The Little Albert Experiment by Watson and Rayner tested whether an infant could be classically conditioned to fear a neutral stimulus.
By pairing a white rat with a loud noise, they induced fear that later generalized to similar furry objects, demonstrating learned emotional responses – but raising major ethical concerns by today’s standards.
Key Takeaways
- Aim: A 1920 study by John B. Watson and Rosalie Rayner aimed to test whether a human infant could learn fear through classical conditioning.
- Method: A white rat was paired several times with a loud, startling noise until 9-month-old Little Albert showed fear of the rat even without the noise.
- Findings: The fear extended to other furry objects – such as a rabbit, a dog, a fur coat, and a Santa Claus mask – demonstrating stimulus generalization.
- Applications: The experiment became a classic teaching example for understanding how phobias might develop and for illustrating principles of classical conditioning.
- Criticism: Later reviews noted methodological weaknesses and raised ethical concerns, including the lack of informed consent and failure to reverse the induced fear.
Little Albert Experiment
-
Can an infant be conditioned to fear an animal that appears simultaneously with a loud, fear-arousing sound?
-
Would such fear transfer to other animals or inanimate objects?
-
How long would such fears persist?
Ivan Pavlov showed that classical conditioning applied to animals. Did it also apply to humans? In a famous (though ethically dubious) experiment, John B. Watson and Rosalie Rayner showed it did.
Conducted at Johns Hopkins University between 1919 and 1920, the experiment aimed to provide experimental evidence for the conditioning of emotional responses in infants.
At the outset, Watson and Rayner worked with a healthy, fearless nine-month-old boy they nicknamed Little Albert.
His exact identity remains uncertain, with researchers debating whether he was Albert Barger or Douglas Merritte.
The researchers obtained permission from Albert’s mother to observe and test him, designing a procedure to see if they could condition a fear response to a previously neutral object.
Participant
Little Albert had been raised from birth in a hospital environment, as his mother worked as a wet nurse at the Harriet Lane Home for Invalid Children.
At nine months old, he was healthy, weighing twenty-one pounds, and was described by Watson and Rayner as “stolid and unemotional.”
This unusually calm temperament, along with his robust health, was one of the main reasons he was chosen as the subject for the study.
His placid nature may also have influenced how he responded during the experiment.
Baseline Session
When Albert was approximately nine months old, the researchers tested his reactions to various neutral stimuli.
He showed no fear of a white rat, a rabbit, a dog, a monkey, masks (with and without hair), cotton wool, or burning newspapers.
The only thing that startled him was the sudden striking of a steel bar behind his head, which caused him to startle and briefly hold his breath.
On the second strike, his lips trembled, and on the third, he cried suddenly. This was the first time a laboratory situation had produced fear in him.
Observations from his mother and hospital staff confirmed that Albert rarely cried and had never shown fear or rage before.
Conditioning Sessions
Two months later, when Albert was just over 11 months old, the conditioning process began.
Over two sessions spaced one week apart, the researchers paired the sight of the white rat with the loud noise a total of seven times.
The noise was produced by striking a steel bar with a hammer behind Albert’s head.
By the end of the second session, Albert reacted to the rat with clear distress – crying and crawling away – even when no noise was present.
Watson and Rayner interpreted this as evidence that a fear response had been conditioned.
By the end of the second conditioning session, when Albert was shown the rat, he reportedly cried and “began to crawl away so rapidly that he was caught with difficulty before reaching the edge of the table” (p. 5). Watson and Rayner interpreted these reactions as evidence of fear conditioning.
Transfer Sessions
In subsequent tests, Albert’s fear response was assessed with other stimuli. When a rabbit was placed in front of him, he leaned away, whimpered, and eventually cried when it touched him.
The dog produced a milder reaction at first, but when it approached his face, he turned away, whimpered, and cried. The fur coat made him withdraw to one side, cry, and attempt to crawl away.
Cotton wool caused initial avoidance, though he eventually played with the paper wrapper while avoiding direct contact.
He was also tested with human hair and a Santa Claus mask, both of which produced negative reactions.
In some cases, he displayed a “conflicting tendency to manipulate” the objects — reaching out to touch but quickly withdrawing.
When moved to a large, well-lit lecture room, his reactions to some stimuli were less pronounced.
In one notable incident, the previously quiet dog barked loudly near his face, startling him into falling over and crying — a reaction that startled the adult observers as well.
Experimental Procedure
| Session | Age | Stimuli Shown |
|---|---|---|
| 1 | 8 months & 26 days | Baseline: rat, rabbit, dog, monkey, masks, cotton wool, burning newspapers (no fear). Loud noise test: startle → lip tremble → crying. |
| 2 | 11 months & 3 days | Conditioning: Rat + loud noise (2 pairings). Startle/whimper. Wooden blocks given between trials to maintain a neutral emotional state. |
| 3 | 11 months & 10 days | Conditioning: Test with rat alone (elicited mild fear). Rat paired with loud noise (5 pairings). Test with rat alone (elicited strong fear). |
| 4 | 11 months & 15 days | Transfer: Rat, rabbit, dog, fur coat, cotton wool, hair, Santa mask. Fear to most stimuli; some manipulation attempts. |
| 5 | 11 months & 20 days | Transfer: In original room: tests with rat, rabbit, and dog; extra rat conditioning; and first direct conditioning of rabbit and dog with noise. In new room: tests with same animals; extra rat conditioning; barking dog incident startled both Albert and adult observers. |
| 6 | 12 months, 21 days | Transfer (one-month follow-up): Santa mask, fur coat, rat, rabbit, and dog. Some weaker reactions (partial extinction), but avoidance persisted. New behaviors included head nodding, shuddering, and thumb-sucking (blocked fear until removed). Albert discharged after this session — no deconditioning attempted. |
Classical Conditioning
-
- Neutral Stimulus (NS): This is a stimulus that, before conditioning, does not naturally bring about the response of interest. In this case, the Neutral Stimulus was the white laboratory rat. Initially, Little Albert had no fear of the rat, he was interested in the rat and wanted to play with it.
-
- Unconditioned Stimulus (US): This is a stimulus that naturally and automatically triggers a response without any learning. In the experiment, the unconditioned stimulus was the loud, frightening noise. This noise was produced by Watson and Rayner striking a steel bar with a hammer behind Albert’s back.
-
- Unconditioned Response (UR): This is the natural response that occurs when the Unconditioned Stimulus is presented. It is unlearned and occurs without previous conditioning. In this case, the Unconditioned Response was Albert’s fear response to the loud noise – crying and showing distress.
-
- Conditioning Process: Watson and Rayner then began the conditioning process. They presented the rat (NS) to Albert, and then, while he was interacting with the rat, they made a loud noise (US). This was done repeatedly, pairing the sight of the rat with the frightening noise. As a result, Albert started associating the rat with the fear he experienced due to the loud noise.
-
- Conditioned Stimulus (CS): After several pairings, the previously Neutral Stimulus (the rat) becomes the conditioned stimulus, as it now elicits the fear response even without the presence of the loud noise.
-
- Conditioned Response (CR): This is the learned response to the previously neutral stimulus, which is now the Conditioned Stimulus. In this case, the Conditioned Response was Albert’s fear of the rat. Even without the loud noise, he became upset and showed signs of fear whenever he saw the rat.
One-Month Follow-Up
Thirty-one days later, Albert’s fear responses were retested. Some were less intense, suggesting partial extinction, but avoidance remained.
New behaviors appeared, such as nodding his head, shuddering, and covering his face with his hands. With the rabbit, he alternated between tentative touching and withdrawing in distress, at one point vocalizing “da da” while leaning away.
He sometimes sucked his thumb during testing — a behavior that appeared to block his fear reaction. Watson and Rayner had to remove his thumb from his mouth to elicit the conditioned responses.
Extinction and Generalization of Fear
In this experiment, a baby who had previously shown no fear of a white rat was conditioned to become afraid of it.
The study also illustrated two important concepts from Pavlov’s work on classical conditioning.
-
- Extinction occurs when a conditioned association gradually weakens and eventually disappears if it is not reinforced. Although the conditioned fear response can be strong at first, repeated exposure to the conditioned stimulus without the unconditioned stimulus will cause the reaction to fade.
-
- Generalization happens when a conditioned response extends to stimuli that are similar to the original one. For example, a child who develops a negative association with one teacher may begin to feel the same way about other teachers, even without direct negative experiences.
In the weeks following conditioning, Little Albert was observed several times. Ten days after the final pairing, his fear of the rat was noticeably weaker, demonstrating extinction in progress.
However, even after a month, some avoidance remained, and the response could be renewed by repeating the original rat-and-noise pairing.
The study ended abruptly when Albert’s mother withdrew him from the hospital on the same day as the last tests.
This meant Watson and Rayner could not carry out their planned deconditioning procedures to remove the learned fear.
Planned Deconditioning
Watson and Rayner intended to explore methods for removing conditioned fear but were unable to do so when Albert left the hospital. They outlined four strategies they would have tried:
- Habituation — repeated exposure to the feared stimulus until the response diminished.
- Tactile reconditioning — pairing the feared object with pleasant tactile stimulation of erogenous zones (lips, nipples, etc.).
- Feeding association — presenting the feared object while giving food.
- Constructive play — encouraging the child to manipulate or play with the feared object through imitation.
Critical Evaluation
Methodological Limitations
The study is often cited as evidence that phobias can develop through classical conditioning. However, critics have questioned whether conditioning actually occurred due to methodological flaws (Powell & Schmaltz, 2022).
The study did not control for pseudoconditioning, meaning the results might not reflect true classical conditioning.
Pseudoconditioning occurs when a subject becomes generally more fearful or reactive to stimuli simply because of repeated exposure to a startling event, rather than forming a specific association between two stimuli.
In Albert’s case, repeated exposure to the loud noise could have made him more wary of any new object, regardless of whether it had been paired with the noise.
This weakens the claim that the white rat specifically caused the fear response. If Albert’s reactions were due to general sensitization, the experiment would not be strong evidence for the conditioning of phobias in humans.
The study failed to control for maturation, making it unclear whether age-related changes influenced Albert’s reactions.
Albert was 11 months old at the start of conditioning and 12 months by the final test.
Developmental psychology shows that fear responses can naturally emerge as infants grow and become more aware of their surroundings.
Without a control group or baseline over time, it’s impossible to rule out that his increased fearfulness was a normal part of development.
If maturation contributed to Albert’s behavior, the results cannot be attributed solely to the conditioning procedure, reducing the study’s validity.
Albert’s reactions were inconsistent, and the conditioned fear appeared weak in later sessions.
The historical record and film analysis show that Albert sometimes interacted with the rat without clear distress and reacted only mildly to other furry stimuli.
These inconsistencies suggest that the conditioning was not robust or long-lasting.
Weak and variable responses limit the study’s usefulness in demonstrating how strong or persistent conditioned fears can be, making the findings less applicable to real-world phobias.
The experimental design confounded conditioning and generalization.
Watson and Rayner used the same stimuli for conditioning and for testing generalization.
For example, the rabbit and dog were both conditioned with noise and later tested as generalized stimuli.
This means it’s unclear whether fear responses to these animals were due to generalization from the rat or direct conditioning.
This reduces the clarity of the results, making it harder to draw firm conclusions about the mechanisms of generalization in humans.
Other methodological criticisms include:
-
- Some doubts exist as to whether or not this fear response was actually a phobia. When Albert was allowed to suck his thumb he showed no response whatsoever. This stimulus made him forget about the loud sound. It took more than 30 times for Watson to finally take Albert’s thumb out to observe a fear response.
-
- Other limitations included no control subject and no objective measurement of the fear response in Little Albert (e.g., the dependent variable was not operationalized).
-
- As this was an experiment of one individual, the findings cannot be generalized to others (e.g., low external validity). Albert had been reared in a hospital environment from birth and he was unusual as he had never been seen to show fear or rage by staff. Therefore, Little Albert may have responded differently in this experiment to how other young children may have, these findings will therefore be unique to him.
Theoretical Limitations
The study ignored the role of cognitive processes in fear responses.
The behavioral model underlying the experiment focuses on observable stimulus–response links, but phobias also involve cognitive elements such as irrational beliefs or selective attention.
Later research, such as Tomarken et al. (1989), showed that phobic individuals tend to overestimate the presence of feared stimuli.
Tomarken et al. (1989) presented a series of slides of snakes and neutral images (e.g., trees) to phobic and non-phobic participants.
The phobics tended to overestimate the number of snake images presented.
By excluding cognitive processes, the study presents an incomplete explanation of phobia development, limiting its theoretical relevance in modern psychology.
The Little Albert Film
Powell and Schmaltz (2022) examined film footage of the study for evidence of conditioning. Clips showed Albert’s reactions during baseline and final transfer tests but not the conditioning trials. Analysis of his reactions did not provide strong evidence of conditioning:
-
- With the rat, Albert was initially indifferent and tried to crawl over it. He only cried when the rat was placed on his hand, likely just startled.
-
- With the rabbit, dog, fur coat, and mask, his reactions could be explained by being startled, innate wariness of looming objects, and other factors. Reactions were inconsistent and mild.
Overall, Albert’s reactions seem well within the normal range for an infant and can be readily explained without conditioning. The footage provides little evidence he acquired conditioned fear.
The belief the film shows conditioning may stem from:
-
- Viewer expectation – titles state conditioning occurred and viewers expect to see it.
-
- A tendency to perceive stronger evidence of conditioning than actually exists.
-
- An ongoing perception of behaviorism as manipulative, making Watson’s conditioning of a “helpless” infant seem plausible.
Rather than an accurate depiction, the film may have been a promotional device for Watson’s research. He hoped to use it to attract funding for a facility to closely study child development.
This could explain anomalies like the lack of conditioning trials and rearrangement of test clips.
Ethical Issues
The Little Albert Experiment was conducted in 1920 before ethical guidelines were established for human experiments in psychology.
When judged by today’s standards, the study has several concerning ethical issues:
-
- There was no informed consent obtained from Albert’s parents. They were misled about the true aims of the research and did not know their child would be intentionally frightened. This represents a lack of transparency and a violation of personal autonomy.
-
- Intentionally inducing a fear response in an infant is concerning from a nonmaleficence perspective, as it involved deliberate psychological harm. The distress exhibited by Albert suggests the conditioning procedure was unethical by today’s standards.
-
- Watson and Rayner did not attempt to decondition or desensitize Albert to the fear response before the study ended abruptly. This meant they did not remove the psychological trauma they had induced, violating the principle of beneficence.
Albert was possibly left in a state of fear, which could have long-lasting developmental effects. Watson also published no follow-up data on Albert’s later emotional development.
- Watson and Rayner did not attempt to decondition or desensitize Albert to the fear response before the study ended abruptly. This meant they did not remove the psychological trauma they had induced, violating the principle of beneficence.
Learning Check
-
- Summarise the process of classical conditioning in Watson and Rayner’s study.
-
- Explain how Watson and Rayner’s methodology is an improvement on Pavlov’s.
-
- What happened during the transfer sessions? What did this demonstrate?
-
- Why is Albert’s reaction to similar furry objects important for the interpretation of the study?
-
- Comment on the ethics of Watson and Rayner’s study.
-
- Support the claim that in ignoring the internal processes of the human mind, behaviorism reduces people to mindless automata (robots).
References
Beck, H. P., Levinson, S., & Irons, G. (2009). Finding Little Albert: A journey to John B. Watson’s
infant laboratory. American Psychologist, 64, 605–614.
Digdon, N., Powell, R. A., & Harris, B. (2014). Little Albert’s alleged neurological persist impairment: Watson, Rayner, and historical revision. History of Psychology, 17, 312–324.
Fridlund, A. J., Beck, H. P., Goldie, W. D., & Irons, G. (2012). Little Albert: A neurologically impaired child. History of Psychology, 15, 1–34.
Griggs, R. A. (2015). Psychology’s lost boy: Will the real Little Albert please stand up? Teaching of Psychology, 42, 14–18.
Harris, B. (1979). Whatever happened to little Albert?. American Psychologist, 34(2), 151.
Harris, B. (2011). Letting go of Little Albert: Disciplinary memory, history, and the uses of myth. Journal of the History of the Behavioral Sciences, 47, 1–17.
Harris, B. (2020). Journals, referees and gatekeepers in the dispute over Little Albert, 2009–2014. History of Psychology, 23, 103–121.
Powell, R. A., Digdon, N., Harris, B., & Smithson, C. (2014). Correcting the record on Watson, Rayner, and Little Albert: Albert Barger as “psychology’s lost boy.” American Psychologist, 69, 600–611.
Powell, R. A., & Schmaltz, R. M. (2021). Did Little Albert actually acquire a conditioned fear of furry animals? What the film evidence tells us. History of Psychology, 24(2), 164.
Todd, J. T. (1994). What psychology has to say about John B. Watson: Classical behaviorism in psychology textbooks. In J. T. Todd & E. K. Morris (Eds.), Modern perspectives on John B. Watson and classical behaviorism (pp. 74–107). Westport, CT: Greenwood Press.
Tomarken, A. J., Mineka, S., & Cook, M. (1989). Fear-relevant selective associations and covariation bias. Journal of Abnormal Psychology, 98 (4), 381.
Watson, J.B. (1913). Psychology as the behaviorist Views It. Psychological Review, 20, 158-177.
Watson, J. B., & Rayner, R. (1920). Conditioned emotional reactions. Journal of Experimental Psychology, 3 (1), 1.
Watson, J. B., & Watson, R. R. (1928). Psychological care of infant and child. New York, NY: Norton.
Further Information
