Remembering Henry M.

The single most famous case study in the history of neuropsychology is that of an anonymous memory-impaired man usually referred to only by the initials H.M. This patient has one of the most severe cases of amnesia ever observed; he has been followed for over 40 years by more than 100 researchers, and is the subject of dozens of research papers and book chapters. The early studies of H.M. provide a basis for modern neuropsychology, and the findings of those who have studied him are today a cornerstone in memory research.

H.M. (sometimes referred to as Henry M.) was born in Hartford, Connecticut, in 1926. His amnesia was the result of neurosurgery performed on him in an effort to alleviate the symptoms of his epilepsy. The origins of H.M.’s epilepsy are unclear. His condition is sometimes attributed to a bicycle accident he had the age of 9 – he sustained a head injury when he was knocked down by someone on a bicycle, and subsequently lost consciousness for about 5 minutes. Soon afterwards, he began suffering from minor epileptic seizures, and had his first major seizure on his 16th birthday. However, H.M. also had a family history of epilepsy – three of his first cousins, from his father’s side of the family, also had the condition. Regardless of the aetiology of his epilepsy, when H.M. reached his twenties, the frequency and severity of his seizures grew progressively worse. By the early 1950s, he was having up to 10 seizures and blackouts a week. The seizures became incapacitating and, as a result, H. M. was unable to continue work as a motor winder.

H.M.’s condition was intractable – he proved to be completely unresponsive to the various anti-convulsant drugs he was prescribed, even when given the maximum dosage. In 1953, at the age of 27, he was referred to William Beecher Scoville, the founder and director of the Department of Neurosurgery at Hartford Hospital. At the time, Scoville was experimenting with surgery as a means of treating psychosis. He localized H.M.’s seizures to the temporal lobe and, on September 1st of that year, performed on him an experimental surgical procedure called a bilateral medial temporal lobe resection. This involved the removal of large portions of the temporal lobe from both hemispheres of the brain. H.M.’s condition justified drastic measures, and the surgery he underwent was extensive and radical. Tissue was removed from the anterior (front) tips of the temporal lobes on the medial (inner) surface of the brain, extending 8cm backwards; the resection therefore completely removed the amygdala, the entorhinal and perirhinal cortices, and about two-thirds of the hippocampus (see diagram below).

Photo Sharing and Video Hosting at Photobucket

The surgery was successful in alleviating the symptoms of H.M.’s epilepsy – afterwards, he had about two seizures a year. But it had been performed as a last resort, and it devastating consequences: H.M. had been left with profound anterograde amnesia and partial retrograde amnesia. To this day, he has no memory of anything that has happened since he underwent surgery, and cannot acquire new factual knowledge about the world around him. He is unable to retain any kind of new information for more than several minutes. He would be seen re-reading the same magazine over and over again, without ever remembering that he had read it before. And he cannot remember much of what happened in the decade prior to his surgery.

Psychologist Brenda Milner has studied H.M. more extensively than any other investigator. Milner worked at the Montreal Neurological Institute the pioneering neurosurgeon Wilder Penfield, who, in the 1930s, had developed the surgical technique on which H.M.’s treatment was based. She used formal memory and intelligence tests to assess the cognitive abilities of patients on whom both Penfield and Scoville had performed surgery. In 1957, four years after H.M.’s treatment, Milner and Scoville published the first written account his treatment and the extent of his amnesia. The paper, published in the Journal of Neurology, Neurosurgery and Psychiatry, also contained accounts of 8 psychotic patients on whom Scoville had carried performed surgery similar to that of H.M.’s.

There has been one striking and totally unexpected behavioural result: a grave loss of recent memory in those cases in which the medial temporal-lobe resection was so extensive as to involve the major portion of the hippocampal complex bilaterally…In [H.M.]…the loss was immediately apparent. After operation this young man could no longer recognize the hospital staff nor find his way to the bathroom, and he seemed to recall nothing of the day-to-day events of his hospital life…he did not remember the death of a favourite uncle three years previously, yet could recall some trivial events that had occurred just before his admission…His early memories were apparently vivid and intact.

H.M. was also unable to recall that his family had moved house or, later, that his father had died. Yet, some aspects of H.M.’s memory remained intact. He could, for example, learn basic motor tasks, but had to be riminded that he knew how to perform them! He could also remember the floor plan of the house he had lived in for about 8 years, and could draw it accurately. But when asked to draw the floor plan after moving house, he would draw a plan of his old residence. And he never recognized Milner, despite meeting with her regularly for testing for over 30 years. He was, she once observed, “chained to the past”.

With the tests she administered, Milner found that, despite his amnesia, H.M. had an above average IQ of 118. Unlike some patients who sustain frontal lobe damage, his social behaviour has always been normal. And, apart from a subtle speech deficit, which may have been present prior to his surgery, or was perhaps caused by some slight damage to the lateral temporal lobe, his capacity for language, including comprehension, knowledge of grammar and memory of words, appeared unaffected. He could, for example, understand the meaning of jokes. Later studies show that H.M.’s memory loss is extremely persistent, and vocabulary and language comprehension tests performed up to the year 2000 show that his capacity for language has remained stable. Milner also showed that H.M. could learn simple sensorimotor skills; his performance on tasks such as tracing a simple diagram improved with practice, suggesting that some learning was taking place. He was therefore able to hold a simple job at a residential home. He could not, however, describe the job after having worked at it for about 6 months.

That the 1957 paper by Scoville and Milner has been cited nearly 1,800 times since its publication is testament to the importance of H.M. to memory research. Milner’s early studies of H.M.’s amnesia revolutionized the field, and changed the way researchers thought about memory. Until then, the ideas of Karl Lashley had dominated the field. Lashley had taught rats to find their way through a maze, then made incisions in their cortices to try and erase the memory trace, or, as he called it, the “engram”. He failed to erase the rats’ memories of the route through the maze, and concluded that memory could not be localized to a single area in the brain.

The case of H.M. conflicted with Lashley’s conclusions, as it showed the importance of a specific part of the brain – the hippocampal formation – in specific types of memory. H.M., and Scoville’s psychotic patients who had undergone similar treatments, showed that the degree of memory loss was dependent on the extent of damage to the hippocampus. In H.M. and another patient, the resection of tissue extending 8 cm. from the front tip of the temporal lobes had caused severe deficits in memory; but other patients who had less tissue removed displayed less severe amnesia, or none at all.

H.M.’s case also supports the dual memory model, as proposed by Atkinson and Schiffrin. According to this model, new information needs to be transferred from short-term to long-term memory (STM and LTM, respectively) if it is to be remembered later on. H.M.’s ability to retain small amounts of verbal information for 15 seconds, or longer with rehearsal, suggests that his STM is generally normal but that he is unable to transfer information into LTM storage. It also supports a model in which there are multiple types of memory for different types of information. For example, H.M.’s ability to recall the floor plan of his house suggests that he retains at least some aspects of his spatial memory, and his inability to remember personal events or the faces of public figures suggested a distinction between declarative memory (that is, memory for factual knowledge) and episodic memory (memory for autobiographical data).

Thus, H.M. appears to be missing those parts of the brain that are crucial for the formation of both declarative and episodic memories, but at least some components of the neural circuitry encoding spatial memories are still present. H.M. is an invaluable source of information for memory researchers. He is now 80 years old, and suffers from osteoporosis, a side effect of phenytoin, the anti-convulsive drug he has been taking. Otherwise, he is in good health. But death is, of course, inevitable, and arrangements have already been made for post-mortem examination of his brain. Undoubtedly, an examination of H.M.’s brain will reveal a great deal more about the anatomy of memory.


Corkin, S. (2002). What’s new with the amnesic patient H.M.? Nat. Rev. Neurosci. 3: 153-160. [Full text]

Scoville, W. B. & Milner, B. (1957). Loss of recent memory after bilateral hippocampal lesions. J. Neurol. Neurosurg. Psychiat. 20: 11-21. [Full text]

A neural implant for epileptics

neuropace_x220.jpgThe Responsive Neurostimulator (RNS) is being developed by NeuroPace, a company based in Mountain View, California. The device, which is smaller than a playing card, is battery-powered and controlled by microprocessors. It is curved so that, like the cochlear implants on which it is modelled, it can fit into a small cavity carved out of the skull. Two electrodes, connected to the device by thin wires, are implanted into the seizure focus (that is, the part of the brain generating the seizures), which is identified beforehand using electroencephalography (EEG) and neuroimaging.

Unlike earlier devices, such as the vagal and trigeminal nerve stimulators, which stimulate the brain continuously, the RNS is designed to do so only when a seizure is imminent. The electrodes monitor the neuronal activity in the part of the brain into which the electrodes are implanted; when abnormal activity is detected, they emit a burst of electrical stimulation that suppresses the imminent seizure by preventing that abnormal activity from spreading through the brain.

The RNS can store data about the electrical activity in patients’ brains. This information can be transmitted wirelessly to a laptop running specialized analytical software. It can be used by a physician to monitor the brain’s activity in real time, and to detect the “signature” of an imminent seizure.

It is unclear how effective the RNS might be in alleviating the symptoms of epilepsy. The efficacy of the device depends largely on the correct identification of the abnormal electrical activity that is characteristic of an imminent seizure, and the generation of the correct response to that activity. A clinical trial involving of 180 patients is now under way. The trial is expected to last up to 3 years, and is being conducted in about 20 centres across the United States.

The RNS would be especially beneficial to those epileptic patients who either experience severe side effects from, or are completely resistant to, anti-convulsive medication. Such patients constitute up to 50% of all epileptics; in some cases, surgery can be performed to remove the seizure focus; in others, however, the focus is located within or near to areas of the brain that cannot be removed. Surgery would not, for example, be performed on a patient whose seizure focus is close to or within the speech centres in the left hemisphere, as damage to or removal of these parts of the brain would lead to severe speech deficits.