Where it is used
Each entry identifies the volume, edition, chapter or appendix in which the work appears.
The evidence behind the series
Follow the research.
Explore every source.
Browse the scientific research and professional literature used across all seven volumes of The Science of Learning for Educators.
7book volumes
8author ranges
1searchable library
Search the collection
Find a paper, author or idea.
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.
4 references
-
- Vol 7 RE: App D
- Vol 7 RE: Chapters 5, 6, 7, 11, 13, 14, App B
large language modelsartificial intelligence literacygenerated content verificationalgorithmic biasdata privacyassessment redesignhuman oversightJournal article Evidence summaryAt a glance
Kasneci and colleagues survey opportunities and challenges created by large language models in education shortly after ChatGPT’s public release. Potential uses include explanations, examples, dialogue, feedback, language support, writing assistance, assessment development, programming help, and teacher preparation of materials.
Key ideas
These systems may broaden access and personalise interaction, but fluent output is not reliable evidence of truth or understanding. Models can fabricate information, reproduce bias, expose private data, obscure sources, enable plagiarism, and encourage overreliance or reduced human judgement. Effects differ by learner knowledge: novices may be least able to detect plausible errors. The authors discuss implications for students, teachers, researchers, and institutions and argue for AI literacy, verification, transparent policies, redesigned assessment, and human oversight. LLMs should augment rather than replace educators or the social purposes of learning.
Why it matters
The article is a multidisciplinary position and review, not an evaluation demonstrating learning gains from ChatGPT. Its durable contribution is a balanced agenda: explore pedagogically grounded uses while studying accuracy, equity, privacy, agency, assessment validity, and the competencies needed to question and responsibly use generated content.
-
- Vol 4 RE: App A, App C
- Vol 4 RE: Chapters 25, 27
artificial intelligenceself-regulated learningmetacognitive awarenessadaptive feedbacklearner agencyhuman-centred regulationlearning analyticsalgorithmic biasJournal article Evidence summaryAt a glance
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.
Key ideas
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.
Why it matters
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.
-
- Vol 5 RE: Chapter 15
generative artificial intelligenceautomated feedbackhigher educationfeedback literacylarge language modelspersonalised feedbackinstructor oversightalgorithmic biasJournal article Evidence summaryAt a glance
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.
Key ideas
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.
Why it matters
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.
-
- Vol 3 RE: Chapters 5, 9, 13, 14, 15, 16, App D
artificial intelligence literacymachine learninglearning analyticsadaptive feedbackhuman–ai collaborationalgorithmic biasdata privacyethical ai adoptionBook Evidence summaryAt a glance
This practical introduction helps teachers and school leaders understand artificial intelligence well enough to make deliberate educational decisions without becoming technical specialists.
Key ideas
The authors distinguish AI from ordinary automation, explain the role of data and machine learning, and ask educators to begin with valued learning goals rather than available products. Schools are encouraged to become “AI ready” by identifying genuine educational challenges, examining what data are collected and what those data represent, and deciding which tasks require human judgment, relationships, and contextual knowledge. Examples show how AI can support analysis, feedback, personalization, and planning, while also exposing limitations involving bias, privacy, transparency, accountability, and unequal access. The book treats effective adoption as an organizational learning process: educators need shared language, evidence-informed experimentation, ethical review, and a strategy for combining artificial and human intelligence.
Why it matters
Its central message is that AI should expand educational opportunity and strengthen professional capacity, with teachers retaining responsibility for defining success and judging whether a system actually serves learners.
A transparent research trail
More than a bibliography.
What it contributes
Evidence summaries explain the central idea and why the source matters to educators.
How to access it
Entries link to lawful full text where available, or to a publisher, library or purchase page.
Edition abbreviations
- TE
- Teacher Edition
- PE
- Practice Edition
- RE
- Research Edition
Citation checks
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