The question
In the 1950s, two very different schools of psychology agreed on one thing about babies. Behaviourists explained a baby’s love for its mother as learning: she brings food, food relieves hunger, and so her presence becomes rewarding. Many psychoanalysts also placed feeding, and the mouth, at the centre of the first bond. The psychiatrist John Bowlby, in a 1958 paper, called this the “cupboard-love” theory, and he doubted it.
Harry Harlow, a psychologist at the University of Wisconsin, was in a position to test it. He had a simple question: if feeding and comfort are separated, which one does an infant attach to?
How it was done
Harlow’s laboratory had begun breeding rhesus monkeys in the early 1950s. Hand-rearing sharply cut infant deaths in the colony, so infants were taken from their mothers 6 to 12 hours after birth and bottle-fed, each in its own cage. Staff noticed that the infants clung to the soft cloth pads on the cage floor and became upset when the pads were taken away to be cleaned.
That observation became an experiment, which began in 1957. The idea of setting a cloth surrogate against a wire one came from William Mason, then in Harlow’s laboratory; Harlow added a head with a face. Mason, who disagreed with adding the face, stepped back, and Robert Zimmermann took over the work with Harlow.
Two artificial “mothers” were built, the same size and shape:
- The cloth mother: a block of wood covered in sponge rubber and soft terry cloth, warmed from behind by a light bulb.
- The wire mother: the same body made of bare wire mesh, also warmed.
Each could hold a feeding bottle. In the main study, eight newborn monkeys each lived with both mothers. For four of them, only the cloth mother gave milk; for the other four, only the wire mother did. Small control groups were raised with only one kind of mother, or with none.
A machine recorded how long each infant spent on each mother. The researchers also watched what happened when something frightening appeared, such as a moving toy bear, and when an infant was placed in an unfamiliar room, with or without the cloth mother there.
Harlow presented the first results in his presidential address to the American Psychological Association in August 1958, under the title “The nature of love”. He and Zimmermann published a fuller report in Science in 1959.
What came out
The answer was clear, and it was the same in both groups.
- Comfort beat milk. Infants spent most of their time clinging to the cloth mother, whichever mother fed them. Harlow described the effect of feeding on the preference as negligible. Over 165 days, the infants fed by the wire mother moved even further towards the cloth one.
- Same growth. Infants raised with only a wire mother or only a cloth mother drank the same amount of milk and gained weight at the same rate. Harlow’s summary: the wire mother was “biologically adequate but psychologically inept.”
- A safe place to run to. When a strange moving toy appeared, frightened infants ran to the cloth mother and clung to her, again regardless of who had fed them. The preference grew stronger with age.
- A base for exploring. In an unfamiliar room with the cloth mother present, infants first clung to her and then began to explore, returning to her from time to time. Without her, they showed strong signs of distress: crying out, crouching, rocking. Their distress scores were roughly halved when she was there.
- A lasting bond. When the infants were separated from the surrogates at about five and a half months, they still rushed to the cloth mother at reunion months later. Harlow argued that a bond learned only through feeding should have faded; this one did not.
A group of control infants first received both mothers at about eight months of age. They were distressed for a day or two, then gradually took to the cloth mother, soon spending about nine hours a day on her and under an hour and a half on the wire one. Harlow thought their attachment was real but probably weaker than if it had begun at birth.
His conclusion was that contact comfort, the reassurance of touching something soft, is central to the bond between infant and mother, and that nursing plays a minor part.
What really happened (replications and critiques)
The central finding held up. The preference for the cloth mother appeared in every group Harlow tested and in follow-up tests over many months. Later reviewers, including Stephen Suomi, who worked in Harlow’s laboratory, say it undermined both the learning-theory and the psychoanalytic accounts that tied attachment to feeding.
But the experiment had limits. Each group was small: four infants per feeding condition in the main study. Warmth was held equal, but other features, such as the face that Mason had objected to, were not tested systematically. Harlow himself listed rocking, the mother’s face, sounds and postural support as variables still to be studied. And the step from rhesus monkeys to human babies rests on analogy, not on direct tests with children.
The isolation studies. In the 1960s the laboratory went further, and this is the part of the work many readers find hardest to read about. Some monkeys were raised with no social contact at all, for periods ranging from the first three months to the whole first year. Monkeys isolated for the longer periods grew up with serious and lasting difficulties in relating to other monkeys. A 1971 study by Harlow and Suomi found a form of recovery: isolated six-month-olds placed with younger, normally raised three-month-olds made a nearly complete social recovery in the situations tested. That finding suggested early harm is not always permanent, though the question of limits was left open.
Harlow and Bowlby. The two men were introduced through the ethologist Robert Hinde and went on to correspond. Harlow apparently sent Bowlby a draft of his 1958 address before it was published, and Bowlby mentioned the early, still unpublished results in a footnote of his own 1958 paper. Harlow, in turn, said it was Bowlby who drew him towards studying the social and emotional effects of isolation, and his laboratory’s separation studies borrowed Bowlby’s terms protest, despair and detachment. Harlow’s work gave Bowlby’s attachment theory something it had lacked: experimental evidence that attachment is a need in itself, not a side effect of feeding. Suomi and his co-authors note that Mary Ainsworth’s later classification of attachment styles in children owed little to the monkey work, but that her idea of the mother as a “secure base” fits closely with what Harlow saw in the unfamiliar room.
Ethics
- Newborns were taken from their mothers, first to reduce infant deaths in the colony, and then because the studies depended on it.
- The isolation studies caused deliberate, long-lasting harm to young animals whose capacity for distress was the very thing being shown.
The psychologist John Gluck, once a graduate student in Harlow’s laboratory, has called this an ethical paradox. Harlow’s work did more than almost any other to show that monkeys have rich emotional lives, with fear, comfort and attachment. Yet the same work treated that suffering as acceptable. Gluck later stopped his own primate research and turned to research ethics, describing his journey in a 2016 book.
The debate continues. In the mid-2010s, an animal-rights campaign against studies of infant monkeys at a US National Institutes of Health laboratory helped prompt Congress to ask the NIH to review its ethics policies for primate research, and the NIH held a workshop on the subject in 2016. At that time, roughly two in five research primates in the United States were still housed alone or in pairs.
What it means today
The core idea has become part of how we think about care: babies need warmth, touch and a dependable presence, not only food. That view is now built into attachment theory, which remains central in developmental psychology.
Some practical thoughts follow from it, though they go well beyond what Harlow tested, and his studies were on monkeys, not children:
- Holding and comforting a baby is not “spoiling”; it is part of what helps a child feel safe enough to explore.
- A trusted person nearby can make new places and new people less frightening, for children and, often, for adults too.
- Early hardship matters, but the recovery seen in the 1971 study is a reminder that warm relationships later on can still help.
The studies also leave a harder lesson about science. A question can be important and the answer valuable, and the way it was obtained can still be wrong.
Would you have passed?
Coming soon. Here you are not the subject; you are the review board. You will read short descriptions of what was planned, from the first cloth-and-wire comparison to the isolation studies, and decide at each point whether you would approve it, ask for changes or refuse it. At the end you will see where today’s animal-welfare rules would draw the line, and how your choices compare.
Questions people ask
What did Harlow's monkey experiment show?
Infant rhesus monkeys were given two artificial mothers: one made of bare wire and one covered in soft cloth. Even when only the wire mother gave milk, the infants spent most of their time clinging to the cloth mother and ran to her when frightened. Harlow concluded that comfort from contact matters far more for attachment than feeding.
Why was Harlow's study important?
At the time, many psychologists believed that babies love their mothers mainly because mothers feed them. Harlow's results undermined that view and gave the psychiatrist John Bowlby experimental support for his idea that attachment is a basic need in its own right.
What were Harlow's isolation experiments?
In the 1960s, Harlow's laboratory raised some monkeys with no social contact at all for periods of up to a year. Isolation of six months or more left them with serious, lasting difficulties with other monkeys. Later work found that gentle contact with younger, playful monkeys led to substantial recovery.
Were Harlow's experiments ethical?
By today's standards they are hard to defend, the isolation studies above all. They showed that monkeys have rich emotional lives, and in doing so caused them serious suffering. One of Harlow's former students, John Gluck, later became a leading voice in the ethics of animal research.
Sources
- Harlow, H. F. (1958). The nature of love. American Psychologist, 13(12), 673–685. doi:10.1037/h0047884
- Harlow, H. F., & Zimmermann, R. R. (1959). Affectional response in the infant monkey. Science, 130(3373), 421–432. doi:10.1126/science.130.3373.421
- Harlow, H. F. (1959). Love in infant monkeys. Scientific American, 200(6), 68–74.
- Bowlby, J. (1958). The nature of the child's tie to his mother. International Journal of Psycho-Analysis, 39, 350–373.
- Harlow, H. F., Dodsworth, R. O., & Harlow, M. K. (1965). Total social isolation in monkeys. Proceedings of the National Academy of Sciences, 54(1), 90–97. doi:10.1073/pnas.54.1.90
- Harlow, H. F., & Suomi, S. J. (1971). Social recovery by isolation-reared monkeys. Proceedings of the National Academy of Sciences, 68(7), 1534–1538. doi:10.1073/pnas.68.7.1534
- Suomi, S. J., van der Horst, F. C. P., & van der Veer, R. (2008). Rigorous experiments on monkey love: An account of Harry F. Harlow's role in the history of attachment theory. Integrative Psychological and Behavioral Science, 42, 354–369. doi:10.1007/s12124-008-9072-9
- Gluck, J. P. (1997). Harry F. Harlow and animal research: Reflection on the ethical paradox. Ethics & Behavior, 7(2), 149–161.
- Gluck, J. P. (2016). Voracious science and vulnerable animals: A primate scientist's ethical journey. University of Chicago Press.
- Keim, B. (2016, December 4). Reconsidering research on primates. The Chronicle of Higher Education. www.chronicle.com/article/Reconsidering-Research-on/238540