Mary Cover Jones

Introduction

Mary Cover Jones (1 September 1896 – 22 July 1987) is called the mother of behaviour therapy, a title given to her by Joseph Wolpe in the 1970s — roughly half a century after the work that earned it.

In 1924, working at Columbia University Teachers College, she took a three-year-old boy who was terrified of rabbits and removed the fear. The method was the inverse of the experiment she had heard described in a lecture: where John B. Watson had conditioned a child into a fear, she conditioned one out of it, by pairing the feared object with something pleasant and bringing it gradually closer.

The study was published and then largely forgotten. It was not her dissertation, it was a single case, and the field that would have valued it did not exist yet. When behaviour therapy emerged in the 1950s and 1960s, its practitioners found that someone had already done the founding experiment — and had spent the intervening decades doing something else entirely, a longitudinal study of adolescence that turned out to be one of the most important in developmental psychology.

Early Life and Education

Mary Cover was born in Johnstown, Pennsylvania, in 1896, into a family that valued her education and sent her to Vassar College in 1915. She took as many psychology courses as she could find and did laboratory work under Margaret Floy Washburn, the first woman to receive a doctorate in psychology.

Her introduction to behaviourism was close to accidental. While visiting New York she was taken to a lecture by Watson, who described and showed film of the Little Albert study — the conditioning of a fear of a white rat in an infant. What struck her was not the demonstration but its direction. If fears could be produced this way, they could presumably be removed.

She enrolled for a master's degree at Columbia in 1919 and completed it in 1920. That same year she married Harold Jones, a fellow psychology graduate student. In 1923 she became a research associate at the Institute of Educational Research at Columbia Teachers College.

Watson, by then, had been forced out of Johns Hopkins and was working in advertising in New York. Cover Jones had been a friend and classmate of his second wife, Rosalie Rayner, and through that connection received his advice on the Peter study. Her later account of the work was careful to credit him, and equally careful to note the method as her own development.

Academic and Professional Career

The Peter study was published in 1924. She completed her doctorate at Columbia in 1926 with a dissertation on the development of imitation in infants, based on observations of 365 babies — deliberately large-scale, because she understood that a single case would not carry the weight she wanted it to.

In 1927 her husband was appointed director of the newly founded Institute of Child Welfare at the University of California, Berkeley, and the family moved. She joined the institute as a research associate, which is where she spent the rest of her career and where she did the work that occupied most of her professional life.

Her advancement there was blocked for two decades. University rules against nepotism — her husband was the institute's director — prevented her from holding a regular academic appointment. She became a lecturer in psychology only in 1946, an associate professor of education in 1952, having spent seven years in an assistant professorship, and a full professor in 1959, one year before she retired.

She retired in 1960 and was elected president of the APA's Division of Developmental Psychology the same year. Harold Jones died of a heart attack in March of that year, shortly after their retirement.

Alongside the research, she and Harold produced what is generally described as the first educational television course on child development, broadcast in 1952.

Major Contributions to Psychology

The case of Peter

Peter was a healthy, active three-year-old with an intense fear of rabbits that generalised to other furry things — white rats, fur coats, feathers, cotton wool. Crucially, his fear had been acquired naturally, not induced by an experiment; he was in every other respect an ordinary child.

Cover Jones tested several methods before settling on what she called direct conditioning. Peter was placed in a high chair and given food he liked. A caged rabbit was put at the far end of the room, twelve feet away, close enough to be noticed but not to disrupt his eating. Over successive sessions the rabbit was moved closer, always paired with the pleasurable activity and never brought near enough to provoke real fear. Eventually Peter could touch the rabbit and let it nibble his fingers.

She documented the progression as a tolerance series — an ordered sequence of steps from "rabbit anywhere in the room produces fear" to "rabbit allowed on the high chair tray." This is the ancestor of the anxiety hierarchy that Joseph Wolpe formalised three decades later as systematic desensitisation, and the technique is still the standard behavioural treatment for specific phobias.

The contrast with Watson is instructive in a way she intended. Watson's study produced a fear and did not remove it; he never deconditioned Albert, and by modern standards the study could not have been ethically completed. Hers removed one, in a child who arrived with it, using a procedure designed to keep him comfortable throughout.

The Oakland Growth Study

The work she spent most of her career on, with Harold, was the Oakland Growth Study, begun in 1932. It followed 212 fifth- and sixth-graders born in 1920–21, assessing them twice a year until they finished high school, and then following them into adulthood at ages 38, 48, and 60.

The aim was to describe ordinary adolescence — physical development, physiological change, peer relationships — at a time when the field had almost no longitudinal data on it. The study produced more than a hundred papers, and its longevity is attributable in no small part to Cover Jones's relationships with the participants; she was still conducting interviews in 1980, at eighty-three.

Infant behaviour and imitation

Her doctoral work on imitation in 365 infants, published with Barbara Burks in 1936, was a deliberate corrective to the single-case limitation of the Peter study, and contributed to the empirical description of early behavioural development.

Major Publications

  • "A Laboratory Study of Fear: The Case of Peter" (1924, Pedagogical Seminary, 31, 308–315)
  • "The Development of Early Behavior Patterns in Young Children" (1936, with Barbara Burks)
  • "Albert, Peter, and John B. Watson" (1974, American Psychologist, 29, 581–584) — her retrospective on the two cases

Impact and Legacy

Every exposure-based treatment for phobia descends from the Peter study, and the tolerance series is the direct ancestor of the graded hierarchy used in systematic desensitisation. Cover Jones received the APA's G. Stanley Hall Award in 1968 (APA Division 7, shared that year with Lois Meek Stolz) for her contributions to developmental psychology.

The Oakland Growth Study is one of the foundational longitudinal datasets in developmental psychology, and one of the few that followed participants across six decades — the kind of study that cannot be designed retrospectively and that later researchers have relied on for questions its designers never asked.

What the record also shows. Her case is one of the clearest examples in the discipline of a contribution being invisible for reasons that had nothing to do with its quality. Three things worked against it, and only one of them was intellectual.

The first was the single case. A study of one child, even a meticulously documented one, did not meet the evidentiary standards of the period, and she knew it — which is why her dissertation was on 365 infants instead. The Peter study was published in a journal with a small readership and was not built upon for decades.

The second was the company it kept. Watson had left academic psychology in disgrace, and behaviourism was not the framework within which developmental research was being done in the 1930s and 1940s. A technique that looked like an application of Watson's programme had nowhere to go while that programme was out of favour.

The third was structural. The anti-nepotism rule at Berkeley cost her roughly twenty years of academic standing. She was doing the Oakland Growth Study as a research associate while her husband held the directorship, and she did not hold a professorial rank until 1952 or a full professorship until the year before she retired. Her promotion record has no relationship to her scientific output.

There is also a complication now attached to the study she was responding to. The Little Albert experiment — the one that inspired her — has been substantially discredited: recent analysis of Watson's film and notes concludes the conditioning was largely ineffective, and the identity of the child is disputed between candidates that include a boy with congenital hydrocephalus. Cover Jones's work does not depend on Watson's being sound, and one of its merits is that it demonstrates the reverse procedure in a child whose fear was ordinary. But the pairing of "Albert" and "Peter" in the literature has linked her clean study to a messy one, and the 1974 paper she wrote disentangling them is worth reading for that reason.

She is sometimes described as having been modest to the point of complicity in her own neglect. That is probably unfair. She published the study, she returned to it in print in 1974, and she continued working. What she could not do was make the field want it before the field was ready.

Learn More

References

  • Jones, M. C. (1924). A laboratory study of fear: The case of Peter. Pedagogical Seminary, 31, 308-315.

Last updated: 2026-10-20

Menu
Loading