The Search for Psychology's Lost Boy

In 2009 the decades-old mystery of 'Little Albert' was finally solved. Or was it?

June 2, 2014

First comes fire, followed by the monkey.

The baby on the floor next to the unsmiling woman. The burning newspaper just beyond the baby’s grasp. The tiny monkey on a leash, led by the man in the tie. The parade continues: scraggly dog, rat, rabbit. Through it all, the baby, plump and serene, gazes on with mild interest. No tears, no tantrums.

These words appear on the screen:

"Fear of an animal may be experimentally set up by stimulating the infant with a loud sound just at the moment the animal is presented. Six combined stimulations produced the marked fear of the rat shown next."

In the following scene, the rat returns. The baby cries, attempts to crawl away. The rabbit and the monkey also return, along with a different dog, and the baby cries each time—even without the loud noise. The once-placid infant is now a wailing wreck.

The grainy, black-and-white footage, filmed in 1919 and 1920, documents what has become a classic psychology experiment, described again and again in articles and books. The idea is that the baby was conditioned to be afraid, instilled with a phobia of all things furry.

John Watson's Studies Upon the Behavior of the Human Infant

Go to 7:10 for the sequence that includes Little Albert

The man in the tie is John Watson, the father of behaviorism, a foundational figure in psychology, a Johns Hopkins University researcher who rose from poverty to prominence only to watch his academic career cut short by scandal. When he’s remembered now, it’s often in connection with this experiment, his legacy forever entwined with the baby nicknamed Little Albert.

Photograph from the Johns Hopkins archive

John Watson conducted the Little Albert experiment at the Johns Hopkins U.’s Phipps Psychology Clinic.

The real identity of that baby has long intrigued students of psychology. Who was he? What happened to him? Did Watson really saddle the poor kid with a lifelong terror of animals?

Those questions were long thought to be unanswerable. Watson provided relatively scant biographical detail in his notes on the experiment, and he burned his papers before his death, leaving the curious without much to go on. Then, in 2009, Hall Beck, a professor of psychology at Appalachian State University, published a paper that shed new light on the case. Beck and his fellow researchers had tracked down leads and sifted through clues for a decade before finally arriving at a conclusion.

What they found cast an even darker shadow over Watson’s flawed, ethically dubious experiment. The history of psychology would need to be rewritten—and, indeed, the discovery has already made its way into a number of textbooks. No one would be able to look at the film, or think about Little Albert, in quite the same way again.

That is, unless Beck got it wrong.

Hall Beck’s office is something of a shrine. On top of a filing cabinet is a framed photo of the baby his research identified as Little Albert. On the wall is a promotional poster from one of his talks about Little Albert. On a shelf is a photo of John Watson, taken in his later years, looking sharp and determined. "To My Biggest Fan" reads the dedication, a gag from a student familiar with Beck’s obsession.

The search began with a question that psychology professors have fielded for generations. Several students asked Beck what happened to the baby in the study. The honest answer is: No one knows. On this day, in 2001, Beck said they could try to find out. He didn’t think it was a particularly good idea—a goose chase, probably—but he didn’t want to squelch their enthusiasm. Besides, he thought they would almost certainly forget about it.

They didn’t forget. Instead they pestered him to get started.

There wasn’t much of a document trail. Watson had burned his personal papers because, as he put it, "When you’re dead, you’re all dead." He did provide a handful of clues in the published article, including the first initial of the baby’s last name (Albert B.), but considering that everyone connected with the experiment was probably long gone, the chances of tracking him down were remote.

Beck searched the archives at Johns Hopkins and, judging by memos Watson had sent to the university’s president, he figured that the first phase of the experiment was probably filmed in either late November or early December 1919.

One of the memos Watson sent to the university’s president

According to Watson’s paper, Little Albert was eight months, 26 days old at the time, which meant he was born between March 2 and March 16 of the same year. His mother, Watson had written, was a wet nurse at the Harriet Lane Home for Invalid Children, on the Johns Hopkins campus. Records from that time mention higher-ups at the institution, but nothing about individual wet nurses.

Watson on Albert

"Albert's life was normal: he was healthy from birth and one of the best developed youngsters ever brought to the hospital, weigh- ing twenty-one pounds at nine months of age. He was on the whole stolid and unemotional. His stability was one of the principal reasons for using him as a subject in this test. We felt that we could do him relatively little harm by carrying out such experiments as those outlined below."

From "Conditioned Emotional Reactions," by John Watson and Rosalie Rayner, The Journal of Experimental Psychology, February 1920

Beck then turned to census records. The records show three women listed as "foster mothers" living at Harriet Lane in 1920. Their names were Ethel Carter, Arvilla Merritte, and Pearl Barger.

Ethel Carter was African-American, and Little Albert was clearly white, so they ruled her out. Pearl Barger was intriguing because of her last name. Could Barger be the "B" name they were looking for? After many hours—hundreds, according to Beck—he and the paper’s co-author, Sharman Levinson, an associate professor of psychology at the American University in Paris, found neither evidence that Barger had a child, nor any information about what had happened to her after she left Harriet Lane. The promising lead turned out to be a dead end.

Photograph from the Johns Hopkins archive

Watson wrote that Little Albert’s mother was a wet nurse at the Harriet Lane Home for Invalid Children.

That left Arvilla Merritte. Beck found records indicating that she had given birth to a son on March 9, 1919, which was within the window they had calculated. After more digging, Beck located one of Merritte’s descendants, a grandson named Gary Irons, who confirmed that she had worked at Harriet Lane and had given birth to a son on that date. The boy’s name was Douglas Merritte.

Douglas had the right skin color. His mother had worked as a wet nurse at Harriet Lane. And, crucially, he was born at the right time. Everything seemed to fit. Except for the name.

The Little Albert experiment was conducted long before institutional review boards came along to make sure that subject anonymity was honored. (Any reasonable review board these days would laugh Watson out of the room.) So it would have been fine for Watson to use his subjects’ real names. Why didn’t he just call the baby Douglas M.? If Merritte was the baby in the film, wouldn’t it be known now as the Little Douglas experiment?

Beck mentioned the problem to Charles Brewer, a professor of psychology at Furman University and an authority on the life of Watson. Brewer reminded Beck that Watson’s middle name was Broadus, after a well-known preacher of the time, John Albert Broadus. Maybe calling the baby Albert B. was a wink in the direction of his namesake. It was a guess, but that’s all they had.

Then there was the photographic evidence. Beck went through the Little Albert film frame by frame, searching for the best stills of the baby’s face. Unfortunately, he found no clear close-ups, and the quality of the film—this was shot in 1919, remember—was lousy. Beck did extract a couple of images, though they were far from ideal.

Courtesy of Hall Beck

A comparison of Douglas Merritte (left) and Little Albert, prepared by Hall Beck.

He took those stills, along with baby photos of Merritte provided by Gary Irons, to William Rodriguez, a forensic anthropologist and a former chief deputy medical examiner at the Armed Forces Institute of Pathology, in Washington. Rodriguez determined that while there were "certain facial similarities" between the photos of Merritte and Little Albert, it was impossible to say for sure that they were the same child, in part because of the poor image quality and in part because babies’ faces change so rapidly that such matching is tricky.

That was the bad news. The good news was that Rodriguez didn’t see anything that eliminated the possibility that Merritte was Little Albert. In his professional opinion, "the two photographs could be the same individual."

That left Beck with a pile of evidence, some of which pointed to Merritte, and some of which at least didn’t rule him out. No single fact was a clincher, but there was enough to make a case. Scientists tend to eschew absolute statements, but Beck seemed pretty confident. "The available evidence strongly supports the hypothesis that Douglas Merritte is Little Albert," he wrote.

The last line of the paper was even bolder: "Psychology’s lost boy has come home."

That confidence bothered Russell Powell, a professor of psychology at MacEwan University, in Edmonton. He started reading the paper eagerly—Little Albert had been found!—but finished with a different feeling. The evidence felt too weak to draw such a firm conclusion. He thought the back-of-the-envelope calculation about the odds that another baby would be born at the same time at Harriet Lane was suspect.

Powell published a note in 2011 in the journal History of Psychology summarizing his doubts. He could have let it drop then. Whether Beck and others choose to believe that Douglas Merritte was Little Albert is not, in the end, the most pressing issue in psychology. And one could argue—as has Ben Harris, a professor of psychology at the University of New Hampshire—that attempting to pin down the identity of Little Albert is foolish, a grand hunt for a trivial treasure.

Powell on Beck

"The reality is that in significant ways some of the reported information we have about Albert—such as age at baseline and reported adoption—does not match the recorded facts about Douglas. While it remains possible that Douglas Merritte was Albert, the Little Albert case is far from closed and warrants further investigation."

Russell Powell "Little Albert, Lost or Found: Further Difficulties With the Douglas Merritte Hypothesis," History of Psychology, 2011

Powell did let it drop, for a while anyway. Then, in 2012, his interest was sparked again when Beck, along with Alan Fridlund and William Goldie, as well as Gary Irons, published a follow-up paper suggesting that Little Albert, aka Douglas Merritte, was developmentally delayed. A close examination of the film, they argued, indicates that the baby was more passive and less sophisticated in his movements than a typical baby of his age.

Fridlund, an associate professor of psychology at the University of California at Santa Barbara, had contacted Beck after reading the first paper and watching the Little Albert film. He noticed what he believed were signs that Little Albert, whom Beck had originally deemed "inquisitive," might have suffered from a neurological disorder. They brought in William Goldie, a pediatric neurologist, who often treats developmentally delayed infants. Goldie watched the film "blind"—that is, not knowing that it was the famous Little Albert and not knowing anything about Merritte’s medical history or Fridlund’s suspicions.

Goldie agreed that something seemed off.

CLIP (hover to view)

John Watson’s Studies Upon the Behavior of the Human Infant

Before Watson introduced the loud noise, a small monkey didn’t appear to bother Little Albert.

They observed that the baby in the film is "alarmingly unresponsive"; at no point does the camera catch him smiling. Nor does he engage in so-called gaze monitoring, or watching to see where others are looking. "No evidence is provided of mutual gaze or that Albert sees Watson or is responding to any of Watson’s specific actions," they wrote.

What’s more, when Little Albert reaches for objects. he uses "hand-scooping in lieu of pincer-grasp movements." When you use a thumb and a forefinger to seize an object, that’s a pincer grasp, a developmental milestone that usually happens around nine months—and Little Albert fails to demonstrate it.

All of that combined led the researchers to conclude that "Albert’s temperament and behavior are not within the normal range for his age, and the abnormalities observed on film cannot solely be attributed to the hospital environment or the physical context of filming."

What they had found would require a re-evaluation of Watson’s already checkered legacy.

To believe that Little Albert is Douglas Merritte, you need to believe that Watson either lied or was extremely confused about which baby had taken part in the experiment. He described Little Albert as "healthy from birth and one of the best-developed youngsters ever brought to the hospital." Those words don’t begin to describe Douglas Merritte.

Merritte was very ill. He was hydrocephalic back before the condition could be effectively treated. The fluid accumulation put him repeatedly in the hospital.

We know that Watson was not the most scrupulous researcher. His attention to detail never equaled his ingenuity and ambition. What he wrote in his paper doesn’t always match what appeared in the film. (For instance, the paper mentions the baby nearly falling off a table, though there is no table in the film.) When he would write again about the experiment, years later, some of the details conflicted with the first paper.

Carelessness is one thing. But why would he lie?

Perhaps, Fridlund and Beck speculate, because it would be easier to use a passive child, one who responded in the manner Watson wished him to, and that he then fibbed to cover his tracks. And Merritte had such severe health problems that it makes sense that he would be slower and more passive than an average infant. If that’s true, then Watson’s experiment wasn’t just flawed, it was fraudulent.

It wasn’t long after the publication of the Little Albert paper that Watson’s career in psychology came to a premature end. He was in his professional prime and had plans for more research on human development, including an institution where babies could be raised according to his theories, including his belief that children should not be hugged or kissed, for fear that such shows of affection would lead to emotional trouble when they were adults. Johns Hopkins had recently upped his salary, and students had voted him "handsomest professor" on campus. But Watson, who was married, had been carrying on an affair with Rosalie Rayner, a graduate student and co-author of the Little Albert paper. She is the unsmiling woman in the film.

Watson's Scandal

"Dr. Watson's friends in the university were not prepared, however, for allegations made in Mrs. Watson's suit, tending to show that the celebrated psychologist and author of note has become involved in an affair of the 'affinity' type."

Baltimore Sun, November 1920

His wife found out, and the divorce proceedings were front-page news. "Affectionate Letters Said To Have Been Obtained By Wife," read the headline from one of many articles dissecting the separation. Despite the publicity, Watson never imagined that the affair would force him out of the discipline. His work was too important to be sidetracked by what he considered a relatively minor personal matter.

He was wrong. After what was undoubtedly an unpleasant conversation with the university’s president, Watson scribbled a short resignation letter on university stationery and, just like that, the professor’s academic career was over.

Watson’s resignation letter

Photograph of John Watson from the Johns Hopkins archive

John Watson; at right, the article that ran in The Sun, in Baltimore, about his affair.

Not that he disappeared from public life. Watson reinvented himself as an advertising executive, married Rayner, and became known as an authority on raising children with his 1928 best seller Psychological Care of the Infant and Child. A year before his death, in 1958, the American Psychological Association invited him to a ceremony honoring his work as a "point of departure for continuing lines of fruitful research." It was to be a homecoming, a return to the field that had exiled him decades before.

Watson couldn’t bring himself to attend.

Russell Powell showed the Fridlund paper to Nancy Digdon, a colleague at MacEwan. She was familiar with the Little Albert film and had shown it in her history-of-psychology course. She had never noticed signs that Little Albert might be delayed. Now she watched the film again, keeping an eye out for the evidence that Fridlund and company had cited. She didn’t see anything. It was as if she were watching a different film. What were they talking about?

CLIP (hover to view)

John Watson’s Studies Upon the Behavior of the Human Infant

Digdon and Powell saw evidence that Little Albert used a pincer grasp.

In a scene in which a rabbit is dropped in Little Albert’s lap, Digdon saw evidence of gaze monitoring—a look toward Watson that said, "Hey, what’s going on here?" It appeared to Digdon that the baby was looking for reassurance, just as one would expect from a normal infant. As for the hand movements, Digdon identified a scene in which Little Albert appeared to use the pincer grasp to pick up a marble.

Those discrepancies aside, Digdon and Powell contend that you can’t diagnose a baby on the basis of a few minutes of grainy film. Previous research suggests that such diagnoses tend to result in false positives even under the best of conditions. "If I give them an inadequate sampling of behavior, their expertise can’t compensate for that," Digdon says. "They don’t have the kind of material that they would need to make an informed assessment of that baby."

Photograph by Jason Franson for the Chronicle

Nancy Digdon and Russell Powell

Digdon and Powell thought the evidence for Little Albert’s supposed identity was shaky and the proof of developmental delay nonexistent. But rather than simply poke holes in the work of Beck and Fridlund, they decided to go a step further. "We thought, ‘Wouldn’t it be amazing if we could find Little Albert?’ " says Powell. "What would be the odds?"

Building on Beck’s legwork, they focused first on Pearl Barger. Beck’s paper suggested that every avenue had been explored in an attempt to uncover whether she had given birth. It was unlikely that they could succeed where Beck and his fellow diligent researchers had failed, but given her last name, they couldn’t resist. Powell started by typing "Pearl Barger" and "Albert" into Google.

Lo and behold, there she was.

The website he stumbled across revealed that Pearl had married a man with the last name "Martinek." Powell, deciding that such research was not his forte, turned the work over to a professional genealogist, Christopher Smithson, who in about a week uncovered considerably more information about Pearl Barger, including her obituary in The Sun, in Baltimore. "I found it pretty quick," he says.

Her oldest child, William, was born on March 9, 1919, according to his birth certificate—the same day as Douglas Merritte. William’s death certificate says he was born on March 10, but either date would put him in the running to be Little Albert. Again, though, there was the first name; there was no Little William experiment. William’s middle name, however, was Albert, and that’s what he was called by family and friends. That’s also how he was listed in his medical records.

Albert Barger’s admittance record at the Harriet Lane Home

Could this be Albert B.?

Albert Barger and Douglas Merritte, while they may have been born at the same time, were very different babies. Barger was remarkably healthy, well-developed and thriving—exactly as Watson had claimed.

Perhaps the strongest evidence in favor of Barger is his weight. Watson wrote in his paper that Albert B. weighed 21 pounds in the first phase of the experiment, when he was a few days shy of nine months old, well within the normal range. Indeed, the baby in the film appears well-fed, even plump. Merritte, however, was severely underweight during the first year of his life. At nine months, he weighed 14 pounds, 15 ounces, according to Powell’s review of the records. (Beck says the records show Merritte at nearly 16 pounds.) Barger, at nine months, weighed 21 pounds

I wrote a short article about the Fridlund paper when it was published, two years ago, and I stayed in touch with Beck and Fridlund. I considered writing something more in-depth about the search. The time and effort they had invested in the search, the pains­taking analysis of the Little Albert film, were impressive. They brought a sense of mission to the project. They felt that, by trying to tell this boy’s story, they were giving him the respect he had been denied when he was alive.

When I heard about Powell and Digdon’s findings, which will published in American Psychologist, I was surprised. Was there really serious doubt about whether Beck had found Little Albert? He had made such a strong, if circumstantial, case. Yet the fact that Powell and Digdon had located the son of a wet nurse, a boy named Albert B., who was born at the right time, who was the right weight, was awfully persuasive.

At a coffee shop near the Appalachian State campus, I asked Beck: Do you believe Albert Barger is the baby in the film?

Beck on Merritte

"Standing beside Douglas’ grave, my prevailing feeling was one of loneliness. Douglas never grew up; our search was longer than the child’s life. The quest, which had for so long been a part of my life, was over. I put flowers beside my little friend and said goodbye."

Hall Beck, "Finding Little Albert," The Psychologist, May 2011

"No," he said. "But that’s just my opinion. I don’t think you should put too much weight on my opinion." What if he had found Barger first? Wouldn’t Barger have been his primary candidate rather than Merritte? "I would have explored both directions," he said. "I would have looked up the whole series of things that we know about Albert to see to what degree they lined up. ... I wouldn’t have gone in just one direction."

Later I pressed him on the weight and the name. Knowing what we know now, isn’t it more probable that Little Albert was pretty much who Watson said he was—that is, a healthy baby? Beck countered that there are 28 attributes we know about Little Albert, and that no one attribute, or set of attributes, is decisive. What you have to do, he said, is match up as many as possible and then make an interpretation.

Beck still insists that he is willing to change his mind. "I don’t want to solidify anything in print or give people the impression that I’m not open to receiving new information," he says. He wants to enlist more pediatric neurologists to review the film. His case for Merritte, the boy whose framed photo looks down on him from atop a filing cabinet, now rests largely on the neurological condition of Little Albert. "Is it autism? Is it because he’s visually impaired?," Beck asks. "I think the data is going to come back that the boy in the film is delayed."

CLIP (hover to view)

John Watson’s Studies Upon the Behavior of the Human Infant

Digdon and Powell saw evidence of “gaze monitoring” in Little Albert.

His co-authors, Goldie and Fridlund, say they, too, stand by their conclusions. After seeing an early copy of Digdon and Powell’s paper, Fridlund responded to The Chronicle via email, making it clear that, as far as he was concerned, the new findings changed nothing. "The evidence that Powell et al. produce in their article does not weaken our findings or our conclusion," he writes. He also counters a few of their more specific criticisms, including the question of gaze monitoring: "Albert never references socially during the entire film, regardless of whether he is judged to be ‘looking’ at anyone."

Occam's Razor

"Arguments and counterarguments can be raised for almost any historical possibility, as the prevalence and persistence of conspiracy theories readily demonstrate. Fortunately, Occam’s razor, with its preference for explanations requiring the fewest assumptions, is often an effective antidote in such cases and may be highly applicable in the present case."

From the forthcoming paper "Correcting the Record on Watson, Rayner, and Little Albert" by Russell A. Powell Nancy Digdon Ben Harris Christopher Smithson, to be published in American Psychologist

Digdon and Powell believe it’s over, at least when it comes to whether Little Albert is Douglas Merritte. Albert Barger, they contend, is a much better candidate. They hold open the possibility that a third candidate could emerge, though that would mean that yet another wet nurse, one unaccounted for in the census or other records, had a baby at the same time who somehow fit the criteria even better than Barger—a long shot, but strange things can happen.

Both Digdon and Powell express dismay that the Merritte thesis has made it this far, even into textbooks, given what they see as its myriad shortcomings. For Digdon, continuing to believe that Merritte could be Little Albert is akin to believing in a conspiracy theory. "I understand that if [Beck] quite widely disseminated an idea, it’s hard to back up and reverse it, but that’s called science," she says. "It’s actually a good teaching example of the danger of confirmation bias," the tendency to favor evidence that support your hypothesis., Acroterion;

Douglas Merritte’s grave, in Mt. Airy, Md.

When Beck imagined meeting Little Albert in the flesh. He pictured sitting down with Albert—who would have been in his 80s when Beck started searching for him—and watching the Little Albert video together. How surreal that would have been.

Douglas Merritte died in 1925, at age 6, from the hydrocephaly. According to stories passed down by his family, Merritte never learned to walk and either crawled or had to be carried. It’s unclear whether he ever spoke. He is buried in a cemetery about 45 minutes west of Baltimore.

The other baby, Albert Barger, lived a long life, but not quite long enough for researchers to discover him. He died in 2007, at age 87.

I visited Barger’s niece, Dorothy Parthree, at her home in Baltimore last fall. She and Barger were close, more like father and daughter, perhaps because Barger never had children of his own.

At first Parthree was understandably reluctant to delve into family history with a stranger, but she warmed to the task. She remembered her uncle as a dapper dresser, a guy who liked to take long, aimless drives listening to the radio. He dabbled in sales, drifted between jobs. He moved to Texas, following a girlfriend, then moved back to Baltimore.

Photograph courtesy of Dorothy Partree

William A. Martin (Albert Barger)

He was a bit of a loner. He was married for a while, divorced, though he and his ex-wife remained close. Toward the end of his life, he was in an assisted-living facility, and Parthree would bring him candy and copies of Reader’s Digest.

Parthree produced an envelope with photographs of Barger. There he is, in a suit and tie, during the holidays. There he is, on the couch, posing with the rest of the family. She had also held on to his wallet with his driver’s license and bank card. That envelope contained pretty much everything he’d left behind.

She remembered that Barger didn’t like dogs. When I spoke with her, her four chihuahuas—Rocky, Peppy, Diamond, and Chi Chi—yipped and leapt about. Barger hated that, and she would often sequester them in another room when he came over. His antipathy toward dogs was well known among his friends and family. They teased him about it.

VIDEO (click to play)

Dorothy Parthree discusses her uncle, William Martin.

If Barger was Little Albert, does that mean that the experiment, in a sense, worked? Did Watson instill a lifelong fear of animals in Barger? Did the quixotic quest to locate Little Albert unexpectedly prove, after all these years, that Watson was right?

In their paper, Digdon and Powell consider this, but they caution against overinterpreting Barger’s apparent phobia. "[A]lthough one cannot rule out the possibility that his aversion to dogs and other animals might, at least partially, represent the residual effects of the conditioning procedure to which he had once been exposed, there exists no clear evidence in that regard,"

After spending a couple of hours with Parthree, I drove to the nearby cemetery where Barger is buried. The marker says "William A. Martin," his legal first name and the last name of the man his mother married soon after she left the Harriet Lane home. Like Beck, I imagined meeting Albert in the flesh, showing him the video, watching his reaction. As far as Parthree knows, her uncle was unaware of the experiment and had no idea that his mother had ever been a wet nurse. Pearl Barger, perhaps not wanting her son to know that he was born out of wedlock, must have kept that knowledge from her children.

I asked Parthree what her uncle would have thought about all of this. "Oh," she said, smiling, "he would have been thrilled."

Online presentation by Brian O'Leary, Julia Schmalz, and Tom Bartlett