Lonni Sue Johnson doesn’t remember the scientists she’s been working with over the past few years. She doesn’t remember whether or not she was ever married. She doesn’t remember what happened just a few minutes ago.
But the odd things she does remember have had scientists puzzled -- and may shift conventional thinking about memory.
Johnson was an amateur violist, a licensed single-engine airplane pilot, and an accomplished illustrator working for the likes of the New Yorker. But in 2007, viral encephalitis damaged Johnson’s brain and completely wiped out her hippocampus, a brain structure crucial for forming and recalling memories. Now, she seems to have lost most memories of her life and her general knowledge of the world. She can’t recognize something as common as the McDonald’s logo. She looks at Van Gogh’s "The Starry Night," once one of her favorite paintings, and can’t name the painter -- not even if she's given Van Gogh's name in a list of three names.
Yet she can tell you in exquisite detail how to use watercolor and apply different techniques. She knows what a crescendo looks like on sheet music, even if she can no longer recognize the sound of Beethoven's Fifth symphony or the tune of the Happy Birthday song.
Cognitive scientist Barbara Landau and her colleagues at Johns Hopkins University, who have been working with Johnson for a couple of years, picked up on this peculiar pattern of memory. "We started asking her about artistic techniques and she seemed to know things that we wouldn't have expected her to know," Landau said. "It was very striking to all of us how much she knew and how articulate she was."
In a new study of Johnson's case, published online in the journal Cognitive Neuropsychology June 17, Landau and her team examine Johnson’s ability to remember facts related to performing her skills. The case suggests there is a special category of knowledge within the memory system that has previously been mostly neglected.
A new type of memory?
Memory and knowledge are typically broken down into two types: knowing “what” and knowing “how.” The first is called explicit or declarative memory, such as memories of a person's own life and general facts about the world. This is knowledge you are very aware of and can talk about. The other type of memory, non-declarative, is commonly called implicit memory. This is information that’s difficult to verbalize but enables someone to ride a bike or, in case of an artist like Johnson, to bow a viola or apply skilled brushstrokes.
When Landau and her colleagues met Johnson about six years ago, they were struck by how profound and severe her memory loss was. She couldn’t remember much of her past life. She had lost most of her general knowledge, even in areas where she was once an expert.
“Yet we found this really interesting contrast when we spoke to her about how to do things that she knew how to do. She was suddenly very articulate and full of information,” said Michael McCloskey, a cognitive scientist who worked on Johnson's case and a co-author of the new study. “We thought it was quite striking that that category of knowledge seemed to be spared, despite her severe losses of other kinds of general knowledge.”
These findings pointed to a special category of memory: One that is explicit, dealing with facts, but is closely related to skills -- in other words, a type of “skill-related knowledge.” And this category may be treated differently inside the brain, the scientists thought.
So they decided to test Johnson’s skill-related memory. The team, which included researchers Emma Gregory and Zoe Ovans, tested Johnson's memory for facts related to performing her top skills -- art, music, flying and driving.
For each test, scientists asked her about 80 questions on techniques and equipment in each field. For example, “How is a crescendo represented in sheet music?” or “What three major steps must you complete in order to get a pilot’s license?”
They then compared Johnson’s results with those of other people taking the same tests. In art and driving, Johnson scored nearly as high as people with the same level of expertise as hers. In music and aviation, she performed worse than the experts but significantly better than amateurs.
“We found that although she had losses in these areas, they were much more spared than all the other sorts of general knowledge,” McCloskey said.
Where memories live in the brain
In many ways, Johnson's case resembles that of Henry Molaison, a man who had his hippocampus removed in 1953 to relieve him of debilitating seizures. After the surgery, Molaison’s seizures stopped, but he was no longer able to form new memories. He became an icon in neuroscience history by participating in follow-up studies for the rest of his life, which taught researchers about not only the importance of the hippocampus but also the nature of memory itself.
Molaison lived in the now. Doctors had to reintroduce themselves to him every time they met. But despite his inability to register new information, Molaison showed signs of learning. For example, if he practiced a drawing test, he did better next time -- even though he had no memory of having ever done the test. This suggested that there’s a kind of implicit memory, a “muscle memory,” that doesn’t rely on the hippocampus to be registered by the brain.
It's surprising how little we know about where and how knowledge or memory is stored in the brain. Michael McCloskey, cognitive scientist, Johns Hopkins University
Molaison also showed a peculiar pattern when it came to past memories. He retained a fair amount of his general knowledge about the world and his childhood, but he had hazy memory of events that occurred closer to the time of his surgery. This pattern, again, raised a question: were his long-term memories stored somewhere outside his hippocampus?
There are different views about how memories are encoded. One theory holds that the hippocampus is crucial when you first encode the memory, but over time these memories become independent of the hippocampus and go on to live elsewhere in the brain. Another view, however, holds that even if memories are not stored in the hippocampus, the hippocampus remains permanently crucial for retrieving them. In other words, it acts like a gate, allowing memories to come in and get out.
“It's surprising how little we know about where and how knowledge or memory is stored in the brain,” McCloskey said. “We can point to the areas of the brain that are important for memory like the hippocampus. But there's still a lot of fundamental disagreement [among scientists] about where our long-term knowledge is stored, how it's stored and how we remember it.”
And Johnson’s case has raised more questions than answers.
If there’s this subdivision of skill-related knowledge within explicit memory, where is it in the brain? Why doesn’t it need the hippocampus like the rest of explicit memory?
“One possibility is that this kind of knowledge is stored in a separate place or in a separate way. Another possibility is that it was more durable in the face of brain damage, given that maybe it was highly connected with other things and learned over a long period of time,” McCloskey said. “That's just guesswork at this point.”
While it may be too soon to rewrite memory textbooks, the scientists suggest that researchers should also pay attention to skill-related knowledge as a possible subcategory of declarative knowledge.
“This type of knowledge is not the skill itself; it's not the ability to bow the violin. It's the knowledge that goes along with those skills, McCloskey said. “When you learn to play tennis or music, you also need information beyond just muscle memory. We are suggesting we may need a reorientation in the way we think about declarative knowledge versus skill and recognize that skill is not out there by itself.”