Don MacKay: A pioneer at the intersection of language and memory

David Briggs
The research of Nicasio resident Don MacKay, who runs the Language, Emotion and Memory Lab at the University of California, Los Angeles, helped shed light on the complex nature of memory and redefine language as one of many kinds of memory. 

In 1957 a man named Henry Molaison, who became one of psychology’s most famous patients, had his hippocampus removed in an attempt to control his life-threatening seizures. After the surgery, he was unable to form new memories without years or even decades of effort. Many at the time believed the damage was limited to his “episodic memory,” or memories of events. But research by Don MacKay, a professor of psychology at the University of California, Los Angeles, demonstrated that his imagination, language production and language memory were also decimated—at a time when most psychologists did not consider language a type of memory. 

Mr. MacKay has taught at Los Angeles since 1967 and directs the university’s Language, Emotion and Memory Lab; during his career he undertook two-dozen experiments with Mr. Molaison, referred to in scientific literature as H.M. until his death in 2008. Dr. MacKay has lived part-time in Nicasio since the 1980s, when he visited friends and fell in love with the beauty, trails and fresh air of West Marin. (“Every breath is delicious,” he said.) He now lives here full time and is retiring from his work this summer. Mr. MacKay sat down with the Light last week to discuss his experiments with H.M., his research demonstrating that language is its own kind of memory and what we can do to stall memory degradation. 

Point Reyes Light: How did you become interested in psychology?

Don MacKay: I got interested in psychology as a 12-year-old living in northern Canada, in a small town about the size of Point Reyes. It is 400 miles north of Toronto, a beautiful scenic place like this, called Larder Lake. I had a chance meeting with an older friend of my brothers who had come back from Toronto while they were in college. He talked about this exciting revolutionary stuff: Freud! [Laughter.] 1900. The psychopathology of everyday life. That fascinated me as a young person, and I wanted to be a professor from a very young age. I had an uncle who was a professor of physiology, and he was a sort of model. So it was a combination of those two things, probably. 

Point Reyes Light: Your research has specifically focused on language and memory. Can you talk about what drew you to that?

Don MacKay: Many things drew me to language. My two brothers both ended up studying or working with language, and Canada is a linguistically divided country. We were living in Quebec, which was half French and half English, and still is (well, we’re still deciding). But basically language was my central interest when I was a grad student at [the Massachusetts Institute of Technology]. Language was a hot topic, Chomsky was there. I’ve always looked at language from a psychological point of view, as very central to human cognition and memory. It’s built into our genes. We think in terms of language. Language dominates everything. Language can influence how we remember things—but that’s a whole field in itself. 

Point Reyes Light: Can you explain how language memory works in the brain?

Don MacKay:  So take the simplest example: I have learned the phrase “Samantha Kimmey.” How did I do that? I knew the word “Samantha”; “Kimmey” I had never encountered, so I had to connect those syllables: “kim,” “me.” I had to join “Samantha” and “Kimmey” and form a new unit that represented that concept. From that new unit I form links to everything I’ve learned about 

you: where she is from, where she went to college, etc. So that’s the simplest case: forming that new unit and connecting it to other units that are proper names. What happens very quickly when I learn your name as a unit is that the hippocampus is engaged. It’s an activating mechanism that causes extreme activation of “Samantha” and “Kimmey” and that activation over a period of seconds—which is an eternity in the brain—burns a new connection to a new unit or neuron representing that connection. That’s called binding. That new unit has to get used or those connections lose their strength. But if I continue to use that unit, those connections get stronger and it can be a long-term memory.

Point Reyes Light: Can you describe what happened to H.M.?

Don MacKay: At age 26 he had life-threatening epilepsy. He could have had a grand mal seizure and fallen down the stairs and killed himself. He was having several seizures a day. The neurosurgeon decided that the source of these epileptic seizures was in the midbrain area, the hippocampal region, and decided to carefully take out that region. He inserted thin metal tubes over the eyes and into the midbrain—he could tell exactly where they were with X-rays—and by suction he removed about half of the hippocampus and all of the amygdala, which is associated with emotion and fear. That was removed, but the cortex was left intact. That’s what made this operation special from a scientific point of view. 

From a personal point of view, it was quickly obvious that something was very wrong. He could not remember things that he had done just hours or even minutes earlier. So they had accidentally removed the engine of forming new memories. The neurosurgeon published this and put out the word: “Never do this operation again.” It’s never been done again. H.M.’s life after that was devoted to science. He participated in perhaps thousands of experiments. He was well taken care of. He died in 2008 and his brain is now the source of intense study. 

Point Reyes Light:  Are there ways your research affected the way you perceive memory and language?

Don MacKay: My view of memory has totally changed. I started with the view that memory was events. But my research with H.M. has shown it’s much more complex. There is language memory and that is largely independent of event memory. There are many subcategories of language memory. We mentioned my memory for your name: “Sam Kimmey.” That learning and remembering is much more difficult than other categories of remembering, such as someone’s occupation. Any older adult will have more difficulty associating an image of a face with “Mr. Carpenter” than learning the association between a face and “this is a carpenter.” One is a name, the other is an occupation. They’re totally different in your brain. Even for young people it is true: proper names are harder to learn than an occupation. 

The reason is that, with common nouns like “carpenter,” we have a whole set of associations of what a baker does and what a carpenter does. But with a proper name—“this is Mr. Baker”—we don’t have that. It’s just a word, so forming that link is more difficult. 

There are different kinds of memory. There is visual memory that is non-verbal, and then there’s language memory. Words can be as complicated as events are. Episodic or event memory involves three things: when something occurred, where it occurred, and what occurred. Those three things have to get linked together by your hippocampus to form the memory. Or reading—people have no idea of the complex processes involved in reading. We have to create the pronunciation of words, because the letters don’t unambiguously represent the pronunciation of a word. We have to know how to segment the word into syllables and learn what to stress; it has to be built into your memory. It’s not in the letters, although it seems to be, because we do it so quickly. H.M.’s breakdowns in reading demonstrated that. His memory was degraded, and as a result he made these spectacular errors in reading—words like “pedestrian,” he would say “ped-AYE-ee-string.” The second “e” is supposed to be short, not long. He’s got the wrong stress, and those other features that have to be derived from memory. 

Point Reyes Light:  You distinguished language memory from episodic memory, but after H.M. had his operation, both his episodic memory and his language memory was affected. So is there is a connection in the brain between those processes?

Don MacKay: Absolutely—in the fundamental process by which they happen. People in the past thought about episodic memory as just memory and language was totally separated.

Point Reyes Light:  They thought language wasn’t memory at all.

Don MacKay:  Yes. The new way of thinking is that episodic memory and language memory involve fundamentally the same processes, mediated by the hippocampus.

Point Reyes Light: In your recent article in Scientific American Mind, I thought it was interesting when you wrote about asking H.M. what “lentil” meant.

Don MacKay: He said, “Combination word: lent and till, area and time of.” Well, “till” is associated with time, but area? I have no idea where that came from. 

Point Reyes Light:  So that is a demonstration of the breakdown in his language memory processes.

Don MacKay: A very dramatic breakdown. Reading, spelling problems, word-finding problems: those things happen to normal older adults, but nothing like as dramatic as H.M. It’s typically slower, whereas H.M.’s was extremely rapid. There are two processes in memory [degradation]. [One is] the result of non-use; that happens with infrequent words but not frequent words, ones that [H.M.] used on a daily basis. If he hasn’t used it in months, then it’s vulnerable. Aging is the other factor. Degradation happens as we age, and you see it after 65. We can relearn words, but [H.M.] wasn’t normal because he was unable to relearn things. 

I had a tip-of-the-tongue experience this morning. I was trying to remember the name of the editor of the Point Reyes Light in 1982 when we first came here. It was a long time ago. I haven’t used the word recently. I knew I knew the word, and I could picture him, and the last name started with M. I finally got “Mitchell.” And then “Dave” popped into my mind after that. That was aging combined with infrequent use. 

The category of proper names is special. They are more difficult to remember than any other type of word. So after his operation, H.M. couldn’t do the follow-up process when he re-encountered the word. Whether it’s the phrase “Dave Mitchell” or the sight of an experimenter he’s worked with for 40 years, he can’t relearn her name, even though he has heard it hundreds of times.

Point Reyes Light:  How long could H.M. carry on a conversation? 

Don MacKay:  Well, that’s the interesting thing. He had a repertoire of stories he would tell from before his operation. He seemed to be perfectly coherent in telling those stories. He had a well-formed memory representation. It was only when he was forced to talk about something new, and that’s what my experiments did. One set of experiments involved describing a picture. In one, participants—H.M. and the control subjects—had to produce a single grammatical sentence that described a picture and used two or three specific words. One was a picture of a father and two little boys at a stoplight with a “Don’t walk” sign. The words were “first,” “cross” and “before.” You have to use the words to describe the situation. So a normal person, like yourself—what would you say?

Point Reyes Light:  “Boys, first you must look both ways before you cross the street.”

Don MacKay: You passed! Typical older adults, the same age as H.M., would say: “The father is telling his sons to look first before walking across the street.” H.M. said, “Before at first you cross across.” He got all three words, but it’s totally ungrammatical. It’s not a sentence. 

Point Reyes Light:  So what is happening in his mind?

Don MacKay: I am glad you asked that question. These are familiar phrases: “at first,” “you cross.” “Cross” and “across” are closely associated. What he’s doing is he’s producing free associations involving familiar phrases and throwing them together. The associations are based on the words provided. He’s not creating anything new to describe the picture. He’s simply using his old memory for these associations with these words. As a result it seems ungrammatical.

Point Reyes Light:  Could he sense he was doing it incorrectly?

Don MacKay:  That’s what’s really fascinating. He makes these errors and has no sense whatsoever that he’s making errors. 

Point Reyes Light: To him it made sense?

Don MacKay: Well, I think he didn’t have a coherent representation by which to judge it against, whether it makes sense or not. When you make an error, you know it doesn’t make sense because you know what you’re intending to say, and what you said doesn’t match it. But he can’t construct the initial intent in order to realize that he’s making an error. 

I conducted a number of experiments to evaluate his ability to detect errors, like ungrammatical sentences, or identify whether a given word was truly a word or not. He had a problem with low-frequency words—which at one time he knew were words—and now he would say, no, they’re not words. He had a bias to say no, it’s not a real word. His error detection was extremely impaired, whether it was his own errors or experimentally produced errors.

Point Reyes Light: How many experiments did you do with H.M.?

Don MacKay: It was on the order or 20 or 25 experiments on different topics. I did 11 experiments on his language comprehension. I did a number on language production and his reading and word knowledge. 

Point Reyes Light:  Did you have a favorite experiment—one that surprised you or had the most significance to you?

Don MacKay: I think my favorite is this one on language production. It was the most informative. The very first I did as a graduate student when I was 23 is described in the article. It was the most surprising. Everybody believed that H.M.’s language was perfectly normal. And here’s this young graduate student saying, “No it’s not normal. His language comprehension is not normal. His language production is not normal. He’s totally incoherent and has word-retrieval problems.”

They wanted this to be a pure memory case. They didn’t see that there is language memory and that it’s on an equal status with event memory. That’s the new message of my research: memory is very complex and there are many different types. They follow different rules. Language memory follows different rules than visual memory and event memory. But there are certain factors that influence all memory. The big one is—can you guess?—it’s exercise. If you’re an older adult it’s extremely important to get lots of exercise. That will improve every single type of memory. It’s a hot research topic right now. It seems that exercise increases the blood flow to the brain, which strengthens the synapses for all types of memory. The data concludes that exercise is extremely important for older adults to keep their memories intact. I go for a walk every hour and a half. I’m in an aqua-aerobics group. I’m in a hiking group. We talk as we walk along. It has changed my life.

Point Reyes Light: To what extent can we strengthen our memory and to what extent is it inevitable that it will start to degrade?

Don MacKay: We don’t have control over the primary causes of degradation: aging and infrequent use. What we have control over is, for one, relearning. Once we’ve totally forgotten something we can relearn it and it revivifies the memory and slows down those basic causes. The other thing we can control is use. Keep active. Keep trying to remember. Stay socially engaged and listen to intellectual discussion. That active approach keeps your memory strong and prevents degradation. Again, degradation happens for infrequent things. Everyday words—the word “he”—you’re never gonna forget the word “he.” One estimate says you use the word “he” 20 million times in your life. Well, the different forms—“he,” “himself,” “him”—don’t worry about them. Worry about infrequent words.

My wife works on the tip-of-the-tongue phenomenon. So for instance, you get a definition for a low-frequency word: a fastener that’s made out of nylon. You think, “Oh, it’s got two syllables. I know this word. It starts with V.” Other words pop into mind maybe. It increases with aging. She’s shown that trying to recall it improves your ability to recall it later. So if you look it up on an iPhone, you won’t remember it as well. 

Point Reyes Light: Like when you were trying to remember Dave Mitchell’s name.

Don MacKay: Yes. So I didn’t ask you, “Who was the editor back in the 80s?” Now I’m not going to forget it for a very long time.

Point Reyes Light: Why?

Don MacKay:  That is such a new finding that we don’t completely understand it. It will take more research. There are many factors to be looked at before we truly understand it, but it’s a well-established fact. A lot of people all over the world are going to be looking at that.

Point Reyes Light: I was surprised that H.M. didn’t start to panic that he kept forgetting things every few minutes. But maybe it’s related to what you said before, that he didn’t have a reference point for errors. Was he aware that he didn’t have the ability to make long-term memories?

Don MacKay:  H.M. could learn new things through repetition. It’s very slow and inefficient. But through massive exposure to association between “Kennedy” and “death”—it occurred long after his lesion, as he died in ‘63, and his brother died in ‘68—he didn’t know whether it was J.F.K. or his brother, but he had an association between the word “Kennedy” and “death.” So repetition is a primitive way of forming new memory. He is participating in thousands of experiments; he knows the protocol and he’s been told hundreds of times that he is helping science by being in these experiments. He himself said over and over that he’s helping science. He was not frustrated, except on extremely rare occasions. He had a few outbursts. He threw his glasses at his mother. [At that moment, a bird flies into Dr. MacKay’s window, and falls.] That wasn’t staged. That’s not a Hollywood effect.

Point Reyes Light: That was quite strange.

Don MacKay: In any case, he had these rare outbursts like throwing his eyeglasses at his mother. What was that about? We can only speculate. But also I mentioned his amygdala was destroyed so he didn’t have normal emotions of fear and regret. That just left his basic personality, which was very cooperative and wanted to be of help. That’s why he was so popular. So many people wanted to work with him. It made him the most studied patient in the history of the world.

Point Reyes Light:  I know applications aren’t part of your research, but I was wondering if you could talk about the potential applications of what you have uncovered.

Don MacKay: My [lunch friends] urged me to say in interviews that pure research isn’t done with applications foremost in mind. Einstein is an example: he came up with a theory in 1905. It wasn’t tested for almost 20 years. The applications are just now happening, and they’re surprising: no one would have imagined them when he created the theory. So research without a clear application is important. Asking, “What’s the use of this?” has a good response, which was Benjamin Franklin’s: “What’s the use of a newborn baby?” But my research does have some clear implications. We touched on some in my personal life. But for people in general who are my age and older, it’s important to use your memories: that’s what keeps them from decaying. It’s important to keep learning. The use process applies to your hippocampus: use it or lose it. Keep learning. Keep it active. Continue being creative and engaging in social activities and come up with solutions to problems, including problems of expressing new ideas. H.M. illustrates the terrible things that happen if you’re unable to use your memories or recreate your memories. They fall apart. That disaster, in a sense, has a silver lining: it points to the positive side rather than just the negative side.

Point Reyes Light:  If he had never had that operation, would the scientific community know as much about memory as it does today?

Don MacKay:  He revolutionized the study of memory and the brain. That revolution is ongoing. There are still books being published about H.M., there’s still data being analyzed. Our understanding of memory took off with this one case. It impacts all of our theories of the mind.