In 1986, two memory researchers at Boston University described a single case study that cast a spotlight on retrograde amnesia, an area of research that most previous studies of memory-impaired patients had neglected. The study addressed whether retrograde memory is equally affected across all time periods, or whether information stored decades before the onset of amnesia is more resilient than that stored closer to the disease onset. The researchers’ carefully crafted experiments underscored the importance of having information about a patient’s past knowledge. Using tests created expressly for their patient P.Z., they uncovered the relative sparing of remote memories coupled with the extensive loss of memories closer to the onset of his amnesia.7
P.Z., an eminent scientist and university professor, was diagnosed with alcoholic Korsakoff syndrome in 1981 at age sixty-five. He had both profound anterograde and retrograde amnesia. Because P.Z. had been a prolific writer, the memory researchers who studied him had an excellent grasp of what he had known before he became amnesic. He had written an autobiography before the onset of his brain damage, and when researchers tested his ability to recall events he had described in his book, he performed poorly overall. Intriguingly, his performance was uneven: he was more likely to give the correct answer about events that occurred further back in time. He remembered his early childhood well, but could answer almost nothing about the few years before the advent of his amnesia. As this and other studies have shown, distant long-term memory is less vulnerable than recent long-term memory.
To study this phenomenon in greater detail, the researchers worked up a list of famous scientists—seventy-five researchers whom P.Z. had known personally and in many cases cited in his own work. These scientists had achieved prominence during different time periods. In one experiment, the researchers showed P.Z. the scientists’ names, one at a time, and asked him to identify each scholar’s main field of study and specific scientific contributions. P.Z.’s worst scores were for the colleagues whose careers peaked after 1965, demonstrating that his retrograde amnesia swallowed the fifteen years prior to the onset of his amnesia in 1981. His best scores were for the period before 1965. We do not yet understand why patients like P.Z. experience this pattern of memory loss.
Henry also experienced retrograde amnesia, although it took decades for us to realize the true nature of what he had forgotten. The results of our experiments put him at the center of a scientific debate on the role of the hippocampus in autobiographical memory. From studying Henry’s amnesia in greater detail, particularly his retrograde amnesia, we learned much about how the mind stores and retrieves different kinds of memories. The brain uses separate processes for retrieving personal episodic knowledge, such the morning your teacher promoted you to the top reading group, than it uses for personal semantic knowledge, such as the name of your elementary school. Our studies of Henry over half a century were crucial to this discovery.
At first, Scoville and Milner believed that Henry’s amnesia was fairly straight forward—he could not remember any new information after the surgery, and had lost significant memory from the period immediately prior to his surgery, but had clear recall of things that had happened earlier in his life. In 1957, they reported that Henry had “a partial retrograde amnesia, inasmuch as he did not remember the death of a favorite uncle three years previously, nor anything of the period in hospital, yet could recall some trivial events that had occurred just before his admission to the hospital. His early memories were apparently vivid and intact.” Similarly, in June 1965, a neurologist noted that Henry had a partial deficit in his knowledge of events that had occurred during the year prior to surgery. For example, he consistently confused a vacation he took a month and a half before the operation with one taken two months before the operation. The neurologist further documented memories that Henry retained—events that took place two or more years prior to the surgery, relatives and friends whom he knew preoperatively, and skills and abilities he once had. In 1968, based on information from Scoville’s office and from unstructured interviews with Henry and his mother, we reported no change in Henry’s capacity to recall remote events antedating his operation, such as incidents from his early school years, a high-school girlfriend, or jobs he held in his late teens and early twenties. His memory seemed vague for the two years immediately preceding his operation, performed when he was twenty-seven years old.8
As remote memory testing became more standardized and sophisticated, we realized that our early impressions were incorrect. From 1982 to 1989, my colleagues and I introduced objective tests to probe Henry’s memory for different kinds of preoperative and postoperative information. The first kind was public knowledge—famous tunes (such as “Cruising down the River” and “Yellow Submarine”), information about widely known historical facts (What agency controlled rationing and price support during World War II? In what Latin American country did President Johnson intervene by sending troops?), and famous scenes (Marines raising the United States flag on Iwo Jima; Neil Armstrong on the moon). These tests intermixed items that Henry encountered before his operation with those that occurred after. We found that for public events that occurred during the 1940s through the 1970s—that is, both prior to and after his surgery—he was surprisingly accurate when choosing from four options. For instance, when asked, “When Franklin Roosevelt ran for a third term, his opponent was? . . . ,” Henry correctly recognized Wendell Willkie. When asked, “Which world leaders did President Carter convene with at Camp David?” he correctly identified Begin and Sadat.9
Our test results revealed that Henry, despise his amnesia, could recognize some high-profile historical figures and events encountered after his operation. How do we explain this obvious indication of declarative memory in a man who for all practical purposes did not remember anything? We turned to his lifestyle for answers. He spent considerable time watching television and reading magazines, providing many opportunities to encode information about current events and celebrities. This repetition established representations—symbolic codes of information—in his cortex that were sufficient to allow him, when tested, to say whether he had encountered that person or event before. On the Famous Scenes Multiple Choice Recognition Test, Henry had to select one answer out of three choices—one correct and two incorrect. For example, when he was shown the image of Marines raising the United States flag on Iwo Jima, he chose from among three real-life events—Iwo Jima, South Pacific; Hanoi, Vietnam; and Seoul, Korea. He was also asked to pick one of three dates—1945 (thirty-nine years ago), 1951 (thirty-three years ago), and 1965 (nineteen years ago). When tested this way, with postoperative scenes, Henry often chose correctly. The memory traces, built up gradually through daily exposure to the media, sparked a sense of familiarity sufficient to support Henry’s recognition memory. He was much less successful, however, when the examiner gave him a harder task—recalling the subject and date of the famous scenes (Iwo Jima, 1945) without any cues. A recall test like this is more challenging for all of us than recognition test because we have to dredge up the answers on our own. Henry’s problem was disproportionately greater for postoperative years than the difficulty experienced by healthy participants. Although he scored normally for events from the 1940s, he was impaired for those from the 1950s through the 1980s.10
We also tested Henry’s autobiographical memory. We asked him to relate personally experienced events cued by each of ten common nouns, such as tree, bird, and star. He could choose a memory from any time period in his life. We scored his responses on a scale ranging from zero to three, depending on how specific they were with respect to the time and place of the memory recalled. The participants earned a score of three for a memory that contained a specific autobiographical event that incorporated the stimulus cue (bird), was specific in time and place, and was rich in detail. For example, a participant might have said, “On my twenty-first birthday, I went to Las Vegas and stayed in a hotel that had green and red parrots in the lobby.” They received a score of two for
a memory that contained a specific autobiographical event that incorporated the stimulus cue but lacked specificity of time and place and showed poverty of detail: “I used to watch birds at a lake near my parents’ house.” They earned a score of one for a memory with autobiographical content but lacking the stimulus cue and specificity: “I used to enjoy bird watching.” We gave participants a score of zero for no response or a general statement with no autobiographical reference: “Birds fly around.”11
The results were telling. Henry drew his memories for personal events entirely from the period more than forty-one years prior to testing—when he was sixteen years old, eleven years before his operation. His surgery eradicated the most recent preoperative memories but spared the most distant. These results from the mid 1980s showed that Henry’s retrograde amnesia was temporally limited, but the duration of the deficit was much more extensive than was reported in the 1950s and 1960s. People with anterograde amnesia, including individuals with dementia, often have lengthy retrograde amnesias, and they can remember events from their youth more clearly than events that happened just before the onset of their memory impairment. The shorthand for this phenomenon is “last in, first out.”12
Subsequent advances in the design of remote-memory experiments gave us two new tools. The first, a more sensitive autobiographical memory interview, assessed our participants’ ability to re-experience single events from specific times and places, including the recollection of contextual details. The second, a companion public-events interview, asked about the context of events that occurred in particular places at particular times. We carried out new studies in 2002 to assess Henry’s memory for events in his distant as well as recent past. We began these experiments with a clue from my 1992 interview that he lacked any episodic, autobiographical memory.13
In this interview, I asked him, “What is your favorite memory of your mother?”
“Well, I—that she’s my mother,” he said.
“But can you remember any particular event that was special—like a holiday, Christmas, birthday, Easter?”
“There I have an argument with myself about Christmastime,” he said.
“What about Christmas?”
“Well, ’cause my daddy was from the South, and they didn’t celebrate down there like they do up here—in the North. Like they don’t have trees or anything like that. And, uh, but he came north even though he was born in Louisiana. And I know the name of the town he was born in.”
Henry’s narrative began with Christmas, but as he continued he distracted himself, forgot the question, and ended on a different topic. Over years of interviews, Henry could not supply a single memory of an event that took place with his mother or his father. His answers were always vague and unvarying. When asked about a major holiday, most of us can recount vivid moments, replete with the sensory details that make memories indelible. Henry focused instead on sorting through facts, using his general knowledge about his family and his upbringing to try to construct a response.
This study, a breakthrough in our evaluation of Henry’s memory, showed that his recollections from the time before his operation were sketchier than initially believed. He could conjure memories that relied on general knowledge—for example, that his father was from the south—but could not recall anything that relied on personal experience, such as a specific Christmas gift his father had given him. He retained only the gist of personally experienced events, plain facts, but no recollection of specific episodes.14
In October 1982, we had an excellent opportunity to explore, in a natural environment, Henry’s memories from his preoperative life. I learned that his high school class was having its thirty-fifth reunion at the Marco Polo Restaurant in East Hartford. Neal Cohen, a postdoctoral fellow in my lab, and I obtained permission from the Bickford staff to take Henry out for a night on the town. We drove to Windsor Locks to escort him to the party, and on arrival found Henry all dressed up and eager to go.
The restaurant was crowded, with roughly a hundred people, classmates and their spouses, attending the reunion. Several of Henry’s classmates remembered him and greeted him warmly; one woman even gave him a kiss, which he seemed to relish. As far as we could determine, however, Henry did not recognize anyone by sight or name. He was not alone in this predicament, however. One classmate confided to us that she did not recognize a single person at the event; unlike many of the attendees, she had moved away from Hartford and had not socialized with any of her classmates for many years.
Nor had Henry, of course—so we could not tell how much of his inability to recognize his classmates was a result of having no contact with any of them for thirty-five years, and how much was due to amnesia. Still, even if people’s faces did not look familiar, he should have had glimmers of recognition when he saw their nametags. He might have said, “Danny McCarthy—I remember you from homeroom!” or “Helen Barker—I sat next to you in English, and you helped me with my homework.” But he did not, suggesting that a significant chunk of his high-school memories had been erased.
As we studied Henry over the years, we learned that his deficit was specific to his autobiographical memory—although he was unable to retrieve unique life experiences, he remained capable of recalling public events with considerable clarity. So, for instance, he could talk about the stock market crash in 1929 (when he was three years old), Teddy Roosevelt’s leading the charge in the battle of San Juan Hill, FDR, and World War II. When it came to personal information, however, his deficit was extreme. For example, he could recount a general picture of drives along the scenic Mohawk Trail in Massachusetts with his parents, but could not provide details of a specific event that occurred on a particular trip. He recalled facts but not experiences.
A giant in the cognitive science world, Endel Tulving, provided the theoretical breakthrough that helped us understand the distinction between the kinds of information Henry could and could not retrieve. In 1972, Tulving proposed two major categories of long-term memory: semantic memory, our store of facts, beliefs, and concepts about the world, and episodic memory, the unique events in our personal lives. Semantic memory is not linked to a particular learning experience—for example, I do not know when and where I learned that Paris is the capital of France. Unlike semantic memory, episodic memory records the flow of events in time and allows us to reflect on our mental representations in terms of what the event was, when and where it occurred, and whether it preceded or followed other events. We can vividly remember the details surrounding the phone call we received telling us that we were hired for the job, and today we can re-experience this exclusive event because of our capacity to mentally travel back in time. Henry was unable to make that voyage.15
What about his personal semantic knowledge—his factual memory for the people and places that pervaded his early life? Henry’s old family photos, which Mrs. Herrick had given me, captured happy times—a wedding, catching a big fish, and celebratory family dinners. Here was physical evidence of his personal story. In 1982, I selected thirty-six of these photos for a test of Henry’s childhood memories, interspersing these shots with an equal number of old photographs of my own family. I was not in any of these photos. I made slides of the photographs, projected them onto a screen in the lab one at a time, and asked Henry whether he recognized the people in each picture and when and where the picture was taken. In one photo, Henry and his father were posing in front of a statue of a Native American on the Mohawk Trail; Henry, age twelve, wears shorts, glasses, a white dress shirt, and a tie, and is looking at the camera with his hands behind his back. His father, tall and lean, also wears a dress shirt and a tie, but is in long pants. He strikes a jaunty stance, looking off in the distance with his hands on his hips and one leg in front of him.
“Do you recognize these people?” I asked Henry.
“Well, yeah, one’s me.”
When I asked him which one, he replied, “The smallest one. And the other seems like my father, and it was taken—I think of the
Mohawk Trail right off. There I have a question with myself, though—is that a statue? Well, I know that’s a statue in the background—of an Indian. But I wasn’t sure of that mountain that’s in the background, further back.”
When I asked him when the picture was taken, he said “about ’38, ’39 . . . ’38. I said it right the first time.” Although I do not know the exact date of the picture, his conservator and I guessed that he was about twelve years old, dating the photo to 1938, so his answer was likely correct.
Henry recognized the people in thirty-three of his thirty-six family photos. Equally significant, he did not recognize anyone in my family photographs. Only three of Henry’s family photos evoked no memories for him. One was of distant relatives at a dinner table, and Henry was not in that picture. He said that the little boy looked familiar but did not know his name, where the photo was taken, or when. It is possible that he had not spent a lot of time with that family. In a photo of Mrs. Molaison and Henry celebrating his fiftieth birthday, he recognized his mother but not himself, seen in profile view. Here, Henry’s error is difficult to explain because he did recognize himself in the other photos. The third photo he missed, taken after a hurricane, showed the exterior of his aunt’s house with the roof missing. The house was in Florida, so Henry may never have been there or seen the photo. In every other case, he had specific knowledge of who and what was portrayed in his family photos. I was impressed that he recognized a landmark in one of my family photos (a shot of my mother holding my daughter in Gartford’s Elizabeth Park, with seven ducks at their feet and a lake and trees in the background). He correctly identified the park, which he must have visited growing up in Hartford, and he was able to tease apart what he knew from what he did not know in this photo.
Permanent Present Tense Page 26