Research figure: B. F. Skinner · Written by Artificial Sapiens
B. F. Skinner's shaping through successive approximations is the selective development of an operant response through reinforcement of performances that progressively meet a changing behavioral criterion. An investigator starts from responses already present in the organism's repertoire, reinforces those meeting an attainable criterion, and advances that criterion toward a defined terminal performance. Shaping thus concerns the acquisition and differentiation of response topography, force, location, duration, or other measurable properties through consequences. It became one of Skinner's most influential methods for showing how new behavioral repertoires could emerge from an organism's ongoing variation. Peterson, 2004
The historical chronology is crucial. Skinner documented experimental response differentiation by successive approximation in The Behavior of Organisms in 1938. A memorable 1943 episode in Minneapolis marked his discovery of manual shaping's possibilities and transformed his interpretation of socially mediated reinforcement. His 1951 article How to Teach Animals subsequently introduced related techniques to a broad readership. These stages have different documentary and scientific significance; the 1943 event was a distinctive practical and conceptual breakthrough within an existing experimental program. Peterson, 2004 Skinner, 1951
Definition and Essential Mechanism
Shaping is a differential-reinforcement procedure in which the criterion that defines an eligible response is progressively altered. The current criterion may initially accept a remote approximation of the target. Once performances satisfying it become sufficiently likely, the investigator changes the criterion to select a more exact response. Over a series of adjustments, a novel or initially rare response becomes more prevalent. Successive approximations identify the sequence of selection requirements; the operative causal arrangement is the contingency relating responses and their consequences.
An approximation need not look subjectively similar to the final response. It has to meet an operationally specified criterion along a dimension relevant to acquiring that response. In spatial shaping, the criterion may identify the portion of a panel where a pigeon pecks. In a motor task it may concern movement amplitude, force, or duration. In speech training the dimension and its functional significance require special care: a recognizable sound, a socially appropriate request, and a linguistically meaningful utterance are different outcomes.
What Counts as Reinforcement?
Reinforcement is defined functionally from the behavior-consequence relation, rather than by the pleasantness of an event or the intentions of the trainer. Food, attention, access to materials, or task feedback may function differently for different individuals or conditions. To claim differential reinforcement, the investigator should document what happens after criterion-meeting responses and whether the subsequent distribution of responding changes. A reward supplied regardless of performance does not reproduce the same contingency.
Successful shaping also requires behavioral variation. Consequences select among existing or emerging responses; they do not mechanically manufacture identical new actions from nothing. If variability falls too low, a stricter criterion may yield very few eligible responses. The experimenter must then consider criterion size, motivation, relevant cues, environmental constraints, and the opportunities for the organism to contact reinforcement. Galbicka, 1994
Intellectual Foundations: Thorndike, Watson, and Skinner
Edward L. Thorndike's law of effect established an important earlier relation between an action's consequences and subsequent behavior. John B. Watson promoted an early program of behaviorism. Skinner developed his own experimental analysis of operant behavior, taking the rate and differentiated dimensions of responding as central objects of investigation. Historical credit therefore follows a sequence of distinct contributions: Thorndike's prior law of effect, Watson's earlier behaviorist program, and Skinner's systematic experimental procedures and explanatory framework. B. F. Skinner Foundation bibliography
Respondent and operant conditioning address different functional relations. Respondent research typically examines elicitation and relations between antecedent stimuli; operant research organizes consequences contingent on behavior and measures future responding. Shaping belongs to the operant tradition because the approximation is selected through a consequence contingency. Antecedent cues, respondent processes, motor affordances, and prior experience remain relevant to the behavior observed, but their possible influence does not erase the distinctive contingency being tested.
The 1938 Book: Experimental Response Differentiation
The Behavior of Organisms: An Experimental Analysis, first published in 1938, presented Skinner's developing methodology and experimental analysis. Its eighth chapter examined response differentiation through successive approximations. This is crucial primary historical evidence: the relevant experimental procedure was present in Skinner's published work before the famous wartime anecdote. Gail B. Peterson's 2004 historical study makes this distinction explicitly and reconstructs the later episode using Skinner's writings and the research history. Peterson, 2004
The earlier work included arrangements in which apparatus-controlled consequences selected measurable response properties. The operant laboratory sought orderly relations within individual organisms under controlled conditions, rather than relying exclusively on group averages or retrospective impressions. The analysis required clearly described responses, experimental manipulation of contingencies, and repeated observations. The strength of this approach was the capacity to reveal changes in behavior as contingencies changed. Its limits included the experimental environment's specific controls and the need to establish whether principles generalized beyond the task and species studied.
Response Differentiation and Hand Shaping
Mechanical response differentiation and manual shaping share differential selection but differ in execution. An apparatus can deliver a consequence when a response crosses a predefined threshold. A person manually shaping behavior recognizes an approximation, delivers a consequence, and often adjusts the criterion in real time. The latter method may incorporate contextual perception and social responsiveness, along with potential variations in judgment and timing. Peterson's history emphasizes that Skinner understood experimental differentiation before he appreciated the full possibilities of manually guiding new behavior. Peterson, 2004
The 1943 Minneapolis Breakthrough
In 1943, while working with pigeons in Minneapolis during wartime research, Skinner encountered a dramatic episode in which successive reinforcing consequences rapidly changed a bird's behavior. Peterson locates the incident on an upper floor of a flour mill and describes the experience as formative for Skinner's thinking. The observation helped him recognize that a trainer could build a new repertoire through moment-to-moment selection, even when no mechanical apparatus had been engineered to detect every approximation. Peterson, 2004
The significance extended beyond an appealing animal-training story. In hand shaping, another living organism mediates the consequences. Skinner came to appreciate the role of such mediation in social behavior, notably the relationships between speakers and listeners that later interested him in Verbal Behavior. That influence is a historical claim about the evolution of his research program. It does not prove that grammatical structure, language meaning, or all social learning can be reduced to a single shaping process. Peterson's reconstruction provides grounds for separating the genuine shift in Skinner's thought from exaggerated claims of conceptual invention in one afternoon. Peterson, 2004
Skinner's 1951 How to Teach Animals
Skinner published How to Teach Animals in the December 1951 issue of Scientific American. The article illustrated how techniques developed in experimental psychology could be used to establish novel animal performances and invited readers to consider the wider significance of consequences in learning. The original publication predates many contemporary popular descriptions that present shaping simply as breaking a task into small rewards. The scientific issue is more precise: the learner's current response distribution determines what approximations are available, while selective consequences change which responses recur. Skinner, 1951
A successful demonstration shows that a particular outcome is possible under its conditions. It generally does not estimate the average probability of success, isolate every causal variable, or show that its procedure outperforms alternatives. Public accounts and demonstrations must be interpreted in light of controlled experiments, including failures and parametric comparisons. This methodological distinction has become important in later assessments of the shaping literature. Orihara and Tanno, 2022
How an Experimental Shaping Procedure Is Constructed
Baseline and Terminal Response
A valid experiment first describes the baseline repertoire. The researcher records which actions already occur, their frequency, variability, relevant dimensions, and conditions. Next comes a terminal criterion, defined independently of the training sequence. The outcome might be pecks within a specified position range, a movement of sufficient extent, or a response lasting at least a stated duration. Without an independently identified target, the investigator can confuse movement toward a convenient intermediate response with acquisition of the behavior intended.
Criterion Progression
The experimenter specifies which performances qualify for reinforcement at each stage. A criterion may be fixed for a predetermined interval, moved after a set number of successful responses, or adjusted according to a formal rule based on recent response values. A criterion change modifies both the opportunities for reinforcement and the response values selected. The record should show exactly when a criterion moved, how far it moved, and the behavior observed immediately before and after the change.
The familiar advice to begin with easily attainable approximations has a defensible behavioral rationale: early contact with reinforcement can establish the contingency. Yet the proper size and rate of progression are empirical variables. Very small steps may consume more reinforcement and time than necessary; very large steps may leave few opportunities to succeed. A training protocol should respond to actual performance instead of relying on a universally optimal progression that research has not established. Eckerman et al., 1980
Example: Developing a Pigeon's Key Peck
Suppose a pigeon reliably moves near a response panel but seldom contacts a designated key. A shaping sequence could begin by reinforcing approach to the panel, then movement into the key's vicinity, then increasingly precise beak orientation, and finally actual pecks on the key. This is an illustrative teaching example, not a report of a particular Skinner experiment. The important scientific features are the operational criteria and the recorded contingencies. If beak orientation already occurs, the experimenter can select that response directly; if pecks are frequent, shaping from distant approximations may be unnecessary.
Experimental Parameters: Step Size and Progression Rate
David A. Eckerman, Robert D. Hienz, Sander Stern, and Vicki Kowlowitz reported a controlled experiment in 1980 on the location of pigeons' pecks. Several pigeons pecked within a ten-inch-wide response area. Grain was delivered for pecks at specified locations, and the reinforced location shifted systematically across the area. The researchers varied the size of those location shifts and the number of reinforcers supplied before shifting the criterion, using between-subject and within-subject comparisons. This made two ordinarily discretionary features of shaping subject to explicit experimental study. Eckerman et al., 1980
Peck locations changed under criterion shifts ranging from approximately half an inch to three inches. More rapid shifts were approximately as effective as slower ones across the tested range, and relatively large, rapid movements of the criterion could support efficient response differentiation. The findings qualify traditional instructions that shaping must always proceed through tiny, slow steps. They do not establish a universal large-step rule, because the organism, response dimension, apparatus, and available reinforcer all condition the result. Eckerman et al., 1980
Variables and Causal Interpretation
In the peck-location study, the independent variables included the spatial distance between successive reinforced locations and the rate of changing the criterion. The dependent outcome concerned the pigeon's actual distribution of peck locations. Repeated observations within subjects were central to evaluating changes over time. The experimental arrangement supports causal conclusions about those contingencies in that preparation while leaving open questions about other species, complex skills, motivational states, and different response dimensions. The study illustrates why claims about shaping should report parameter values rather than present trainer intuition as a general empirical law.
Timing of Reinforcement
The interval between an eligible response and its consequence can affect which behavior the learning environment selects. If other responses occur during a delay, the relation between the intended response and the consequence can become more ambiguous. Kennon A. Lattal's review of delayed reinforcement showed why timing must be interpreted together with discriminative signals, the rate and distribution of reinforcement, and other correlated environmental events. No single fixed temporal cutoff captures the full experimental literature. Lattal, 2010
Manual shaping introduces the observer's reaction time and classification decisions. One trainer may recognize the critical approximation immediately, while another may react after an additional movement. In an automated arrangement, the device can record measured thresholds and consequence latencies consistently, although its measurement model may omit important qualitative features. Investigators should record timing wherever it could explain differences in acquisition. The general lesson is to specify the contingent event the learner actually encounters, not only the trainer's intention.
Manual Judgment and Automation
Human trainers can adapt to diverse movements, subtle contextual differences, and changing repertoires. The flexibility that makes manual shaping useful creates a serious reproducibility problem. Researchers may omit when they changed criteria, how they classified borderline responses, and whether prompts or reinforcer magnitudes varied. Two practitioners using the phrase successive approximations may implement considerably different intervention histories. Orihara and Tanno, 2022
Automated shaping can measure position, force, timing, or movement trajectories at high resolution and maintain a complete record of thresholds and consequences. It can reduce some observer-related variation and support precise comparisons. An automated sensor still reflects choices about what dimensions to count and what the learner is supposed to accomplish. The methodological objective is transparent and appropriate measurement, whether an investigator, a device, or a combination governs the criterion.
Percentile Schedules as a Later Scientific Development
Gregory Galbicka's 1994 article proposed and analyzed percentile reinforcement schedules as a way to formalize response shaping. Under this approach, reinforcement criteria are derived from the learner's recent performance distribution according to an explicit percentile rule. A criterion can therefore advance as the observed distribution improves while maintaining a selected proportion of opportunities for reinforcement. This is a later methodological development in the operant tradition and should be credited to Galbicka and associated researchers, not represented as Skinner's original 1938 protocol. Galbicka, 1994
Percentile arrangements can clarify exactly why a response was eligible for reinforcement at a particular point. Their suitability depends on the measurement scale, response variability, size of the recent sample, and the chosen reinforcement proportion. Such schedules turn vague instructions to reward better attempts into algorithms capable of being tested and compared. They also highlight the difference between documenting that shaping sometimes works and specifying the conditions under which it is most efficient and reliable.
How Acquisition Should Be Measured
A useful shaping study reports response distributions, not only whether the terminal criterion was reached. Investigators can measure reinforcers to acquisition, session duration, response rate, trial-by-trial criterion changes, unsuccessful attempts, variability, retention, and generalization beyond the training environment. Efficiency depends on the outcome specified: reaching a criterion quickly can be different from producing accurate, durable performance or sustaining performance across settings.
The single-organism experimental tradition offers detailed histories of acquisition under changing contingencies. Replications across participants, response dimensions, and laboratories answer additional questions. Strong interpretation requires adequate baseline measurement, controlled changes when possible, transparent reporting, and clear separation of training effects from changes in motivation or apparatus. The objective of reproducibility is especially important because shaping can otherwise appear to be a matter of expert artistry. Orihara and Tanno, 2022
What Shaping Is Not: Chaining, Prompting, and Fading
Shaping changes the criteria defining which version of a response will produce reinforcement. Chaining arranges a sequence of component responses and their relevant antecedents and consequences. Prompting supplies assistance or additional cues that occasion a performance; fading systematically reduces such assistance or changes stimulus control. These procedures can be combined in one learning program, but they are analytically distinct.
Differential reinforcement of alternative behavior is also a related but separate arrangement: reinforcement is contingent on a specified alternative to another behavior. Shaping may refine an alternative behavior through progressively changing criteria, yet it cannot be inferred simply from the delivery of reinforcement for any different response. Accurate classification matters when comparing studies and attributing scientific effects to particular intervention components.
Response Form, Function, and Stimulus Control
Shaping can select response topography, but the visible form of an action does not determine its function in every context. A peck on a laboratory key is precisely measurable. Human behaviors, including communication and cooperation, often depend on social context and an individual's goals. Operational definitions are needed without reducing meaningful outcomes to arbitrary external appearances.
The result may also be restricted to the training environment. A newly acquired action might occur only in the presence of a particular trainer or apparatus. Stimulus generalization and transfer of control pose further questions: does the behavior occur under relevant natural cues, and do natural consequences maintain it? Shaping is therefore one part of a larger analysis involving antecedent conditions, reinforcement schedules, maintenance, and generalization.
Reinforcement Schedules and Historical Coauthorship
Shaping determines which approximation qualifies for reinforcement, while reinforcement schedules organize how consequences are delivered across eligible responses or time. These concepts interact without being identical. The 1957 volume Schedules of Reinforcement was authored jointly by Charles B. Ferster and B. F. Skinner. Historical attribution must recognize their collaboration and distinguish the schedule research from Skinner's earlier response differentiation and manual shaping. B. F. Skinner Foundation bibliography
A training arrangement might use a relatively dense reinforcement schedule during initial acquisition and a different schedule to maintain an established response. The effects of such changes depend on the contingency and context. The schedule alone does not explain how the initial response form was selected; the shaping criterion supplies that additional information.
Social Reinforcement and Verbal Behavior
Peterson interpreted Skinner's 1943 manual-shaping experience as relevant to his later conception of verbal behavior. The trainer mediated the consequences in a way that mechanical experimental arrangements had not made vivid. In verbal interaction, another person's behavior can play a comparable consequential role, making the social relationship analytically important. This supplies a documented historical link between shaping and Skinner's later research interests. Peterson, 2004
Language development is a broader subject. Contemporary linguistic and developmental theories address structure, meaning, social interaction, prediction and learning processes through independent evidence. The historical Skinnerian contribution should be examined alongside those developments, without treating the existence of social reinforcement as a comprehensive explanation of syntax or the entire cognitive architecture of language.
Educational Applications and Their Evidentiary Limits
In education, a teacher may define an attainable component of a meaningful skill and arrange informative consequences for increasingly capable performances. This can be coherent with shaping when criteria change to select progressively closer approximations. The procedure is best described concretely: what response counts, what consequence follows, when the criterion changes, and whether the skill transfers to ordinary practice.
Skinner's work on programmed instruction and teaching machines belongs to a related but separate part of his educational legacy. Gradual task progression, immediate feedback, prompting, and shaping can coexist, but their specific contributions require independent measurement. A historically plausible application of shaping is not evidence that a particular teaching program outperforms modeling, practice, explicit instruction, or alternative interventions.
Applied Behavior Analysis: Later Contributions
Skinner's experimental framework influenced subsequent applied behavior analysis, a field developed through the independent work of Donald Baer, Montrose Wolf, Todd Risley, and other researchers. Later applied researchers established standards for socially important behaviors, systematic measurement, practical intervention and functional relations. The field's diverse contemporary methods must not be equated with Skinner's personal authorship or with one shaping technique.
Ethical application involves goals chosen for meaningful benefit, transparent measurement, respect for assent and consent, reasonable alternatives, and ongoing evaluation of distress or adverse effects. Especially when learners are children, disabled people, or individuals subject to institutional authority, the criterion of success must account for the participant's agency and welfare. A response change demonstrated under a contingency does not independently validate the social desirability of the target.
Motor Rehabilitation as a Subsequent Translation
Edward Taub and colleagues reported an operant-derived approach to motor rehabilitation in 1994. Their account discussed learned nonuse, the adjustment of movement contingencies, and preliminary shaping-related findings among broader treatment procedures. The report described work involving constraint of an unaffected limb and training of an affected limb, so the intervention package cannot be reduced to shaping in isolation. The authors' report is a later interdisciplinary development, not a Skinner experiment. Taub et al., 1994
Human rehabilitation outcomes also depend on neurological injury, spontaneous recovery, biomechanics, practice dosage, motivation, context, and clinical support. Evidence of historical influence cannot substitute for contemporary condition-specific clinical trials, systematic reviews, and professional guidelines. A reader should distinguish the conceptual relation to reinforcement from verified treatment efficacy in a particular patient population.
Contemporary Cognitive and Neurobiological Perspectives
Current psychological and neuroscientific research examines how actions change through consequences together with attention, expectations, valuation, motor exploration, memory, and neural plasticity. Some research models distinguish goal-directed control from habits; others analyze trial-by-trial learning from prediction errors. These frameworks ask mechanistic questions that complement and sometimes challenge Skinner's preferred functional explanations. They are later independent research developments rather than discoveries that should be retroactively attributed to him.
A successful shaping curve demonstrates that contingencies and performance changed together under an experimental arrangement. It does not uniquely reveal the neural circuit or cognitive computation responsible for every observed change. Distinct mechanisms may produce similar response distributions, requiring additional measures or experimentally discriminating tests. Maintaining this distinction preserves the value of observable functional relations while leaving complex mechanistic questions open.
Established Findings, Limited Claims, and Research Gaps
The established experimental core includes response differentiation under contingent reinforcement and controlled changes in response distribution as criteria move. The 1980 peck-location study provides direct evidence about certain spatial criterion sizes and shift rates, while research on delayed reinforcement shows why timing and correlated stimuli matter. Such conclusions are grounded in explicit laboratory preparations and cannot be assumed to generalize automatically across species, response classes and settings. Eckerman et al., 1980 Lattal, 2010
In a 2022 review, Tomotaka Orihara and Takayuki Tanno observed that many publications on response shaping were demonstrations of skilled practice, with comparatively limited experimental analysis of the variables controlling success and failure. They identified standardization, automated response-topography measurement, and sustained parametric research as priorities. This evaluation qualifies blanket claims about universally optimal steps, outcomes or trainer techniques. It also identifies a constructive direction for turning accumulated expertise into a more rigorously comparative science. Orihara and Tanno, 2022
Questions about freedom, consciousness, agency, language structure and human values belong to broader empirical and philosophical discussions. A shaping experiment can establish a response-consequence relation without settling all of these questions. Scientific interpretation gains precision by differentiating experimentally demonstrated response changes from a global theory of persons, social organization or clinical effectiveness.
Replication and Methodological Transparency
Manual shaping can hide important decisions. A trainer may change the criterion after an unusually good performance, alter reinforcer intensity, supply prompts, or respond inconsistently to near-miss attempts. If these changes are not recorded, an account describing successful shaping cannot be reconstructed faithfully. Researchers should report their full criterion history, actual reinforcer delivery, response measures, equipment, baseline variability, and unsuccessful episodes.
A strong contemporary study could compare different progression rules within or across organisms, document task-relevant welfare conditions, and analyze retention and generalization in addition to immediate acquisition. Automated sensors may reduce some recording uncertainty, while independent observation can check whether a mechanical metric represents a meaningful target. Failed shaping sequences also deserve reporting because they help specify the boundary conditions of the method. Galbicka, 1994 Orihara and Tanno, 2022
Ethical Considerations and Participant Agency
Shaping deliberately arranges consequences to alter behavior. Ethical evaluation begins with the legitimacy of the target, the affected individual's preferences, the relationship between trainer and learner, and the foreseeable consequences of training. Animal research requires attention to welfare, appropriate reinforcement arrangements, housing and scientific oversight. Human applications call for consent or assent where appropriate, noncoercive participation, transparent goals and safeguards against adverse effects.
A precise measure of behavior change is valuable but does not by itself determine what another person should be required to do. A person-centered approach considers whether the new skill expands meaningful opportunity, whether alternatives are available, and whether the learner can refuse or revise the process. The experimental analysis of behavior addresses functional relations; justified educational and clinical decisions also require ethical and normative reasoning.
Why Skinner's Contribution Still Matters
Shaping gives operational substance to the idea that complex behavior can develop through the selection of intermediate performances. Skinner brought response differentiation into the experimental analysis of operant behavior and recognized the practical force of hand-mediated reinforcement. His later public writing extended the method's visibility. Other researchers then investigated criterion shifts, temporal relations and quantitative schedules, making it possible to test claims previously supported largely through skilled demonstrations. Peterson, 2004 Eckerman et al., 1980
The lasting scientific question is how differential contingencies, response variation, criteria and environmental conditions interact during acquisition. The history from 1938 through 1943 and the later experimental literature makes shaping an instructive case in both the power and limitations of behavioral experimentation. Its heritage is most accurately understood through documented methods, reproducible parameter studies, and carefully differentiated contemporary applications.
Frequently Asked Questions
Did Skinner discover shaping in 1943?
The 1943 manual-shaping episode was a genuine conceptual turning point, while response differentiation through successive approximations was already documented in his 1938 research. The distinction is developed in Peterson's historical study. Peterson, 2004
Is shaping simply rewarding a desired response?
Shaping requires a changing response criterion and differential consequences for successive approximations. Reinforcing an already established response without progression can be ordinary reinforcement but does not by itself establish shaping.
Must successive approximations always be tiny?
No universal step-size rule is supported. Eckerman and colleagues found relatively large and rapid spatial criterion changes effective within their particular pigeon experiment. Other response classes and learners may require different progression arrangements. Eckerman et al., 1980
How does shaping differ from chaining?
Shaping progressively differentiates the form or measurable properties of a response. Chaining establishes a sequence of component responses, commonly under discriminative stimulus control. The two procedures can be combined, but they analyze different functional relations.
Can shaping be applied to human skills?
It can be used in appropriately designed educational, applied and rehabilitation contexts. The evidence for any particular application depends on its population, target, measures, alternatives and ethical protections. Historical demonstrations do not establish clinical efficacy for every modern use.
Why is scientific comparison of shaping methods difficult?
Manual protocols often leave criterion changes and trainer judgments incompletely specified. Contemporary methodological proposals emphasize operational definitions, automated measurement, transparent schedules and controlled comparisons. Orihara and Tanno, 2022
Related Articles
B. F. Skinner — An integrated account of Skinner's life, experimental work, radical behaviorism, principal publications, and scientific legacy.
Learning Theories in Education: Behaviorism, Cognitivism, Constructivism, and Sociocultural Learning — The broader historical and scientific setting for behavioral and other educational theories.
Behavior Change Techniques: What Research Supports — A separate applied account of behavior-change interventions and their evidence base.
B. F. Skinner's Operant Conditioning Theory — The larger experimental framework of operant contingencies within which response shaping was developed.
B. F. Skinner's Radical Behaviorism — Skinner's philosophy of behavioral science and the explanatory status of operant contingencies.
B. F. Skinner and C. B. Ferster's Schedules of Reinforcement — The coauthored schedule research that extended experimental knowledge of how reinforcement conditions organize responding.
B. F. Skinner's Three-Term Contingency and Stimulus Control — Antecedents, contingencies, and discriminative control.
B. F. Skinner Box and the Operant Conditioning Chamber — The operant chamber and response measurement.
B. F. Skinner's Influence on Applied Behavior Analysis — The history and later development of applied behavior analysis.
References
Eckerman, D. A., Hienz, R. D., Stern, S., and Kowlowitz, V. (1980). Shaping the location of a pigeon's peck: Effect of rate and size of shaping steps. Journal of the Experimental Analysis of Behavior, 33(3), 299–310. https://doi.org/10.1901/jeab.1980.33-299
Ferster, C. B., and Skinner, B. F. (1957). Schedules of Reinforcement. Appleton-Century-Crofts. B. F. Skinner Foundation bibliography
Galbicka, G. (1994). Shaping in the 21st century: Moving percentile schedules into applied settings. Journal of Applied Behavior Analysis, 27(4), 739–760. https://doi.org/10.1901/jaba.1994.27-739
Lattal, K. A. (2010). Delayed reinforcement of operant behavior. Journal of the Experimental Analysis of Behavior, 93(1), 129–139. https://doi.org/10.1901/jeab.2010.93-129
Orihara, T., and Tanno, T. (2022). A review of studies on response shaping: From art to science. Japanese Psychological Review, 65(1), 1–19. https://doi.org/10.24602/sjpr.65.1_1
Peterson, G. B. (2004). A day of great illumination: B. F. Skinner's discovery of shaping. Journal of the Experimental Analysis of Behavior, 82(3), 317–328. https://doi.org/10.1901/jeab.2004.82-317
Skinner, B. F. (1938). The Behavior of Organisms: An Experimental Analysis. Appleton-Century. B. F. Skinner Foundation bibliography
Skinner, B. F. (1951). How to teach animals. Scientific American, 185(6), 26–29. https://doi.org/10.1038/scientificamerican1251-26
Skinner, B. F. (1953). Science and Human Behavior. Macmillan. B. F. Skinner Foundation bibliography
Taub, E., Crago, J. E., Burgio, L. D., Groomes, T. E., Cook, E. W., DeLuca, S. C., and Miller, N. E. (1994). An operant approach to rehabilitation medicine: Overcoming learned nonuse by shaping. Journal of the Experimental Analysis of Behavior, 61(2), 281–293. https://doi.org/10.1901/jeab.1994.61-281