By Ivan Flis
Every scientific discipline has its famous experiments. The case is no different for psychology. In the company of famous psychological experiments, one study is often mentioned as the fulcrum of the behaviorist revolution of American psychology – the story of a little boy named Albert and the attempt to teach him fear.
The Little Albert experiment has been analyzed through and through and built into the folklore of the discipline – a small boy in a room, being scared by clanging noises and furry animals. Aside from the experiment’s relevance for the discipline’s history, the destiny of the boy experimenters Watson and Rayner named Albert has ever since piqued the interest of many a psychologist. ‘What happened to little Albert after the experiment?’ has become a common question. The ethical norms on psychological research instituted since the 1920s have thoroughly changed and so has our understanding of what qualifies as acceptable psychological research – but is this enough to free us from the nagging feeling that Albert was put into a situation that should have never happened? Are psychologists maybe trying to find absolution, or at least reassurance, in finding what happened to Little Albert?
This makes the Little Albert experiment an extremely interesting piece of history of psychology – if we could find out what happened to the boy after the experiments, could we lay our ethical concerns to rest? Could we place these concerns far away into the historical context of Watson and Rayner’s experiment? Or is the story of little Albert still relevant to the ethical and methodological set-up of psychological experiments today?
Before we discuss the curious historical interest in Albert’s destiny, let’s have a look at Watson and Rayner’s experiment itself.
John B. Watson and Rosalie Rayner wanted to test four hypotheses:
- Can we teach the child to associate the noise produced by striking a steel bar with an animal, thus conditioning Albert to fear the animal as he hears the sound?
- If a fearful response can be conditioned in such a way, can it be transferred to other animals or objects?
- Will the child stop associating the clang with the animals after some time has passed?
- If, after a period of time, the fearful response has not extinguished, what methods could be used to remove it?
To test these hypotheses Watson and Rayner tried to condition fear in a boy they called Albert. At the start of the experiments, little Albert was nine months old and, according to the experimenters, the son of a wet nurse working in the hospital. First, they presented Albert with a number of stimuli to which he had not shown fearful responses (at least as judged by the experimenters): a rat, a rabbit, a dog, and a monkey; and various objects: cotton, masks, and burning newspaper. This was done to ascertain he did not fear any of those things. However, they added one stimulus to which Albert had shown fear: striking a hammer against a steel bar behind his back, making a loud clang.
In the following months the experimenters tried to condition the boy by continuously presenting a rat and clanging the hammer against a steel bar behind him. After several sessions, Albert started reacting fearfully to the rat itself – presumably connecting the loud noise that frightened him with the rat that did not initially evoke fear. Later, Albert was presented with a range of different stimuli. Among them, he showed a fear response to the rat, the rabbit, the dog, and a sealskin coat. 31 days after the last conditioning, Albert still showed fear when presented with the chosen stimuli. Following these last observations, Albert’s mother removed him from the experiment. Watson and Reyner do not explicitly explain why the boy was removed from the experiment, other than stating he left the hospital and that they knew he would leave a month in advance.
The results of the experiment were pushed as proof that fear can be conditioned, and speculated this to be one of the probable causes of phobias. Watson and Rayner’s paper from 1920 concluded with a strong jab directed at Freudians dominating psychopathological theorizing at the time: “Emotional disturbances in adults cannot be traced back to sex alone. They must be retraced along at least three collateral lines — to conditioned and transferred responses set up in infancy and early youth in all three of the fundamental human emotions” (1920, p. 14).
In the end, Watson and Reyner did not answer all of their initial questions (and even the answers they provided were a point of sometimes hotheaded discussions in the following years) but the experiment still left a mark on the collective consciousness of the discipline. Many ethical questions arose – was it justified to teach a child to fear something just for the sake of criticizing Freud’s theories? Interestingly, by asking this question, Watson’s method immediately gained scientific legitimacy because it implies Albert indeed was taught fear. By questioning the ethics, one implicitly acknowledges that Watson’s behavioral conditioning was effective.
The 21st century detective story
Aside from the widely discussed ethical, methodological, and theoretical questions that the Little Albert study has opened among psychologists in the following decades (or maybe because of these), in the last couple of years a number of scholars have (again) started discussing the oft-asked questions in coffee breaks at psychology conferences: what actually happened to Little Albert? Did he suffer long term consequences from the study? What was his life like after his mother took him from the hospital? These questions are of obvious historical interest, but then – do they also have to do with finally putting to rest a disciplinary skeleton in the closet? Did psychological research emotionally scar that boy, or did he grow up completely unscathed by Watson and Rayner’s conditioning?
The most detailed inquiry into the true identity of the child is that of Beck, Levinson and Irons. In 2009 they published a paper under the title Finding Little Albert: A journey to John B. Watson’s infant laboratory. Aside from ignoring Watson’s associate in the title, like most psychologists do, the scholars have told a detailed account of a true detective story, trying to ascertain the identity of a nine month old boy living in the university hospital almost ninety years ago.
They first aspect Beck et al. had to determine was when exactly the study took place. The documents did not sufficiently indicate exact dates. Watson and Reyner were relatively clear on the timeline and how long it took, but not when it actually started. So what our detectives first did was try to specify the possible dates when the study was conducted as precisely as possible – by comparing when the paper was published, when Watson acquired the necessary equipment (he filmed parts of the experiment, and for that he needed a camera!), and when the universities received the journal volume in question. Their conclusion was that “[a]lthough we were unable to establish the month of publication, we found no evidence indicating that Watson did not complete data collection in late March or early April of 1920.” (p. 608).
After setting a loose timeframe, they had to determine how many wet nurses resided in the hospital at the time, and how many of them had children who were then approximately nine months old. The hospital records did not go that far back, so they ventured into quite a bit of (tasking!) reconstruction through the census information from 1920. Out of the three wet nurses, the one who fit the profile the best was Arvilla Merritte and her son Douglas. Arvilla was a wet nurse at the hospital and Douglas was born on March 9. Arvilla was therefore probably lactating (making her suitable for the job of a wet nurse) at the time of the Watson and Rayner’s experiment. Douglas also fit three known attributes of little Albert: he was male, Caucasian and born between March 2 and March 16.
Beck et al. also provide some extra proof going in the direction of at least not excluding Douglas as Albert: comparison of a photo of Douglas to the child in Watson and Rayner video of the experiment, speculation on how and why would Watson come up with the pseudonym Albert B., and a probability analysis of another boy fitting Albert’s profile as much as Douglas did.
Being persuaded by the presented evidence that Douglas is indeed Albert, the investigators reconstructed what happened to Douglas after the left the hospital at Johns Hopkins. The end of the investigation was quite anticlimactic for psychologists waiting for absolution from the disciplinary guilt of Albert suffering long-term consequences from the experiment: “After leaving the Harriet Lane Home, the robust child shown in Watson’s (1923) film became sickly. According to his death certificate (Department of Health, Bureau of Vital Statistics, 1925), Douglas developed hydrocephalus in 1922. Acquired hydrocephalus is often caused by a disease or condition such as encephalitis, meningitis, or a brain tumor (Turkington, 2002).” (p. 613).
Have we finally found out what happened to Albert? And does it matter?
This detective story published in 2009, delving into the life of a little boy born in 1919, spawned quite a little discussion among psychologists. A number of authors provided counter-evidence and interpretations stating that Douglas could not be Albert (Powell, 2010; Reese, 2010), then Beck, Levinson and Irons reasserted their claims (2009b), while other authors joined our primary investigators and went to speculate even more about the validity of the Little Albert experiment, considering that if he was indeed Douglas Merritte, some newly discovered medical records indicated that the infant suffered from “congenital obstructive hydrocephalus, iatrogenic streptococcal meningitis/ventriculitis, and retinal and optic nerve atrophy” (Friduland, Beck, Goldie, and Irons, 2012).
Is our Albert really Douglas Merritte? Was he a robust healthy child, as described by Watson and Rayner, or a neurologically impaired one as the contemporary investigators claim? If he was impaired from the beginning of the study, the Little Albert experiment loses much of its persuasive power in pushing the behaviorist research program – becoming more of a publicity stunt than sound and persuasive research.
There is no conclusive evidence to answer those questions. However, we can still observe how a confluence of ethical, methodological and psychologically substantive discussions woven around an experiment for decades have bred an interest into the destiny of a little boy born to a wet nurse in a university hospital at the beginning of the 20th century. Through the embellished textbook descriptions and retelling of the story throughout psychological literature, the primarily ethical concern asking if Albert suffered long term consequences got grand proportions and is still discussed among psychologists and historians of psychology today.
Did psychologists find absolution in identifying Douglas as Albert? Maybe. The child died only years after the experiments at the Johns Hopkins hospital – he did not live long enough to suffer psychological long term ill effects like phobias. On the other hand, putting forward the argument that he was neurologically impaired from the beginning raises yet another boogeyman: Watson and Rayner consciously deceiving the research community in saying that he was healthy.
Keeping this in mind, contemporary scholarship on Albert’s identity and fate sketches a historical background of questionable practices against which the modern ‘high’ standards of conduct stand in stark contrast. Does this offer absolution? Not so much – but it does offer comfort in the holy progress of science. We have laid the boogeyman to rest by our ethical committees and strict control on human experimentation, and the stories of research barbarity are left to quirky anecdotes from the history of the discipline– seedy mattresses in university hospitals and unscrupulous researchers with cameras are a thing of the past. Maybe psychologists then have not found absolution in Albert’s destiny, but in the newly reinforced myth of progress.
Ivan Flis is a psychology graduate from the Centre for Croatian Studies at the University of Zagreb, and since January 2013 a PhD student in History and Philosophy of Science at the Descartes Centre, Utrecht University. His work focuses on the development of quantitative methodologies in psychology in the post-war period. He also likes to think, write and talk about open access publishing, higher education and research in general.
Beck, H. P., Levinson, S., & Irons, G. (2009a). Finding Little Albert: a journey to John B. Watson’s infant laboratory. The American Psychologist, 64(7), 605–614. doi:10.1037/a0017234
Beck, H. P., Levinson, S., & Irons, G. (2009b). The Evidence Supports Douglas Merritte as Little Albert Hall. The American Psychologist, 64(7), 605–14. doi:10.1037/a0017234
Fridlund, A. J., Beck, H. P., Goldie, W. D., & Irons, G. (2012). Little Albert: A neurologically impaired child. History of Psychology, 15(4), 302–27. doi:10.1037/a0026720
Watson, J., & Rayner, R. (1920). Conditioned emotional reactions. Journal of Experimental Psychology, 3(1), 1–14.
http://b.vimeocdn.com/ts/282/055/282055155_640.jpg, retrieved 19-5-2014