single prescriptive method, the authors show how different evidence-informed practices address different parts of teaching. Recurring themes include specifying worthwhile outcomes, connecting new material to prior knowledge, managing cognitive demands, checking understanding, giving learners supported practice, and adapting instruction on the basis of evidence. The book also warns against oversimplified slogans and measures that become targets rather than useful indicators.
Its purpose is to help teachers, school leaders, and teacher educators understand the research foundations beneath effective classroom decisions and exercise informed professional judgment instead of applying isolated techniques mechanically.
View or buy the cited work >>Read a similar article free online >>
Kirschner, Sweller, and Clark argue that minimally guided instruction is generally ineffective and inefficient for novices acquiring biologically secondary knowledge. Their case combines cognitive architecture with evidence comparing guided teaching and discovery-oriented approaches.
Working memory is severely limited when processing unfamiliar information, whereas organised schemas in long-term memory enable experts to manage complex tasks. Novices asked to search a large problem space while simultaneously learning its structure can therefore experience avoidable cognitive load. Guidance through explanations, modelling, worked examples, prompts, sequencing, and practice reduces unproductive search and supports schema construction. As learners gain domain knowledge, support can fade because prior knowledge provides internal guidance; methods effective for experts may be unsuitable for beginners and vice versa.
The article groups constructivist, discovery, problem-based, experiential, and inquiry approaches under a minimal-guidance critique, a framing contested by researchers whose implementations include extensive scaffolding. Its strongest implication is conditional rather than a rejection of active learning: instructional activity should be matched to learner expertise, and authentic problem solving needs deliberate guidance, feedback, and knowledge-building support rather than assuming learners will discover essential principles unaided.
Kluger and DeNisi review the feedback literature, meta-analyse 131 eligible studies, and propose Feedback Intervention Theory to explain highly variable effects. Feedback interventions improve performance on average, but more than one third of observed effects are negative.
The theory holds that feedback changes where attention is directed within a hierarchy of control. Information focused on task details, processes, or discrepancies that the learner can act upon is more likely to support performance. Feedback that shifts attention upward toward the self—through praise, threat, evaluation, or ego involvement—can consume resources, provoke affect, and reduce task-focused regulation. Effects also depend on goal clarity, task complexity, perceived control, and whether feedback supplies useful cues or merely signals success and failure. More feedback is therefore not automatically better.
The synthesis includes heterogeneous tasks and intervention types, and the authors’ explanatory model is preliminary rather than a simple recipe. Designers should keep attention close to the work, provide specific information for adjustment, avoid unnecessary self-comparison, and evaluate whether an intervention changes behaviour and learning rather than assuming that receiving results is beneficial.
View or buy the cited work >>Read a similar article free online >>
Knowles presents self-directed learning as a process in which learners take increasing initiative in diagnosing needs, formulating goals, locating resources, selecting strategies, and evaluating evidence of accomplishment. The guide has three parts for learners, teachers, and learning resources.
Inquiry projects help readers examine assumptions about self-direction, identify competencies, and construct learning plans or contracts. For teachers, the role shifts from content transmitter toward facilitator, procedural guide, consultant, and resource who creates trust and supports diagnosis and evaluation. Practical materials include relationship-building exercises, consultation practice, self-assessment tools, guidance for writing objectives, questioning techniques, proactive reading, use of people as resources, and ways to match evidence with objectives. The book is a practitioner guide grounded in adult education, not a controlled study showing that unguided independence improves every learner or task. Self-direction still involves structure, feedback, expertise, and negotiated accountability.
Educators can make goals and evidence explicit, offer meaningful choice, teach planning and resource-seeking, and gradually transfer responsibility. Readiness, knowledge, confidence, and context should shape guidance rather than treating autonomy as all-or-nothing.
View or buy the cited work >>Read a similar article free online >>
Kools and Stoll synthesize organizational-learning research into an integrated model of the school as a learning organisation. Such a school develops and shares a vision centered on every student's learning; creates continuous learning opportunities for all staff; promotes team learning and collaboration; establishes inquiry, innovation, and exploration; builds systems for collecting and exchanging knowledge; learns with and from its external environment; and grows leadership that models learning.
The seven dimensions are mutually reinforcing rather than a checklist of isolated initiatives. Professional learning should be sustained, connected to student needs and school goals, and supported by trust, reflection, experimentation, and access to evidence. Leadership is distributed while remaining responsible for coherent direction and conditions. Feedback loops help staff examine whether changes improve practice and outcomes, and partnerships prevent inward-looking improvement.
The authors operationalize each dimension with observable characteristics to support diagnosis and development, while warning that context matters. Becoming a learning organisation is an ongoing adaptive process that aligns individual, team, and whole-school learning around equitable student success.
This longitudinal study examines Finnish students’ digital fluency before and during the COVID-19 shift to distance and hybrid learning. Two Helsinki cohorts were followed from grades 5 to 6 and grades 7 to 8, with analyzed samples of 947 and 771 students.
The Sociodigital Practices Inventory measured perceived school practices, academic, artistic, and technical digital competences, and a sociodigital mindset combining digital efficacy with engagement. Changes were mixed rather than uniformly positive: primary pupils reported more basic school practices, while middle-school students perceived fewer advanced practices. Self-rated academic competences increased for girls and boys, whereas artistic and technical competences declined. Primary-school boys’ mindset strengthened, but middle-school girls’ mindset weakened. Latent-profile analysis identified Inexperienced, Enthusiastic, Humble, and Driven groups, demonstrating that skill and willingness do not always coincide.
The authors argue that access and frequent technology use are insufficient proxies for digital fluency. Schools need structured opportunities for creative, academic, collaborative, and technically challenging work, along with support for confidence and engagement tailored to differing student profiles.
Koriat and Bjork examine why learners can feel more competent during study than later tests justify. Judgements of learning are often made while both a cue and its target are visible, whereas testing presents the cue alone and requires target retrieval.
Features that make the pair seem strongly related in the presence of the answer can inflate confidence even when the cue is weak at eliciting that answer independently. Across paired-associate experiments, participants’ predictions were influenced by the apparent association from cue to target and failed to discount privileged access to the target during study. This creates an illusion of competence: fluent processing and perceived relatedness are mistaken for retrievability. The authors distinguish a priori cue-to-target accessibility from a posteriori coherence experienced after the answer is shown.
More diagnostic monitoring requires recreating test conditions, for example by delaying judgements, hiding the target, attempting retrieval, and then checking the response. The findings concern specific laboratory tasks but explain a broad study error: rereading with answers present can feel convincing while providing poor evidence that knowledge can be produced later without support.
Kornell, Hays, and Bjork ask whether attempting retrieval before receiving an answer helps or harms learning when the attempt is certain to fail. Participants saw unfamiliar or fictional general-knowledge questions, tried to generate answers, and then studied the correct responses.
Later memory was better than after studying intact question-and-answer pairs without the prior attempt. Experiments ruled out several simple explanations and showed that an unsuccessful search can prepare the learner to encode feedback more effectively. A question may activate related knowledge, focus attention on the relevant gap, create curiosity, and make the eventual answer more distinctive. The result does not justify prolonged guessing without correction. Benefits depend on prompt access to accurate feedback and a task in which the failed attempt engages relevant processing rather than reinforcing an attractive misconception.
Errorless study can feel efficient, yet it may produce weaker encoding than a brief challenge followed by resolution. The practical pattern is pretesting or prediction with low stakes: ask learners to commit to an answer, then reveal and explain the correct response and revisit it through later retrieval.
Korthagen critiques professional development that assumes teachers simply receive theory, understand it, and transfer it into practice. Teacher learning is often implicit, emotionally charged, motivational, social, and embedded in immediate classroom situations.
Behaviour is influenced at several interconnected levels: environment, behaviour, competencies, beliefs, professional identity, and personal mission or core qualities. Approaches that address only techniques or propositional knowledge may therefore leave deeper patterns unchanged. “Professional development 3.0” connects experience with systematic reflection, supports awareness of thoughts, feelings, wants, and behaviour, and helps teachers identify strengths and ideals as resources for action. The teacher as a person is not an obstacle to be standardised away but part of the mechanism of sustainable change. Korthagen draws on research and programmes to argue for coaching, collaborative reflection, small experiments in practice, and alignment across levels.
The article is a critical synthesis rather than a comparative meta-analysis, and its categories should guide inquiry rather than diagnose teachers mechanically. Effective development links theory and practice through repeated, supported learning grounded in authentic concerns and agency.
Kouhia, Kangas, and Kokko analyse Finnish craft educators’ experiences during emergency remote teaching in 2020. Data came from groupwork documents produced in two webinars involving educators across basic, vocational, higher, liberal adult, and arts education, and were examined through qualitative content analysis.
Remote craft pedagogy created opportunities to connect making with homes, families, everyday materials, local environments, and digital documentation. Teachers developed demonstrations, individual feedback, flexible assignments, and practices that could remain useful beyond the pandemic. Substantial difficulties arose from unequal access to tools and materials, safety concerns, limited technical and social resources, and the challenge of teaching embodied material skills at a distance. Interaction became more teacher-centred and task-oriented; one-to-one support could increase, while spontaneous peer exchange, shared making, and awareness of students’ processes were harder to sustain.
The study captures educator reports from an exceptional period rather than comparative learning outcomes. Its practical lesson is that remote craft education must design explicitly for material access, safety, demonstration, feedback, peer interaction, and the productive intersection of digital and hands-on activity.
Kozma reframes the media-effects debate from asking whether media independently cause learning to asking how media capabilities and instructional methods interact with learners’ cognitive and social processes. Media differ in symbol systems, processing capabilities, and the ways they can represent, transform, store, and connect information.
These capabilities become educationally consequential only when a design recruits them for meaningful activity. Kozma illustrates the argument with computer and video environments in which learners manipulate dynamic representations, coordinate multiple forms of information, discuss interpretations, and build knowledge. The same hardware can support very different processes, so a comparison that treats “computer” or “video” as a uniform treatment obscures the mechanism. His position answers Clark’s claim that methods, not delivery vehicles, determine learning, while avoiding simple technological determinism. Media do not guarantee outcomes and should not be evaluated by novelty alone.
Research should specify the representational affordance, task, learner knowledge, social organisation, and cognitive process that link a medium with an outcome. Design should begin with learning goals and deliberately exploit capabilities that make otherwise difficult thinking or collaboration possible.
Krishnamurti argues that education should help people understand life as a whole, not merely accumulate information, pass examinations, or become efficient workers. Technical knowledge has a necessary place, but without self-knowledge and intelligence it can intensify competition, fear, nationalism, and conflict.
He distinguishes intelligence from cleverness or memory: intelligence involves perceiving what is essential, understanding one’s conditioning, and meeting experience without rigid ideology. The “right kind” of education therefore seeks integrated human beings whose minds and emotions are not divided. It should cultivate freedom from fear, inquiry, sensitivity, creativity, relationship, and responsibility rather than conformity to authority or reward. Teachers cannot accomplish this through fixed systems alone; what educators are and how they relate to students matter deeply.
Small, caring environments and teachers engaged in their own self-understanding are central to his vision. The book connects personal transformation with social peace, maintaining that education reproduces disorder when it trains ambition and division, but can contribute to a different culture when it awakens intelligence, love, and direct understanding.
View or buy the cited work >>Read a similar article free online >>
Kuhlmann, Bernacki, and Greene connect cognitive theory of multimedia learning with the demands of self-regulated learning in computer-based higher education. Students must define tasks, plan, select strategies, monitor progress, and reflect, all of which consume limited mental resources.
The authors argue that well-designed multimedia can reduce unnecessary processing and leave more capacity for these regulatory activities. They illustrate how the multimedia principle can coordinate words and relevant graphics, personalization can make explanations more conversational, and generative activities can prompt learners to select, organize, and integrate content. These principles are presented as design supports, not replacements for students’ agency or explicit self-regulation instruction. The chapter is conceptual and practice-oriented rather than a comparative intervention study, so it does not establish that every multimedia feature improves regulation.
Its practical contribution is a useful design test: remove avoidable cognitive burden, align representations with the learning goal, and include activities that require meaningful processing so students can devote attention to both understanding content and managing their learning.
Lan and Zhou qualitatively synthesise research on artificial-intelligence applications supporting self-regulated learning in higher education. They organise findings around phases and functions of regulation, including goal setting and planning, performance and monitoring, feedback and strategy adjustment, and reflection.
AI systems can collect and visualise learning data, recommend resources or pathways, provide adaptive prompts and feedback, support metacognitive awareness, and help learners make decisions. The review distinguishes human-centred regulation, in which learners retain access and control, from designs that risk outsourcing judgement to opaque automation. Evidence is uneven: studies concentrate on selected phases, tools, and short-term outcomes, while emotional and motivational regulation, long-term development, diverse populations, and transfer receive less attention. Challenges include privacy, bias, explainability, data quality, overreliance, learner agency, and whether recommendations genuinely improve regulation rather than compliance.
Because included interventions and methods are heterogeneous, the review does not establish a single effect size or prove that AI causes better learning. It calls for theory-grounded, longitudinal, and human-centred design in which AI augments learners’ awareness and choice and teachers remain responsible for pedagogical interpretation.
This open, peer-reviewed handbook surveys learning analytics as a field concerned with collecting and interpreting data to understand and improve learning.
Its chapters introduce definitions and foundations, measurement, predictive modelling, network, language, multimodal, and temporal methods, and applications to self-regulation, collaboration, writing, discourse, assessment, and institutional decision-making. Later sections address implementation at scale, data literacy, fairness and bias, human-centred feedback, policy, privacy, and global K–12 perspectives. Across these topics, contributors repeatedly connect analytic techniques to learning theory, educational purposes, stakeholder interpretation, and responsible action rather than treating prediction as an end in itself. The volume is a field-level synthesis written by many authors, not evidence that every analytics intervention improves outcomes.
For educators and leaders, its key contribution is a map of design choices and risks: clarify the learning problem, select valid indicators, involve intended users, interpret data in context, evaluate consequences, and attend to equity, agency, and governance when analytics are deployed.
Lavonen analyses how Finland pursued twenty-first-century competences through parallel national-curriculum and teacher-education reforms in a decentralised system. The curriculum combines subject knowledge with transversal competences such as thinking and learning to learn, multiliteracy, digital competence, participation, working-life capabilities, cultural competence, and sustainable futures.
Municipalities, schools, and teachers interpret national goals locally, reflecting high professional autonomy rather than scripted implementation. Reform design involved teachers, teacher educators, ministries, municipalities, unions, students, and principals through collaboration, consultation, pilots, networks, seminars, and local support. A national teacher-education development programme similarly sought coherent research-based preparation, continuous professional learning, collaborative cultures, and capacity for curriculum enactment. Lavonen presents evaluations indicating progress while recognising that policy texts do not automatically change classroom practice. Coherence depends on shared understanding, local leadership, resources, professional agency, and sustained networks between universities and schools.
The chapter is a country case and policy analysis, not causal evidence that a particular reform raises achievement. Its broader lesson is that ambitious competence goals require aligned curriculum, assessment, teacher education, stakeholder participation, and implementation support over time.
Lee and Moore systematically review ten peer-reviewed empirical studies published from 2019 through 2023 that used generative AI for automated feedback in higher education. Systems operated across several instructional contexts and purposes, producing written guidance, conversational responses, cognitive support, emotional encouragement, and feedback linked to assessment or learning activity.
Reported possibilities include faster and more personalised responses, greater accessibility, lower anxiety when seeking help, and reduced instructor effort on routine feedback, potentially freeing time for complex teaching. The small and heterogeneous evidence base limits strong conclusions, however. Tools, disciplines, outcomes, and study designs varied, and rapid technological change means many systems predate current large language models. Generated feedback can be inaccurate, generic, biased, difficult to explain, or misaligned with learning goals. Students may overtrust it, and automation can weaken dialogue or instructor awareness.
The authors therefore frame GenAI as augmentation rather than replacement. Effective implementation requires instructors to design criteria and prompts, monitor quality, teach feedback literacy, preserve human interaction, and study how learners interpret, verify, and act on advice over time.
A transparent research trail
Each entry identifies the volume, edition, chapter or appendix in which the work appears.
Evidence summaries explain the central idea and why the source matters to educators.
Entries link to lawful full text where available, or to a publisher, library or purchase page.
References are checked against publisher records, scholarly indexes, DOI or ISBN metadata, repositories and author records where available. An entry marked [Citation not verified] preserves the wording found in the relevant volume without attributing findings to an unconfirmed work.
From sources to a coherent pathway