Charles Ferster
Introduction
Charles Bohris Ferster (1 November 1922 – 3 February 1981) was an American behavioral psychologist who stood at the hinge between the experimental analysis of behavior and its clinical application. As B. F. Skinner's principal collaborator during the five years of research that produced Schedules of Reinforcement (1957), he helped systematise how the timing and probability of reward determine the pattern of behaviour that follows. As the founder and first executive editor of the Journal of the Experimental Analysis of Behavior, he gave the field a venue when the mainstream journals were rejecting its single-subject methods. And as one of the first researchers to apply errorless learning and token economies to autistic children and to adults with depression, he carried laboratory principles into the clinic years before applied behavior analysis existed as a named discipline.
Ferster spent the last decade of his life moving in the opposite direction from most of his peers: away from the pigeon laboratory and toward education, psychotherapy, and the design of learning environments. He died of a heart attack in Washington, D.C., in 1981, at the age of 58.
Early Life and Education
Ferster was born in Freehold, New Jersey, the second son of Julius Ferster and Mollie Ferster, née Madwin, Jewish immigrants who had arrived from Russian-ruled Poland in 1910 and 1912. He entered Rutgers University in 1940 with the support of an uncle, had his studies interrupted by military service from 1943 to 1946, and returned to complete a B.S. in 1947.
He then enrolled at Columbia University for graduate study, taking his M.A. in 1948 and his doctorate in 1950, and studying with two of the leading figures in American learning theory: Fred S. Keller, author of Principles of Psychology, and William N. Schoenfeld. Their approach — the experimental analysis of the individual organism rather than the statistical comparison of groups — became Ferster's own. He had begun working with Skinner on pigeon research even before completing his doctoral examinations.
Academic and Professional Career
From 1950 to 1955 Ferster was a research fellow at Harvard University, working in Skinner's pigeon laboratory. This was the period of the collaboration that produced Schedules of Reinforcement, and the period in which he developed errorless learning, a training method that arranges conditions so that a learner is never given the opportunity to make a mistake that then has to be extinguished.
He then spent two years, 1955 to 1957, at the Yerkes Laboratory of Primate Biology, working with chimpanzees. In 1957 he moved to the Indiana University School of Medicine as an assistant professor of psychology, where he remained until 1962. There he did two quite different things at once: with John I. Nurnberger and James P. Brady he built a laboratory that recorded physiological responses in freely behaving pigeons, and with a separate clinical group he applied errorless learning to teach speech to young autistic children — work that is among the earliest documented applications of operant procedures to autism, and among the first uses of what came to be called a token economy.
In 1958, frustrated by the difficulty behavior analysts had getting single-subject research published in the mainstream experimental journals of the day, Ferster took the lead in organising a dedicated outlet: the Journal of the Experimental Analysis of Behavior, which he founded and served as its first executive editor.
From 1962 to 1968 he worked at the Institute for Behavioral Research in Silver Spring, Maryland, first as executive director and later as a senior research associate, studying depression, overeating, and smoking as behaviour maintained by its consequences. After a brief professorship at Georgetown University he joined American University in Washington, D.C., in 1969, chaired the psychology department from 1970 to 1973, and later directed the University Learning Center, an experiment in undergraduate education built directly on operant principles. He remained at American University until his death in 1981.
Major Contributions to Psychology
Schedules of reinforcement
The 1957 book Ferster wrote with Skinner was not a summary but a compendium: more than seven hundred pages of cumulative records documenting how pigeons' rates and patterns of responding change under fixed and variable ratios, fixed and variable intervals, and the compound and concurrent schedules built from them. Its contribution was to show that reinforcement schedule is not a nuisance variable to be controlled but the primary independent variable of interest — that the shape of a behaviour over time is a lawful function of the contingency in force. The vocabulary it established (ratio, interval, fixed, variable, concurrent, chained) is still standard.
Errorless learning
Ferster's errorless learning procedure addressed a practical problem: when a learner makes an error, the consequences that follow it are themselves part of what is learned, and extinguishing errors costs time and often produces emotional side effects. His solution was to construct a teaching sequence in which stimuli are faded in gradually enough that the correct response is always available and wrong responses are effectively prevented. The technique became standard in special education and in training programmes for people with developmental disabilities.
The behavioural analysis of autism
Ferster's work at Indiana with autistic children — and later his involvement with the Linwood Center, one of the first community programmes for autistic children in the United States — established that the acquisition of language and social behaviour in these children could be treated as an instructional problem with measurable inputs and outputs. His procedures were taken up and extended by Ivar Lovaas at UCLA, whose discrete trial training became the dominant early intervention model. Donald Baer and Sidney Bijou, who founded the Journal of Applied Behavior Analysis, also drew on his work.
The primary reports are short and specific: Ferster's “Positive reinforcement and behavioral deficits of autistic children” (Child Development, 32, 437–456, 1961), Ferster and DeMyer's “The development of performances in autistic children in an automatically controlled environment” (Journal of Chronic Diseases, 13, 312–345, 1961), and their method paper in the American Journal of Orthopsychiatry (32, 89–98, 1962). Children worked alone in a room with devices operated by a coin or a key; the coins were exchangeable for food, trinkets, a music handset or a picture viewer, and contingent delivery shaped behaviours such as matching figures and pointing to a target picture.
A functional analysis of depression
In a widely cited 1973 paper in the American Psychologist, Ferster argued that depression should be understood as a drastic reduction in the frequency of positively reinforced behaviour: the depressed person's repertoire narrows, activity that once produced reinforcement ceases to produce it, and the resulting passivity further reduces contact with reinforcement. The account was influential because it reframed a diagnostic category as something with identifiable maintaining conditions, and it anticipated the behavioural activation therapies that later proved comparable to cognitive therapy in controlled trials.
The University Learning Center
Ferster's last major project applied the same principles to undergraduate education. The Learning Center at American University was an open, interdisciplinary space staffed by faculty from psychology, physics, anthropology, philosophy, mathematics, and psychiatry, in which instruction was organised around immediate reinforcement, successive approximation, and explicit discriminative stimuli. It was a radical and, to many of his experimental colleagues, an eccentric undertaking — but it followed logically from his conviction that the units of analysis developed in the pigeon laboratory were general.
Major Publications
- Ferster, C. B. & Skinner, B. F. (1957). Schedules of Reinforcement. New York: Appleton-Century-Crofts.
- Ferster, C. B., Nurnberger, J. I. & Brady, J. P. (1963). An Introduction to the Science of Human Behavior. New York: Appleton-Century-Crofts.
- Ferster, C. B. & Perrott, M. C. (1968). Behavior Principles. New York: Appleton-Century-Crofts.
- Ferster, C. B. (1973). "A functional analysis of depression." American Psychologist, 28(10), 857–870.
- Ferster, C. B. (1979). "The evolution of behavior analysis." In M. D. Zeiler & P. Harzem (Eds.), Reinforcement and the Organization of Behavior. Chichester: Wiley.
Impact and Legacy
Ferster's founding of JEAB had consequences well beyond the convenience of a publication venue. The journal institutionalised the single-subject, cumulative-record methodology as a legitimate form of evidence at a moment when mainstream experimental psychology was organised around group designs and inferential statistics, and it created the institutional core around which behavior analysis coalesced as a discipline. Applied behavior analysis as it exists today — with its own journals, certification, and professional practice — is the downstream product.
His work also carried a substantial cost, and the criticisms fall into three groups.
His contribution has been absorbed into Skinner's. The standard apparatus of the field is universally called the Skinner box, the research programme is described as Skinnerian, and Schedules of Reinforcement is usually remembered as a Skinner book. Ferster was the principal investigator on the experiments the book reports, and historians of the field have noted that the naming convention has obscured his role. He was, by the accounts of colleagues including Richard Herrnstein, the most aggrieved of the early behavior analysts about the journal rejections that prompted JEAB's founding — and the one who pushed the plan through.
The behavioural account of depression proved hard to test. Ferster's formulation has been criticised as close to circular: if depression is defined by a reduction in positively reinforced behaviour, and the reduction in reinforcement is inferred from the person's reduced activity, then the explanation and the phenomenon are difficult to separate. Identifying the operative reinforcers independently of the behaviour they are supposed to maintain — the central empirical requirement of the account — has proved much harder in practice than in principle. Behavioural activation, the therapy most indebted to it, works, but researchers disagree about whether it works for the reason Ferster gave.
The autism lineage is contested. Ferster's own procedures relied on positive reinforcement, and he was explicitly concerned with avoiding the punitive methods common in institutions of the period. But the programme he helped start led directly to Lovaas's discrete trial training, which did use aversives, and to a broader applied behavior analysis tradition that has been criticised on several grounds by autistic self-advocates and by the neurodiversity movement: for targeting the reduction of visibly autistic behaviour as a therapeutic goal, for the intensity of the regimens imposed on very young children, and for measuring success by similarity to non-autistic peers rather than by the person's own well-being. Ferster did not live to see that debate, and he is not among its principal targets, but his work is in the lineage it concerns.
Two developments since 2019 have changed what the lineage looks like. The first is evidential. A 2020 audit found that 84% of autism intervention studies in ABA journals had at least one author with a clinical or consultancy conflict of interest and only 2% disclosed it. The five-year update is worse rather than better: Kristen Bottema-Beutel, Jessica Hinson-Williams, Yueyang Shen, Ruoxi Guo, Samantha Brayton, Jacqueline Alicea and Shannon LaPoint examined 119 studies and 450 authors across eight journals and found that 78% of authors held such conflicts, 93% of studies had at least one, 8% disclosed any conflict at all, and 93% of statements declaring no conflict were false (Autism, 30(7), 1887–1894, 2026). Whatever else this is, it is not the evidential culture Ferster built JEAB to protect.
The second is that the argument is now conducted from inside the field rather than against it. David Jackson-Perry, Elsa Suckle, Nick Chown and Jonathan Tarbox — an interdisciplinary team of critical autism and behavior-analytic scholars — proposed in 2025 a “critical behavioral studies” that would make history, culture and self-examination part of behavior-analytic training (Perspectives on Behavior Science, 48, 539–545), while Amy Lutz, in the same issue (48, 547–553), made the opposite case: that the neurodiversity critique rests on an “intact mind” assumption which fails the profoundly autistic people for whom the practical goals are physical safety, communication and basic life skills. Ferster's own procedures were built on positive reinforcement at a time when the institutional alternative was punitive, and that is the part of the lineage his defenders claim; what his critics answer is what the lineage became once it was an industry.
One correction has gone his way. In a 2024 invited address to the Association for Behavior Analysis International, Edward K. Morris argued that Ferster's five 1961–1962 publications preceded every other behavior-analytic contribution to autism, and that his collaborator, the psychiatrist Marian K. DeMyer (1922–2021), had been rendered invisible in the field's standard history of applied behavior analysis.
Finally, Ferster's own trajectory — from the pigeon laboratory to the clinic to the design of a university — was out of step with the experimental mainstream he had helped build. Whether that represents the logic of behavior analysis pursued consistently or a departure from what it does best remains an open question in the field.