Controversy Around Patient H.M.

The Letter  |  The Burning  |  The Lesion  |  The Shredding  |  A Question of Consent

For many decades prior to the publication of my book, a small group of researchers controlled Henry Molaison’s story, much like they controlled Henry Molaison himself. Henry, according to the tale they told, was an ever-willing participant to a half-century of gentle experimentation overseen by a host of selfless, compassionate, and ethical neuroscientists. My book pushed back against that story in a variety of ways. It’s not surprising, then, that some in the neuroscience community pushed back against my book. Below, you’ll find summaries of each their major criticisms, as well as my responses.

First, though, I’d like to bring up one of the book’s genuine shortcomings. The hardcover edition did not include endnotes. It should have. As a reviewer at The New York Times put it, “any book with neuroscience this complex and content this provocative really needs footnotes. This book has none. Zero. And that’s just weird.” I took her point to heart, which is why, if you purchase the paperback edition, you’ll find it includes an extensive new endnotes section.

Also, if you haven’t read the book yet, I’d suggest you wait to read these until afterwards, since they contain what could be described as spoilers.

A few days before the launch of Patient H.M., the Massachusetts Institute of Technology sent a letter to the editor of The New York Times Magazine, which had just published an excerpt of the book. The letter, which was signed by more than 200 neuroscientists, stated that I had described Henry Molaison’s experiences as a research subject at MIT in “what we believe are biased and misleading ways.” The letter took particular issue with my depiction of a recently deceased MIT scientist named Suzanne Corkin, who had been the principal investigator in Henry’s case for more than four decades. What my book revealed about Corkin, the letter stated, was “contrary to everything we have known about her.”

What the letter did not do was address any of my book’s specific claims. Because of that, I don’t have a specific rebuttal to the letter. What I can say is that I’m pretty sure only one of the 200-plus signatories had actually read the book. I know this because that person had previously written me a series of very complimentary emails, in which she told me she’d received an advance copy of the book, which she described as “magnificent.” Evidently she found it difficult to maintain this opinion once her peers in the neuroscience community became upset. Closing ranks in the face of criticism is an understandable and very human reaction. It is also, I have to say, a very unscientific one.

Read the Letter to the Editor of the New York Times Magazine

- by International Community of Scientists | August 5, 2016

Among the hundreds of experiments Henry Molaison took part in were ones that tested his ability to withstand pain. As I described in the book, researchers would tell Henry to hold a heat-producing device called a dolorimeter against his skin—sometimes his chest, sometimes his arms—and then measure how many seconds elapsed before the pain became too much for him to bear. They found that Henry was highly pain resistant, and that, unlike the control subjects, “H.M. did not label any stimuli painful, no matter how intense they were.” In fact, I wrote, he would leave the device on for so long that “his skin began to burn and turn red.”

John Gabrieli and Howard Eichenbaum, memory researchers employed by MIT and Boston University, respectively, strongly objected to my account of the pain experiments conducted with Henry. (Eichenbaum was himself one of the researchers involved with those experiments.) Here’s what Gabrieli and Eichenbaum wrote, in their criticism of me: “In the experiment on thermal pain perception, the heat level was never raised to the level of skin burning. This inflammatory statement is a lie that could only be used for effect on the readers. Also, the experiment used a standard procedure, done on controls and pain patients and the redness of the skin simply indicates that the skin is warm, not burned.”

My source for the description of Henry’s skin was Nancy Hebben, the lead author of the pain study. Here’s what she told me, on tape and on the record, in 2015, when I asked her about the experiment: “One thing I remember is after it was all over—I think we did it several times just to make sure that it was accurate—but one of the times, afterwards, he went up to the nurse at the desk at the CRC [MIT’s Clinical Research Center] and he, because he left it on so long he basically had, like, twelve sunburns, quarter-size sunburns, on his arms. And he actually went up to the nurse, and, even though it had been painful, he said, ‘Where did I get these?’ He had no memory whatsoever of having done that.” Several years prior to my interview with her, Hebben made the same point, in writing: “H.M. also underwent my pain perception task and had an increased tolerance to pain,” she wrote. “Remarkably, he left the heat gun on so long he had 12 quarter size ‘sunburn-like’ circles of red, six on each arm, and shortly after the study ended, he went up to a charge nurse at the CRC, asking how he had gotten the marks on his arms.”

In other words, the dolorimeter remained on Henry even as his skin began to burn and turn red.

My book describes a dispute that arose when Jacopo Annese attempted to report, in a journal article, a previously unreported lesion he’d discovered in Henry Molaison’s frontal lobes. The book details how, after Corkin and her colleagues were retroactively added as co-authors of the journal article, they attempted to edit out all references to the lesion, arguing that it was probably just a postmortem “handling lesion.” The dispute lasted for several months, until Corkin and her colleagues relented and agreed to allow the lesion to remain in the article, though in a somewhat neutered form.

MIT has pushed back at my account, declaring that “there was no attempt to suppress.” In what strikes me as a strange logical leap, the university has pointed to the lesion’s presence in the published article as evidence that there was never an attempt to keep it out. However, the record clearly shows that Corkin and her colleagues did attempt to excise the new lesion before they agreed to allow it to remain. I won’t speculate as to why they did so—I never actually used the word ‘suppress’—and will leave readers to draw their own conclusions, but because this has been a controversial part of the book, I’ll present some supplemental information here.

First of all, MIT, in its criticism of me, has downplayed the significance of the lesion, arguing that I, as a journalist and not a neuroscientist, “misunderstood” and overstated its significance. I never claim to be a neuroscientist, but as I note in the book, many neuroscientists would indeed view an unreported frontal lesion in Patient H.M. as significant. This includes one of the anonymous peer reviewers of the paper in question. The book quotes some of what that neuroscientist wrote, but here’s the complete version of what he or she said about the frontal lesion: “Of particular importance is that these investigators also show that there was some damage in the inferior polar region of the frontal lobe. Much of the neuropsychological literature on H.M. has made the case that so-called frontal lobe function was intact, since H.M. could perform normally on frontal executive tasks that did not place heavy demands on working memory. Such conclusions were often invoked in neuropsychological reports of H.M. to specify that the memory impairment could not have a frontal executive component. However, Annese and colleagues unequivocally demonstrate that frontal white matter was affected, likely by the surgical approach. Their future investigations that will examine detailed histology in this region, will be of great importance, because if there is significant white matter pathology with in frontal systems, that may have implications for a retrospective reinterpretation of H.M.’s neuropsychological findings.”

As for my assertion that an attempt was made to remove the lesion from Annese’s paper, my source for that is Suzanne Corkin’s own revisions to a draft of the paper, in which she excised all references to the frontal lesion. She even included a note explaining her reasoning: “I removed discussion of Scoville’s surgical procedure since it is of no meaningful scientific contribution. Also, the frontal lobe lesion does not appear on either the in situ scans or the fresh brain photos. Without microscopic evidence of a pre-existing lesion, any consideration of it would be highly misleading.”

Annese eventually presented Corkin and her colleagues with evidence proving that the lesion did in fact exist prior to Henry’s death, and Corkin and the other MIT researchers eventually agreed to allow the lesion to remain in the paper. They continued to argue, however, that Annese was giving it “disproportionate attention” and “over-hyping” it, and suggested that Annese not “highlight it in any major way other than to mention it in passing.” The lesion remains in the final version of the paper, which you can read here, though its description was, as I said, somewhat neutered.

All available evidence suggests that my account of the dispute over the unreported lesion was fair and accurate.

On October 28, 2015, during an interview in her office at MIT, Suzanne Corkin told me that she had shredded much of Henry Molaison’s research data, and planned to shred more in the future. (Sadly, Corkin passed away in May of the following year.) I reported Corkin’s remarks in my book, which prompted MIT to quickly issue a statement objecting to what the university described as my “allegation that research records were or would be destroyed or shredded.” I wasn’t surprised by the pushback—after all, the destruction of research records is a serious matter—but the nature of MIT’s defense always struck me as odd. After all, I wasn’t the person making the allegation that research records were or would be destroyed. Corkin, the person in charge of those research records, was, and I was simply reporting, accurately, what she said. If you’d like to hear a recording of Corkin speaking about the shredding, you can find it here.

The head of the Brain and Cognitive Sciences Department at MIT—James DiCarlo—eventually released a second statement saying that he had checked on the status of Henry’s research data and concluded that the data was “maintained and not destroyed.” If this is true, it means that Corkin was not telling the truth when she spoke with me. If it is not true, it means that Corkin was telling the truth. Either scenario is troubling.

I have an update on the shredding matter: After DiCarlo announced that no files had been shredded, Donald MacKay, a UCLA cognitive psychologist, wrote DiCarlo and said that he was happy to learn “that no records from the MIT H.M. Project were actually destroyed.” Then he continued: “On the assumption that MIT can now make those records available to competent researchers, this note comes to request two specific categories of data that almost certainly exist in Professor Corkin’s HM files,” specifically some of Henry’s crossword-puzzle books and his Beck Depression Inventory questionnaires. DiCarlo responded by saying that he would look into the records request and get back to him soon. As of now, many, many months later, MacKay still has not received any of the data he requested.

If MIT wishes to lay this question to rest, they should do something they have been unwilling to do: Release whatever is left of Henry Molaison’s research data, and make it publicly accessible.

Audio Interview with Suzanne Corkin

Around 300 B.C., in Alexandria, Egypt, the pioneering anatomists Herophilus and Erasistratus undertook a series of experimental human dissections. Most of the people they dissected were dead, but some were not. As a Greek historian later wrote, “they laid open men whilst alive—criminals received out prison from the kings—and whilst these were still breathing, observed parts which beforehand nature had concealed.” A couple thousand years later, a British physician named Edward Jenner repeatedly injected his gardener’s young son with the deadly smallpox virus, hoping that the virus would be defeated by a home-grown vaccine he’d developed. In 1951, a woman named Henrietta Lacks, dying of cervical cancer, had some cells removed from her uterus without her knowledge, and those cells eventually multiplied and propagated into the immortal HeLa line, which is still used in laboratories around the world.

All of these cases, different as they were in details and degrees of horror, had at least one thing in common: each involved human research subjects who gave no meaningful informed consent. That is, they either didn’t know they were being subjected to experiments, or they were in a vulnerable position that made true consent impossible. The concept of informed consent has evolved over the years—its modern roots can be traced to the Nuremberg Trials, which exposed the experimental brutality perpetrated by Nazi doctors in concentration camps—but today it’s recognized as an essential cornerstone of all work involving human research subjects. Without informed consent, history makes clear, terrible things can happen.

While researching my book about Henry Molaison, the most important human research subject of all time, I was startled to find evidence that that the principle of informed consent was violated in his case, too. The Massachusetts Institute of Technology oversaw what it called “An MIT Research Project entitled ‘The Amnesic Patient H.M.’” during the last four decades of Henry’s life, and Henry would spend as long as a month every year living at the university’s Clinical Research Center, where he participated in research ranging from dream studies to tests of his ability to withstand pain. During much of this period, Henry himself was the only one signing his informed consent forms. That’s difficult to justify, since not only would he forget the beginning of a single-page document by the time he reached its end, but the lesions in Henry’s brain also made him much more docile, neurologically predisposed to consent to almost anything. 

Eventually, MIT did recognize that Henry’s deficits, “raised the nagging question of how we obtained informed consent,” and the university arranged for a man who claimed to be Henry’s closest living next of kin to apply to become his conservator. For the last sixteen years of Henry’s life, this man signed Henry’s informed consent forms, and also agreed in advance to donate Henry’s brain to MIT after his death. The problem is, Henry’s conservator was not in fact Henry’s next of kin, and MIT never appears to have made an effort to find Henry’s real closest living relatives, as I discovered when I myself tracked them down.

Interestingly, none of the statements issued by MIT in the wake of my book even contain the word “consent.” The closest the university came to addressing the matter was in a statement which declared that, although “Dittrich identifies some individuals who were genetically closer to Mr. Molaison” than his conservator, there was nevertheless “nothing untoward” about the conservatorship process. When a research institution steps so gingerly around the question of consent, it’s hard not to wonder whether this is because it’s a question for which they don’t have any good answers.

Buy the Book