Harlow's Monkeys: Comfort Over Food — SciHub101

In 1958, Harry Harlow presented findings at the American Psychological Association annual convention that directly contradicted what most American parents had been told about raising children. The dominant scientific consensus — backed by behaviorism's most authoritative voices, from John Watson to Benjamin Spock's first editions — held that babies bonded with their mothers because mothers provided food. Love was a conditioned response to reward. Harlow's infant rhesus monkeys demolished that claim in one of the most definitive experimental results in twentieth-century psychology. They chose comfort. Every time. Even when comfort came with no food at all.

Definition: Attachment — a deep, enduring emotional bond that connects one person (or animal) to another across time and space; in developmental psychology, the concept refers specifically to the bond formed between an infant and a primary caregiver, which influences emotional regulation and social behavior throughout life.

The Setup: Wisconsin, Behaviorism, and the Prevailing Theory of Love

Harry Frederick Harlow was born in 1905 in Fairfield, Iowa, and completed his doctorate in psychology at Stanford in 1930. He joined the faculty at the University of Wisconsin-Madison, where he spent his entire career and established the Wisconsin Regional Primate Research Center.

Source: Harry Harlow — Wikipedia

By the early 1950s, Harlow was breeding rhesus macaques (Macaca mulatta) for learning experiments. For hygiene reasons, the infant monkeys were separated from their mothers at birth and raised in individual wire cages. The cages contained cloth diapers as floor padding. Harlow noticed something that a strict behaviorist would have predicted was impossible: the infant monkeys became strongly attached to the cloth diapers. They clung to them when startled. They showed distress when the diapers were removed for cleaning. They used them as a refuge.

This was not supposed to happen. The dominant learning theory, derived from the drive-reduction models of Clark Hull and the conditioning principles of John Watson and B.F. Skinner, held that attachment developed because a mother associated with the primary drive-reducer: food. The mother who provides milk becomes reinforcing through conditioning. Infants love their mothers because mothers mean food. By this logic, a cloth diaper — which provided no food — should have no psychological significance whatsoever.

Source: Harry Harlow's Classic Studies — Association for Psychological Science

The cloth diaper observation suggested something fundamentally different was happening. Harlow designed a systematic experiment to test whether contact comfort — the physical sensation of warmth and softness — could function as an independent variable in attachment formation, separate from feeding.

John Watson's Warning and What the Experiments Were Challenging

To understand why Harlow's results mattered, you need to know what parents were being told in 1950s America. Watson's 1928 book Psychological Care of Infant and Child advised against picking up or comforting crying infants: "Never hug and kiss them, never let them sit on your lap. If you must, kiss them once on the forehead when they say goodnight. Shake hands with them in the morning." Physical affection would spoil children and produce emotional dependency.

This advice was not fringe. Watson was one of the most prominent psychologists in the United States. The Infant Care bulletins issued by the Children's Bureau echoed similar themes well into the 1940s. The behaviorist framework left no room for physical comfort as a primary need.

The Protocol: Wire Mothers, Cloth Mothers, and What Infants Chose

Harlow's first major experimental series, reported at the APA convention on August 31, 1958, in his address titled "The Nature of Love," used a simple but powerful design.

Source: Harlow's Monkeys — Simply Psychology

Infant rhesus monkeys were separated from their biological mothers within hours of birth and placed individually in cages. Each cage contained two surrogate "mothers" constructed from hardware wire and wood:

Wire mother: A cylindrical frame of wire mesh shaped to suggest a primate body, with a wooden head and a bottle of milk attached so the infant could feed from a nipple in the mother's "chest."

Cloth mother: An identical frame covered with soft terry-cloth toweling. The cloth mother provided warmth and texture but, in one experimental condition, no food.

Two groups of infants were tested:

• Group 1: Milk from the wire mother only; cloth mother present but without food
• Group 2: Milk from the cloth mother; wire mother present but without food

Harlow measured how much time the infants spent in contact with each surrogate, and what happened when the infants were frightened by a novel stimulus (typically a mechanical toy that made noise and moved unpredictably).

Source: Harlow's Monkeys Explained — Modern Therapy Online

The Results Were Unambiguous

Both groups of infants spent the vast majority of their time — approximately 17–18 hours per day — in contact with the cloth mother, regardless of which mother provided food. Infants fed by the wire mother would visit it briefly to nurse and then immediately return to the cloth mother. They did not form attachment to the wire mother even when it was the only food source.

Measure	Wire Mother (food)	Cloth Mother (comfort)
Average daily contact time	~1–2 hours	~17–18 hours
Refuge during fear stimulus	Rarely used	Used consistently
Exploration base (open field)	Not used	Used as secure base
Response to novel objects	Distress without cloth mother	Exploration with cloth mother present

Infants fed exclusively by wire mothers showed identical preference for the cloth mother as infants fed by cloth mothers. Feeding location did not predict attachment location.

Source: Harlow's Classic Studies — Association for Psychological Science

When frightened by the mechanical toy, infants with access to the cloth mother ran to her, clung to her, gradually calmed, and then began to explore the environment again, using the cloth mother as a "secure base" from which to venture. Infants without access to a cloth mother showed prolonged, unresolved distress — rocking, huddling, and self-clutching behaviors that resembled the stereotyped self-stimulation seen in children raised without adequate human contact.

This secure-base concept would prove central to John Bowlby's attachment theory, which Bowlby was developing simultaneously in London. Harlow and Bowlby were in close scientific contact from 1957 through the mid-1970s, and Harlow's empirical findings gave Bowlby the primate evidence he needed to argue that human attachment was a biologically programmed survival mechanism, not a learned response to feeding.

Source: "When Strangers Meet": John Bowlby and Harry Harlow on Attachment Behavior — Springer

The Timeline: From Wire Mothers to the Pit of Despair

Year	Development
1930	Harlow joins University of Wisconsin-Madison faculty
Early 1950s	Harlow breeds rhesus macaques for learning experiments; cloth diaper observation
1957–1958	Formal surrogate mother experiments conducted
August 31, 1958	Harlow presents "The Nature of Love" at APA convention, Washington, D.C.
1959	Cloth/wire mother findings published in Science
1960s	Social isolation experiments begin; total-isolation studies with neonates
Late 1960s	Pit of Despair (vertical chamber apparatus) experiments begin
1971	Howard Hughes Medical Institute funding withdrawn; criticism from within scientific community intensifies
1974	Harlow retires from active research
1981	Harlow dies at age 76
1959–present	Attachment theory research lineage continues through Ainsworth, Main, and others

The Controversy: The Pit of Despair and the Ethics of Induced Suffering

The wire-mother and cloth-mother experiments are the part of Harlow's work that most psychologists discuss favorably. They are not the whole of his work.

Beginning in the 1960s, Harlow's research program extended into deliberate induction of psychopathology in monkeys. His goal was to create animal models of human depression that could be used to study the condition and test treatments.

Source: Harry Harlow's Pit of Despair: Depression in Monkeys and Men — PMC

One apparatus he developed was the vertical chamber apparatus, which Harlow himself called the "pit of despair." It was a stainless steel trough with inward-sloping sides, preventing the monkey from climbing out. Infant monkeys — some only a few months old, some up to three years old, all previously socialized with their mothers or peers — were placed alone in the chamber for periods ranging from a few weeks to several months.

Source: Pit of Despair — Wikipedia

Within a few days of isolation, monkeys confined to the chamber stopped moving. They huddled in corners. They refused food. They showed no response to stimuli that would normally produce curiosity or playfulness. When removed from the chamber, the behaviors persisted: the animals were socially withdrawn, could not form normal peer relationships, and showed self-directed stereotypies (repetitive rocking, self-biting, huddling) that resemble behaviors seen in severely neglected human children and in institutionalized individuals with major depressive disorder.

The research produced animal models of depression that were subsequently used in studying the neurochemistry of mood disorders and in evaluating antidepressant drugs. This is the scientific justification Harlow and his team offered.

The ethical objection is not simply that the animals suffered — animal research often involves suffering, and this is regulated and debated — but that the suffering was induced in animals who had already formed social attachments, maximizing psychological harm, with no treatment offered to the affected animals within the experiment. Several of Harlow's former students later distanced themselves from this phase of the research.

Source: Harry Harlow's Pit of Despair — Wiley / Journal of the History of the Behavioral Sciences

One of Harlow's doctoral students later stated that the animal liberation movement in the United States was effectively born from public response to Harlow's experiments. Another described the work as "violating ordinary sensibilities, that anybody with respect for life or people would find offensive." The Howard Hughes Medical Institute, which had funded some of the isolation research, withdrew its support in 1971.

Harlow himself was diagnosed with depression in the late 1960s and underwent electroconvulsive therapy. His biographers have noted that his personal experience with depression may have intensified his interest in creating animal models of the condition — and may account for some of the extremity of the experimental designs.

Replication and Reanalysis: What the Surrogate Mother Findings Built

The cloth-mother findings have been replicated in multiple species and with numerous variations. The core finding — that contact comfort is an independent variable in attachment formation, not simply a conditioned association with feeding — has not been seriously challenged in the subsequent literature.

Source: Harry Harlow Monkey Experiments — Simply Psychology

John Bowlby's attachment theory, published across three volumes between 1969 and 1980, was explicitly grounded in Harlow's primate data alongside Konrad Lorenz's work on imprinting. Bowlby argued that human infants, like other primates, have a biologically programmed behavioral system oriented toward maintaining proximity to a protective caregiver. This system operates independently of hunger and feeding.

Source: 1950s: Harlow, Bowlby, and Ainsworth — Iowa State Press Books

Mary Ainsworth extended this framework with her "Strange Situation" paradigm in the 1960s and 1970s, which classified human infants into attachment styles (secure, anxious-ambivalent, avoidant) based on their response to brief separations from their caregivers. This work would not have been possible without Harlow's prior demonstration that contact comfort, not feeding, was the primary driver of primate attachment.

Subsequent research has extended the findings to:

• Hospital practice: Harlow's results directly contributed to revisions in neonatal intensive care protocols that introduced "kangaroo care" (skin-to-skin contact) for premature infants; meta-analyses have found that kangaroo care reduces neonatal mortality in low-resource settings
• Adoption policy: The findings supported earlier placement of infants in adoptive families rather than extended stays in institutional settings
• Developmental neuroscience: Research on the neurobiology of attachment has identified specific hormones (oxytocin, vasopressin) and brain circuits (amygdala, prefrontal cortex) involved in the formation and maintenance of social bonds

Source: Unveiling Human Attachment: Insights from Harlow's Monkey Experiments — Achology

The isolation experiments also produced lasting influence, though of a different kind. The vertical chamber apparatus is now used as a teaching example in research ethics courses as an instance where the harm inflicted on research subjects exceeded what was scientifically necessary or ethically justifiable. Modern animal research regulations in the United States (the Animal Welfare Act and its amendments) apply specifically to non-human primates and require environmental enrichment, social housing where possible, and review by Institutional Animal Care and Use Committees (IACUCs).

Where It Stands Today: Attachment Science and the Legacy of Contact Comfort

Harlow's 1958 demonstration that contact comfort functions as a primary reinforcer, independent of food, is now a cornerstone of developmental psychology. It is cited in virtually every textbook treatment of attachment theory and is considered one of the most important experiments in twentieth-century psychology.

The parenting advice landscape has shifted completely. Where Watson warned parents against picking up crying infants in 1928, contemporary guidance from the American Academy of Pediatrics emphasizes responsive caregiving — promptly attending to infant distress — as the foundation of secure attachment. The research base for this guidance runs directly through Harlow, Bowlby, and Ainsworth.

Source: Harry Harlow's Monkey Experiments: 3 Important Findings — Positive Psychology

The ethical legacy is more complicated. Harlow's cloth-mother work would clear modern IRB review in modified form — the manipulation (cloth vs. wire surrogate) is relatively mild, the scientific question is well-defined, and the knowledge gained was substantial. The pit of despair experiments, conducted on socialized animals deliberately isolated to induce maximal psychological damage, would face far more stringent scrutiny under contemporary standards and might not receive approval.

This reflects a shift in how research ethics frames the cost-benefit calculation. It is not enough that the knowledge gained is valuable. The harm inflicted must be proportionate to that value, and alternative methods that produce equivalent knowledge at lower cost to the animals are required where they exist.

Definition: Secure base — in attachment theory, the function of an attachment figure as a refuge to which an infant returns when frightened or stressed, and from which the infant feels safe to explore the environment; Harlow's cloth mother provided this function in the absence of food or biological relationship.

Harlow's work also sits permanently in the history of a broader argument about what science owes to its research subjects. The infant monkeys who clung to cloth mothers in 1957 and 1958 contributed to knowledge that has benefited millions of human children — through changed hospital protocols, changed parenting advice, changed adoption practices, and changed understanding of what human infants need. Whether that benefit justifies what was done to them, and especially what was done to the animals in the isolation studies, is a question the field of research ethics has not resolved and continues to debate.

Callout: Harlow's 1958 APA address "The Nature of Love" reported that infant monkeys fed exclusively by wire mothers still spent approximately 17–18 hours per day in contact with the cloth mother that offered no food — directly disproving behaviorism's claim that maternal attachment is a conditioned response to feeding.

FAQs

Q: What did Harry Harlow's wire-mother experiment actually prove? A: It proved that contact comfort — the physical sensation of warmth, softness, and texture — functions as an independent primary reinforcer in primate attachment. Infants did not form attachment bonds with the source of food (the wire mother); they formed attachment bonds with the source of physical comfort (the cloth mother). This directly contradicted the behaviorist model that love is a conditioned response to feeding.

Q: Were Harlow's experiments considered ethical at the time? A: The cloth-mother and wire-mother experiments were conducted before modern animal research oversight structures existed, and they were not considered particularly controversial at the time within mainstream psychology. The isolation experiments and the pit of despair apparatus, conducted in the 1960s and early 1970s, drew significant criticism from both within the scientific community and from early animal rights activists. They contributed to the development of more stringent animal research regulations in the United States.

Q: How did Harlow's work influence John Bowlby's attachment theory? A: Harlow and Bowlby were in direct scientific contact from 1957 onward. Bowlby was developing his argument that infant attachment was a biologically programmed behavioral system, not a learned response to feeding — but he needed primate evidence to counter the behaviorist orthodoxy. Harlow's cloth-mother data provided exactly that evidence. Bowlby cited Harlow's work explicitly across his three-volume Attachment and Loss series (1969, 1973, 1980).

Q: What was the "pit of despair" and why is it ethically controversial? A: The pit of despair was a stainless steel vertical chamber developed by Harlow in the late 1960s to create an animal model of clinical depression. Infant and juvenile monkeys who had formed normal social bonds were placed alone in the chamber for weeks to months. They reliably developed depression-like symptoms: withdrawal, self-directed stereotypies, refusal of food, and social incapacity after removal. The ethical controversy centers on the fact that the animals were deliberately harmed after having formed social attachments, and that no treatment was provided to affected animals within the experiment itself.

Q: Has the contact-comfort finding been replicated in humans? A: Directly controlled experiments isolating contact comfort as a variable cannot ethically be conducted on human infants. However, the finding is strongly supported by convergent evidence: the success of kangaroo care (skin-to-skin contact) for premature infants, the documented harms of institutional rearing with minimal human contact, and the neurobiological research on oxytocin and attachment all point in the same direction. Mary Ainsworth's Strange Situation paradigm provides a structured way to measure attachment in human infants that has been replicated thousands of times across cultures.

Q: What happened to the monkeys raised by wire mothers in the long term? A: Monkeys raised with wire mothers rather than cloth mothers showed lasting social impairments. They had difficulty forming peer relationships, showed elevated rates of aggression and fearfulness, and had significant difficulties with maternal behavior when they later gave birth — often neglecting or abusing their own infants. This generational transmission of social deficits was one of Harlow's own key later findings, and it maps onto research on the intergenerational transmission of attachment insecurity in humans.

---

Read next:

• The Tuskegee Syphilis Study
• The Rorschach Inkblot Test
• The Milgram Obedience Experiment