Of memories lost and found

Interview with Prof. John Hodges, behavioural neurologist.

HOW do people remember and forget? Why do some people forget what happened a minute ago, while remembering vividly what happened decades ago? Understanding how human memory works is a fascinating part of the neuroscientist's endeavour to understand how the human mind works.

Professor John Hodges, an eminent behavioural neurologist, has spent many years understanding why and how people remember and forget.

Memory is not a homogeneous entity. Instead, it is a "number of interactive subsystems in the brain", observed Prof. Hodges. Conscious human memory is of two types: working memory and long-term memory. Long-term memory can be subdivided into episodic memory and semantic memory. Episodic memory relates to our ability to retain personally experienced events or episodes. Semantic memory has to do with our general storehouse of knowledge - words, objects, faces, facts and so on - that we use in our everyday lives.

Some of the most important advances in our understanding of memory have come from the study of patients suffering from Alzheimer's and other brain disorders. In other words, understanding the abnormal has helped in enhancing our understanding of how memory works in normal people.

People suffering from semantic dementia are unable to access their database of knowledge that has been acquired over a period of time. This is particularly manifest in patients' inability to use language. Prof. Hodges pointed out that brain images of such patients revealed severe atrophy of the temporal lobes, particularly on the left side of the brain.

In this interview, Prof. Hodges provides a bird's eye view of developments in his field. He places the modern understanding of memory and its components in the historical context. He advocates a healthy balance between old-fashioned methods of clinical diagnosis and modern brain imaging techniques. Excerpts:

In your lecture on "Remembering and forgetting: the lessons from behavioural neurology" you mentioned the different types of memory. How did this typology develop? Historically, how has the discipline of neurosciences evolved in answering questions relating to memory, its loss and recovery, and the human mind?

The idea of a temporary short-term memory store really goes back to the time of William James, in the 1880s, studying normal memory. In the early 20th century there were studies of memory. The idea that memory is multifaceted started at this time. Traditionally, this meant studying patients who have suffered brain injuries and seeing how memory is affected and what kinds of memory are lost or preserved.

During the 1880s and 1890s, Sergei Korsakoff, a Russian neurologist, studied patients who became amnesic after prolonged alcohol abuse. This rare complication arises in patients who lose their memory because of the deficiency of Vitamin B1, which leads to brain damage. Such patients are unable to lay down new memories and have great difficulty remembering their past.

Korsakoff described deficits in terms of primary and secondary memory, with primary memory being very short-term memory. We now refer to this as working memory. After this there was a period during which the field went through some kind of a dark age when nothing significant was done in the endeavour to understand how memory works.

After the Second World War, during the 1950s, there was a renaissance of interest in memory, which started with the study of patients who had undergone resection of the temporal lobes to treat severe epilepsy. Brenda Milner, a British-born neuropsychologist, who moved to Montreal, was working with a famous neurosurgeon, Wilder Penfield. Together they studied patients who after the removal of the hippocampi experienced a profound loss of memory. Like patients with Korsakaff's syndrome, these patients were unable to form any new memories but their short-term memory was good. They went on to show that the hippocampi were critical structures in the temporal lobe. Brenda Milner went on to complete many studies involving one famous patient known as case "H.M." I suppose this was the beginning of modern neuropsychology.

Important early work was done by a group of neuropsychologists in London, Oxford and Cambridge, which refined many of the findings in amnesia and led to the separation of conscious and unconscious or implicit memory. For instance, Alan Baddeley, who was the director of the MRC Unit in Cambridge, coined the term working memory. Work on semantic memory began with Elizabeth Warrington in London. She found a number of patients whose memory for their own life was okay but who had lost knowledge about vocabulary, words, following viral encephalitis. The terms episodic and semantic memory was introduced by Endel Tulving. He wrote an influential book in 1978, Elements of Episodic Memory. In 1992, I wrote a paper in the journal Brain with Karalyn Patterson in which we described patients with semantic dementia who had degeneration of the temporal lobes of the brain. This paper has been cited more than 1,000 times and has been very influential.

The next era was the one in which brain imaging, particularly functional brain imaging, played a major part. Putting people through a scanner while they underwent various memory tests yielded useful insights into the ways in which memory worked. This began in the 1980s with PET scanning initially, and later Functional Magnetic Resonance Imaging (FMRI).

This must have come as a boon for neuropsychologists. Has FMRI and other techniques been able to make up for what would have been possible if you were able to take out freely brains of dead patients for study without being concerned about ethical issues? What are the ethical issues at play in your community?

The two approaches - studying patients in the tradition of neuropsychology and studying normal people through imaging - are complementary. Brain imaging of normal people can give us insights. But I think there are limits too. For instance, when a patient in a scanner is doing a memory test, lots of different parts of the brain are active; you do not know which of those areas is more important. But if you have damage to each of those areas then you know that some areas are more important than others. So, the two methods are complementary. I believe that the two have to be married. Some people think that we do not need the old methods any more. They say that we can do everything by brain imaging of normal people. I do not think so.

Is this a big issue in your field?

It is a controversy. There has been a decline in neuropsychology in recent years. The number of people studying patients and interested in the traditional approach of making detailed analyses of individual patients has been declining. This is time-consuming because we have to find the right patients and see them many many times to draw conclusions. Imaging is obviously very expensive and time-consuming. But you just have to get ten normal subjects and put them in a scanner. Scanning has been very important, for instance, in understanding ageing. You can compare young and old persons doing a similar task and then draw conclusions using scanning. There is a fusion beginning now. For instance, researchers are scanning patients in the early stages of Alzheimer's while comparing the results with normal subjects.

What are the ethical dilemmas confronting neuroscientists?

First of all there is a problem of access to brain tissue, for instance to find out what causes Parkinson's or Alzheimer's. Obviously, we have to wait for patients to die before we can have them. But there has been a decline in people's willingness to donate.

Why?

First, there is a feeling that it is not necessary any more because we can understand everything by brain scanning and blood tests. This is not true; we still need the actual brain tissue to understand many things.

Secondly, there were a number of scandals in Europe [including the United Kingdom] a few years ago. Brain tissue and other organs had been removed from the dead without proper consent from the families. However, I should point out that this never happened in Cambridge where we were very careful to involve families and to get proper consent before and after the death of patients. There is also the issue of brain transplantation and stem cell research. Most, if not all, neurodegenerative diseases such as Alzheimer's and Parkinson's result in the death of brain cells. Some cases of Parkinson's have been treated by implanting brain cells from human foetuses. This is a great ethical controversy about the use of aborted human tissue. Scientists are trying to find ways for growing human brain cells by taking cells from just one foetus and keeping them in the laboratory, and producing from those cells lots of other clones of the cell for transplantation. In the human brain, cells do not divide. But stem derived from a foetus are primitive and undifferentiated, which can carry on dividing, making different cell lines. The approach has been to try and understand the mechanisms that control stem cells' growth and division so that they can be grown in the laboratory. The idea is to grow cells for Parkinson's, cells for Alzheimer's and so on, and then implant them into the brains of patients. The whole area is quite controversial.

What are the elements of the controversy? Is it driven by religious opposition or is it a secular opposition based on ethics?

Cloning has been very controversial in the U.S. In fact, a very eminent researcher recently moved from the U.S. to Cambridge because of the religious opposition to `messing with nature'. I am not involved with any of this. In the U.S., opposition is mainly on religious grounds. In Europe, the opposition is more to do with human rights. It is to do with the rights of the unborn.

But is not working without stem cells something like working in the dark? Would not stem cell research take your field forward?

I think eventually the breakthroughs in treatment of diseases like Alzheimer's will come from approaches other than stem cell research. The number of transplants and implants in patients suffering from Parkinson's disease is extremely small. It is very time-consuming, costly and technically demanding. I think it is much more likely that there will be breakthroughs in drug treatment.

Are you involved in this?

My role is much more in diagnosis and evaluation of patients. But I am involved in designing drug trials. Many of these diseases [as in the case of Parkinson's and Alzheimer's] occur because there is abnormal protein accumulation in the brain cells. Modern drug approaches are trying to stop this. A number of radical treatments are being developed. There are also immunological approaches, a kind of vaccination against Alzheimer's, which have also been developed in animal models. That is a very exciting development.