Research figure: B. F. Skinner · Written by Artificial Sapiens
B. F. Skinner's The Technology of Teaching is a 1968 book in which the American psychologist Burrhus Frederic Skinner brought together his most influential writings on the experimental analysis of learning and the design of instruction. Its central argument is that educational achievement depends in substantial part on the contingencies educators arrange: the tasks students perform, the responses those tasks require, the feedback those responses produce, and the progression by which new repertoires are established. The book treats teaching as an activity whose procedures can be examined, improved, and evaluated through observable learning outcomes. Skinner (1968); B. F. Skinner Foundation.
Published by Appleton-Century-Crofts in 1968, the original volume contained 271 pages. It is a collection of related essays rather than a single experiment or a modern instructional-effectiveness trial. Its chapters move from the meaning of teaching and the science of learning through teaching machines, teacher preparation, motivation, thinking, creativity, discipline, and the behavior of educational institutions. The historical question is what Skinner actually argued in this book; the contemporary scientific question is how those arguments compare with later evidence about instruction and learning. Google Books bibliographic record; Google Books contents.
The Book, Its Publication, and Its Historical Setting
The Technology of Teaching belongs to Skinner's mature educational work, following his laboratory investigations of operant behavior and his increasingly explicit efforts to apply behavioral principles to human activities. In The Behavior of Organisms (1938), Skinner had developed an experimental analysis of response–consequence relations. Science and Human Behavior (1953) extended behavioral interpretation to human institutions and practices. By the 1950s and 1960s, he regarded conventional schooling as a setting where instruction had advanced far less systematically than the experimental science of learning. His educational writings therefore addressed a practical institutional problem: how to arrange opportunities to learn effectively when one teacher is responsible for many students with different repertoires and different rates of progress. B. F. Skinner Foundation bibliography; Skinner Foundation, The Technology of Teaching.
A visit to his daughter's mathematics classroom in 1952 became an important episode in Skinner's account of this problem. The Foundation's historical description reports his concern that pupils were answering at different rates while the teacher could offer feedback to only one child at a time. Skinner envisaged instructional arrangements that could present each student with suitable material, require an active answer, and supply a consequence promptly. This episode helps explain his subsequent interest in mechanical devices, although the history of mechanical testing and self-instruction already extended well beyond his own inventions. B. F. Skinner Foundation; Watters, Teaching Machines.
The 1968 publication assembled material with different earlier origins. “The Science of Learning and the Art of Teaching” appeared in Harvard Educational Review in 1954; “Teaching Machines” appeared in Science in 1958; an essay titled “The Technology of Teaching” had appeared in Proceedings of the Royal Society B in 1965. “Why Teachers Fail” also circulated as a 1965 essay. The 1968 book must therefore be interpreted as a synthesis of arguments developed across more than a decade, with its title identifying the collected educational program. Reprints, including a 2003 Foundation edition, have their own bibliographic dates and do not change the first publication date. Skinner Foundation full bibliography; Foundation reprint information; Skinner (1958).
What Skinner Meant by a Technology of Teaching
For Skinner, a technology of teaching meant the deliberate arrangement of learning conditions on the basis of experimentally studied relations between behavior and its consequences. Its starting unit was an educationally relevant response in a particular situation. Its procedures included establishing appropriate prerequisites, presenting discriminable tasks, arranging consequences for responses, revising the sequence when errors occurred, and observing whether the desired performance developed. The term technology referred to an organized application of knowledge about behavior, extending beyond the physical machinery that could present instructional material. Skinner (1968).
The basic distinction is between a teacher's presentation of material and a student's acquisition of a repertoire. Explanations, demonstrations, and assignments may all contribute, but their mere occurrence does not establish that the learner can solve the intended problems independently. In Skinner's analysis, success is expressed through changed performances under the conditions that matter. This approach invited educators to specify what pupils should be able to do and to design instruction backward from that performance. Its most productive contemporary interpretation is as a demand for explicit instructional goals and evidence of learning, while recognizing that worthwhile goals also include transfer, reasoning, collaboration, and creativity. Skinner (1968); Education Endowment Foundation feedback guidance.
Skinner's radical behaviorism provided the philosophical framework. He explained learning through functional relations among environmental events, an organism's history, and behavior; he also held that thoughts and feelings were legitimate events for behavioral interpretation. In the educational book his emphasis was practical: instructors can change prompts, feedback, opportunities for practice, sequence, and the social conditions of a classroom. These are variables that institutions can examine directly. The approach is stronger when the behavioral description is treated as one level of analysis within a larger science of learning rather than as a complete account of every cognitive operation involved in comprehension. B. F. Skinner Foundation bibliography; Dunlosky et al. (2013).
Reading the Structure of the 1968 Volume
The table of contents records eleven substantive chapter titles: “The Etymology of Teaching,” “The Science of Learning and the Art of Teaching,” “Teaching Machines,” “The Technology of Teaching,” “Why Teachers Fail,” “Teaching Thinking,” “The Motivation of the Student,” “The Creative Student,” “Discipline, Ethical Behavior, and Self-Control,” “A Review of Teaching,” and “The Behavior of the Establishment.” This structure matters because the book develops a broader philosophy and institutional program than the commonly reproduced story of a teaching machine. Published table of contents.
From the meaning of teaching to a science of learning
The opening chapters ask what it means to teach and why education needs procedures linked to learning. Skinner's language redirects attention from the activity of instruction to the change in a student's behavior that instruction should make possible. His 1954 essay frames the gap between a laboratory science with reliable operant phenomena and classrooms organized around lectures, intermittent correction, examinations, and threats of failure. Its historical claim is a programmatic argument about the organization of schooling, not an experimental estimate of how much every school could improve through a specific machine. Skinner bibliography, 1954; ERIC book record.
Teaching machines and the technology of instruction
The chapters on teaching machines and teaching technology translate the program into concrete arrangements. In a representative programmed lesson, material is organized into successive frames, each invites a response, and the learner receives information about that response before proceeding. Small steps, a high proportion of successful attempts, active responding, and individual progression are intended to produce orderly cumulative acquisition. Mechanical equipment could make presentation and feedback more reliable and less dependent on the teacher's simultaneous attention to a large group. Skinner's 1958 Science article is an essential primary source for his explicit machine proposals; the 1968 chapter examines their place within a larger theory of teaching. Skinner (1958); Skinner Foundation.
The machine itself was neither the original source of all programmed learning nor the only possible implementation. Sidney L. Pressey had constructed earlier automated educational devices, and later educators created branching programs and other forms of individualized instruction. Audrey Watters's historical study traces Pressey's work, Skinner's devices, commercial development, educational labor, and the recurring promises of machine-mediated personalization. Skinner's contribution is best located in his distinctive behavioral rationale and the engineering of instructional contingencies, with independent priority retained for earlier pioneers and subsequent innovations. Watters (2021).
Why teachers fail: the problem of conditions, preparation, and institutions
“Why Teachers Fail” shifts explanatory responsibility toward educational arrangements. In Skinner's formulation, teachers often receive limited training in experimentally informed instruction while institutions rely on delayed examinations, weak opportunities for correction, and aversive pressure to secure compliance. The point is a diagnosis of contingencies within schools, including the conditions governing teachers' own work. It does not amount to an empirical finding that individual teachers lack skill or care. The book's closing chapter, “The Behavior of the Establishment,” widens the perspective to organizations that maintain established practices. Skinner (1968).
This institutional argument remains fertile because a successful instructional method depends on curricula, workloads, technology procurement, teacher preparation, learner access, and the criteria used to judge success. An isolated device can present excellent feedback and still fail as an educational intervention if pupils cannot use it, the curriculum asks inadequate questions, or teachers lack resources to interpret its results. Contemporary implementation research and feedback guidance similarly emphasize the environment in which a practice is used. EEF (2021).
Thinking, motivation, creativity, discipline, and self-control
The second half of the book treats complex repertoires as instructional subjects. “Teaching Thinking” asks how problem-solving actions can be acquired and occasioned; “The Creative Student” discusses the conditions through which useful novel behavior emerges; “The Motivation of the Student” examines the consequences and circumstances that sustain study. “Discipline, Ethical Behavior, and Self-Control” applies the analysis to classroom conduct and the learner's capacity to arrange conditions affecting future action. The breadth of these chapters corrects the recurring caricature that Skinner meant education to consist only of rote memorization. Contents of The Technology of Teaching; ERIC overview.
These discussions are best understood as Skinner's theoretical interpretation of complicated human performances. Their evidential status varies: laboratory findings on reinforcement and stimulus control provide foundations, but a proposed instructional account of creativity or ethical conduct requires its own outcome measures, comparison conditions, and social evaluation. Naming a cognitive or ethical achievement as a repertoire does not by itself establish that a particular sequence of reinforcement produces that achievement. Dunlosky et al. (2013); Hattie and Timperley (2007).
The Learning Mechanism: Responses, Consequences, and Instructional Sequences
In the operant analysis, a response is affected by the consequences that follow it under particular conditions. Reinforcement is defined through its effect on the future probability or rate of behavior, rather than solely through a teacher's intention to praise or reward. A discriminative stimulus signals a context in which a response has a particular history of consequences. Shaping establishes a target repertoire through successive approximations, and stimulus control allows a learned performance to come under the influence of relevant features of a problem. In teaching, these concepts yield a systematic question: what must the student already do, what new response is required, and what response-dependent information or consequence will support the next step? Skinner's original educational account; B. F. Skinner Foundation full bibliography.
A student learning an equation illustrates the design problem. A beginning task may require identifying a missing value in a simple relation; later tasks may ask the student to transform expressions, discriminate which operation preserves equality, explain a solution, and solve unfamiliar problems. A coherent program connects the exercises to those later performances, examines errors, and provides corrective information when needed. The example is an analytical reconstruction of an instructional sequence, not a historical report of a particular Skinner classroom experiment. The central design principle is conditional progression based on demonstrated learning. Skinner (1958); EEF feedback guidance.
Why active responses matter
Programmed instruction requires learners to do something that can be evaluated. The response may be a written answer, a computation, a selected alternative, a demonstration, or an explanation. This emphasis opposes instructional models that equate exposure to information with learning. However, response production and conceptual understanding have different measurement requirements: repeatedly completing near-identical items can show proficiency on those items while leaving transfer, long-term retention, or deeper explanation uncertain. A contemporary evaluation therefore needs assessments beyond the original instructional frames. Dunlosky et al. (2013); Ma et al. (2014).
Immediate feedback and its boundaries
Skinner placed special emphasis on consequences available close to the response. Immediate confirmation can help a learner locate an error while the task is salient and continue the sequence without waiting for the teacher to return corrected work. Yet informational feedback, reinforcement, grades, and rewards are analytically distinct. Feedback conveys information about performance; reinforcement is defined by the effect of an event on future responding. A correct-answer display may serve an informational function even when its reinforcing effect has not been demonstrated. Modern feedback research also shows that content, timing, task complexity, learner knowledge, and opportunities to act on advice can change results. Hattie and Timperley (2007); Newman et al. (2021).
The lesson for contemporary designers is to make feedback specific enough to guide revision and proportionate to the task. A student solving an algebraic proof needs more than a celebratory mark for the final answer; the program should identify which inferential step succeeded or failed. For some complex problems, asking learners to diagnose their own mistakes, compare alternative solutions, or reflect after a delay may produce more transferable knowledge than constant instant confirmation. The choice is an empirical design question. EEF feedback toolkit.
Individual pacing and mastery
Individual pacing was Skinner's response to a structural mismatch between variable rates of acquisition and group instruction delivered on a shared schedule. A learner should be able to advance after reaching the relevant criterion and spend more time where prerequisites remain uncertain. This claim anticipates later mastery-based and adaptive systems in a broad historical sense, although those systems have distinct conceptual origins, designers, and evaluative records. Benjamin Bloom's later argument about the performance possible under individual tutoring and mastery arrangements provides an important subsequent research program, not an experiment conducted by Skinner. Bloom (1984); Watters (2021).
Mastery criteria also have to be defensible. An arbitrary threshold can encourage repetition without durable understanding; poorly calibrated adaptive sequencing can direct learners toward easy items and conceal persistent misconceptions. Robust instruction measures delayed retention and the ability to use learning in novel settings. Progression decisions should take account of accuracy, reasoning quality, fluency when relevant, and the cost of errors. These are questions of educational measurement in addition to behavioral engineering. Dunlosky et al. (2013).
Research Design and What the 1968 Book Actually Demonstrates
The Technology of Teaching is a synthesis and argument built on earlier experimental work and educational proposals. Its chapters are not a single randomized comparison of students assigned to programmed instruction or traditional teaching. A research claim based on this book must therefore identify whether it concerns an experimental law supported by the broader operant literature, an example of instructional engineering, a social criticism, or a forecast about educational technology. These are different classes of evidence. ERIC publication record; Skinner's 1958 paper.
Skinner's fundamental laboratory approach focused on repeated measurement of individual organisms and on changes in response rate under experimentally arranged conditions. Those methods were highly productive for identifying environmental control of simple operant repertoires. Moving from an experimentally controlled response to classroom achievement introduces additional variables: prior knowledge, language, motivation, teacher and peer relationships, socioeconomic resources, assessment reliability, cultural context, and the meaning of a school outcome. Controlled operant effects are an empirical foundation for instructional hypotheses, not an automatic proof that a specified school curriculum works. B. F. Skinner Foundation bibliography; Ma et al. (2014).
Evidence for a school or university implementation should therefore ask who was taught, in what setting, compared with what alternative, using which materials, over what period, and with what measurements. Important outcomes include learning gains relative to an appropriate comparator, retention after instruction, transfer to unfamiliar problems, engagement, differential effects across learner groups, teacher workload, implementation costs, and unintended effects. Random assignment where feasible, transparent analytic plans, replication, and carefully described materials strengthen causal inference. Direct classroom evidence deserves priority over claims that a procedure must work because it resembles laboratory reinforcement. Newman et al. (2021); EEF guidance.
Early Reception, Programmed Instruction, and Subsequent Evaluation
The reception of Skinner's educational program combined enthusiasm about self-paced instruction with practical difficulties and theoretical criticism. Programmed textbooks and devices appeared in an era of expanding enrollment and hopes that better instructional engineering could improve access and efficiency. Yet commercial distribution, the writing of high-quality instructional frames, maintenance, teacher integration, and sustained student interest proved consequential. Watters's book documents the complex historical path from the experimental machine to commercial products and institutional practices; it also restores earlier and parallel innovators to the history. Watters (2021).
In 1980, James A. Kulik, Peter A. Cohen, and Barbara J. Ebeling published a meta-analysis of programmed instruction in higher education. This is a landmark attempt to synthesize comparative evaluation rather than rely on examples offered by a technology's proponents. Its historical setting and the interventions included must be retained when applying its conclusions: evidence about particular programs used in earlier college settings cannot serve as an effect estimate for all modern adaptive learning platforms. Kulik, Cohen, and Ebeling (1980).
One enduring influence was the belief that teaching materials can be evaluated and revised through learner responses rather than judged only by the elegance of a lecture or the authority of a textbook. Another was the ambition to differentiate instructional sequences without requiring the teacher to deliver a separate lecture to each child. Neither development belongs exclusively to Skinner. Historians place him among a wider set of psychologists, educators, inventors, publishers, and organizations that shaped twentieth-century automated instruction. Watters (2021).
Comparison With Contemporary Cognitive and Educational Psychology
Memory, retrieval, spacing, and transfer
Modern cognitive research has expanded the description of what a learning design must accomplish. Dunlosky and colleagues' comprehensive review rated practice testing and distributed practice highly across varied learning conditions. These findings concern retrieval and spacing, which must be designed and measured directly rather than assumed to result from a sequence of small reinforced responses. A program can provide immediate answers while failing to ask students to retrieve information after meaningful delays. Conversely, a thoughtfully designed sequence can combine active responding with cumulative retrieval, spaced review, and challenging application. Dunlosky et al. (2013).
In addition, learners build conceptual models that can be assessed through explanation and transfer tasks. A school exercise may produce correct short answers through familiar prompts while offering weak evidence of abstract understanding. Cognitive load, background knowledge, linguistic demands, metacognitive regulation, and social discussion influence how students use an instructional sequence. In contemporary practice the most productive dialogue with Skinner is about how measurable performance, internal knowledge representations, and learning conditions can be investigated together. Hattie and Timperley (2007); Dunlosky et al. (2013).
Feedback as information, regulation, and action
Hattie and Timperley's 2007 review characterizes feedback as influential but highly variable. It may target a task, the process of solving it, or the learner's self-regulation; feedback directed at personal traits is often less useful. The Education Endowment Foundation's systematic review and guidance likewise find an important role for feedback while stressing the quality of instruction, actionable content, and the conditions under which pupils can use it. These literatures qualify any interpretation of Skinner that reduces effective teaching to maximizing immediate praise or the frequency of correctness signals. Hattie and Timperley (2007); Newman et al. (2021); EEF toolkit.
The modern instructional design question is how feedback changes later performance. Feedback is only useful to the extent that the learner can understand and employ it, and the resulting change must be measured. This formulation preserves Skinner's concern with consequences and learning outcomes while admitting explanations involving beliefs, goals, metacognitive strategies, and representations. EEF teacher guidance.
Motivation beyond externally delivered consequences
Skinner's educational proposal placed special weight on arranging effective contingencies and reducing aversive instructional control. Contemporary accounts also investigate interest, values, competence, autonomy, social belonging, and how different tasks and rewards affect engagement. These considerations make the exact operationalization of motivation crucial. Increased task completion may be useful but is an incomplete measure of curiosity, intellectual confidence, and willingness to learn independently. Skinner's opposition to punitive schooling remains a historically important concern; assessing positive learning climates requires evidence about student experience as well as observable performance. Skinner (1968); EEF feedback guidance.
The Connection to Intelligent Tutoring and Digital Learning
Digital platforms can implement components recognizable in Skinner's design: presenting a task, recording responses, adapting the next item, and returning feedback. They can also use student models, probabilistic inferences, knowledge tracing, simulations, dialogue, and curriculum structures that were unavailable to the mechanical devices of the 1950s. Historical continuity lies in some instructional problems and design affordances, not in a claim that every intelligent tutor is an implementation of radical behaviorism. Watters (2021); Ma et al. (2014).
A 2014 meta-analysis by Ma, Adesope, Nesbit, and Liu examined intelligent tutoring systems across 107 effect sizes and 14,321 participants. It reported positive average achievement differences relative to several comparison conditions, while finding no statistically significant difference against individual human tutoring or small-group instruction in its analyses. The results concern a heterogeneous family of software and comparators and should not be represented as a direct randomized test of Skinner's 1968 proposals. They do, however, support the broader feasibility of well-designed individualized computer-supported instruction under certain conditions. Ma et al. (2014).
Cavalcanti and colleagues' 2021 systematic review of 63 studies on automatic feedback in online environments highlights both reported benefits and limitations in how educational feedback theory has informed technological systems. Their synthesis warns against assuming that automation necessarily reduces teacher workload or that a response-scoring algorithm provides meaningful guidance. Liao, Zhu, and Wang's 2024 meta-analysis of human–machine feedback reported an average positive effect on learning outcomes across its included comparisons, with significant variation associated with feedback design. The pooled result is an estimate for its specified research corpus, not a universal effect of digital instruction. Cavalcanti et al. (2021); Liao et al. (2024).
For Artificial Era educational technologies, the scientific challenge is to preserve evidence-based goals and human educational judgment while making adaptation interpretable. A system should support teachers in knowing why a learner received a certain activity, what evidence justifies the next step, what the algorithm may have missed, and how its recommendations can be corrected. New generative systems add risks of incorrect explanations, inconsistent feedback, data leakage, and misleading confidence. These require direct evaluation of system accuracy, instructional outcomes, accessibility, and governance rather than analogies to early teaching machines. Cavalcanti et al. (2021); EEF guidance.
Ethical and Methodological Limits of Instructional Control
Education has purposes that extend beyond a locally measured response. Skinner's language of behavioral control can illuminate the real influence of curricula, sanctions, rewards, and classroom arrangements, while also prompting questions about who decides the target outcomes and whose interests those targets serve. Ethical design requires transparent educational purposes, appropriately justified evaluation, learner participation where feasible, accessibility, proportionate data collection, and independent review of harms. Student agency should be evaluated through opportunities to question, choose, reason, and eventually govern one's own learning. Skinner (1968); EEF guidance.
The distinction between positive reinforcement and bribery or coercion is empirical and contextual. A contingent reward can encourage a selected response, yet its value for education depends on whether it supports understanding, durable engagement, and worthwhile goals. A strict completion metric may mask guessing, dependency on prompts, or disengagement. A system trained on aggregate response data can also disadvantage learners whose communication style, disability, language, or prior access differs from the dominant training population. Outcome evaluation should examine these possible differential effects instead of relying only on an average gain. Newman et al. (2021); EEF toolkit.
Historical educational behaviorism and present-day clinical or special-education services must also be distinguished. A claim that reinforcement contingencies can affect learning is not a clinical efficacy estimate, a diagnosis, or a treatment recommendation. Modern behavior-analytic applications and their professional and ethical standards were developed by many subsequent researchers, including Donald Baer, Montrose Wolf, and Todd Risley, whose 1968 foundational work in applied behavior analysis belongs to a distinct applied-research history. The Technology of Teaching is best evaluated here as a text in educational psychology and the history of instructional design. Skinner (1968); B. F. Skinner Foundation bibliography.
What Remains Scientifically Valuable
The book's durable contribution is a set of researchable educational design questions. Are goals stated as performances that matter? Does each learning activity invite a meaningful response? Is feedback accurate and usable? Does the curriculum distinguish prerequisites from advanced applications? Can students progress at different rates without losing access to teachers and peers? Are retention and transfer measured rather than presumed? Can teachers identify weaknesses in a sequence and improve it using evidence? These questions persist across print, classroom conversation, mechanical instruction, adaptive software, and contemporary intelligent tutors. Skinner Foundation; Hattie and Timperley (2007); Ma et al. (2014).
Established scientific findings support the value of many well-designed feedback practices, active retrieval, distributed practice, and some forms of individualized educational software. Evidence is more conditional when one asks whether a particular historical Skinnerian program, automated platform, reward schedule, or classroom policy outperforms an alternative for a specified population and outcome. Contested issues concern the explanatory sufficiency of behavior-only accounts of complex cognition, the cultural aims of instructional control, and the institutional costs of technologically mediated education. Treating these levels separately gives the book a serious intellectual place without projecting twenty-first-century research results backward onto its author. Dunlosky et al. (2013); Newman et al. (2021); Watters (2021).
The most valuable reading of The Technology of Teaching is historical and analytical: it explains why Skinner believed a science of behavior should transform education, how he translated that belief into instructional design, and which questions his proposals left for experimental and democratic scrutiny. The book shows an ambitious attempt to connect laboratory inquiry, school practice, and institutional organization. Its continuing value depends on testing rather than merely repeating its prescriptions. Skinner (1968); Watters (2021).
Frequently Asked Questions
When was The Technology of Teaching first published?
The book first appeared in 1968 through Appleton-Century-Crofts. Its chapters include revised or reprinted essays that Skinner had published earlier, notably material from 1954, 1958, and 1965. The Foundation issued a later reprint in 2003. Those earlier essay dates and later reprint dates should be recorded separately from the book's original 1968 publication. Skinner Foundation books; Foundation book listing.
Is the book mainly about a teaching machine?
Teaching machines are central examples, but the book also treats the purpose of teaching, the scientific analysis of learning, student motivation, problem solving, creativity, discipline, self-control, teacher preparation, and educational institutions. Its eleven substantive chapters demonstrate the wider scope. Published contents.
Did Skinner invent the first teaching machine?
No. Mechanical testing and teaching devices, including Sidney Pressey's earlier work, preceded Skinner's own designs. Skinner made influential contributions to self-paced programmed instruction and to a behavioral account of how instructional contingencies could be arranged. The fuller chronology is documented by historian Audrey Watters. Watters (2021).
Did Skinner claim that teachers should be replaced?
His proposal was that machines could take over some repetitive presentation and correction tasks, leaving teachers more time for intellectual, cultural, and interpersonal work. This claim was part of the book's rationale and should be distinguished from empirical evidence about actual workload savings. Later studies of automated feedback show that the workload question remains an independent outcome to measure. Skinner (1968); Cavalcanti et al. (2021).
Does modern research prove Skinner's entire educational theory?
Modern evidence supports some component practices, including high-quality feedback and particular individualized learning technologies, under specified conditions. The 1968 volume contains broader philosophical and institutional arguments that cannot be established by a single effect estimate. Research on memory, motivation, cognition, classroom relationships, and social consequences also informs a complete educational evaluation. EEF toolkit; Dunlosky et al. (2013).
Related Articles
B. F. Skinner — The central scientific chapter on Skinner's life, experimental analysis of behavior, radical behaviorism, principal publications, and intellectual legacy.
Learning Theories in Education: Behaviorism, Cognitivism, Constructivism, and Sociocultural Learning — Compare the wider theoretical approaches that shape present-day educational research and instruction.
Behavior Change Techniques: What Research Supports — Examine broader evidence and boundaries for methods that seek to change behavior across settings.
B. F. Skinner's Teaching Machines and Programmed Instruction — Teaching machines and programmed instruction in education.
B. F. Skinner's Science and Human Behavior — The 1953 synthesis of behavioral science and social institutions.
B. F. Skinner's Influence on Applied Behavior Analysis — The history and later development of applied behavior analysis.
References
Bloom, B. S. (1984). The 2 sigma problem: The search for methods of group instruction as effective as one-to-one tutoring. Educational Researcher, 13(6), 4–16. https://doi.org/10.3102/0013189X013006004
Cavalcanti, A. P., Barbosa, A., Carvalho, R., Freitas, F., Tsai, Y.-S., Gašević, D., and Mello, R. F. (2021). Automatic feedback in online learning environments: A systematic literature review. Computers and Education: Artificial Intelligence, 2, 100027. https://doi.org/10.1016/j.caeai.2021.100027
Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., and Willingham, D. T. (2013). Improving students' learning with effective learning techniques: Promising directions from cognitive and educational psychology. Psychological Science in the Public Interest, 14(1), 4–58. https://doi.org/10.1177/1529100612453266
Education Endowment Foundation. (2021). Teacher feedback to improve pupil learning: Guidance report. https://educationendowmentfoundation.org.uk/education-evidence/guidance-reports/feedback
Hattie, J., and Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1), 81–112. https://doi.org/10.3102/003465430298487
Kulik, J. A., Cohen, P. A., and Ebeling, B. J. (1980). Effectiveness of programmed instruction in higher education: A meta-analysis of findings. Educational Evaluation and Policy Analysis, 2(6), 51–64. https://doi.org/10.3102/01623737002006051
Liao, M., Zhu, K., and Wang, G. (2024). Can human-machine feedback in a smart learning environment enhance learners' learning performance? A meta-analysis. Frontiers in Psychology, 14, 1288503. https://doi.org/10.3389/fpsyg.2023.1288503
Ma, W., Adesope, O. O., Nesbit, J. C., and Liu, Q. (2014). Intelligent tutoring systems and learning outcomes: A meta-analysis. Journal of Educational Psychology, 106(4), 901–918. https://doi.org/10.1037/a0037123
Newman, M., Kwan, I., Schucan Bird, K., and Hoo, H.-T. (2021). The impact of feedback on student attainment: A systematic review. Education Endowment Foundation. https://eric.ed.gov/?id=ED616009
Skinner, B. F. (1954). The science of learning and the art of teaching. Harvard Educational Review, 24, 86–97. https://www.bfskinner.org/publications/full-bibliography/
Skinner, B. F. (1958). Teaching machines. Science, 128(3330), 969–977. https://doi.org/10.1126/science.128.3330.969
Skinner, B. F. (1968). The technology of teaching. Appleton-Century-Crofts. Original edition; later reprinted by B. F. Skinner Foundation in 2003. https://www.bfskinner.org/product/the-technology-of-teaching/
Watters, A. (2021). Teaching machines: The history of personalized learning. MIT Press. https://doi.org/10.7551/mitpress/12262.001.0001