When Does One-Shot Induction Work?

Pierre-Paul Broca was one of the great neurologists of the 19^th century. A witty and affable man, Broca was both universally liked and highly respected. In 1861, he famously encountered M. Leborgne, a 50-year old syphilitic man who had (more than a decade earlier) lost the ability to utter any syllable other than ‘tan’. Five days after Broca examined him, Tan (as he was known to hospital staff) died. Broca’s autopsy revealed a syphilitic lesion in the mid-frontal lobe, a rather large one, as you can see. By legend—Carl Sagan wrote compellingly about this—Broca concluded that the lesion was the cause of the loss of speech.

The classic view of mental functioning was that it was based on general purpose learning mechanisms. When we seem to see, for instance, a person, we are aware (as Hume said) of “a collection of simple ideas, that are united by imagination”—i.e., by the association of ideas, or what we have come to call “conditioning.” Conditioning is general purpose: any two ideas can be associated. Because it is universal, association cannot be localized within the brain; rather it is a principle of brain function as such.

Broca and Tan toppled associationism by demonstrating anatomical specialization in the brain. Or did they? Could a single instance really cast doubt on a theory as well-entrenched as associationism?

Well, the legend is not quite true. For one thing, the localization of brain function was an idea that had been play ever since Franz Joseph Gall in the early 19^th century. For another, Broca was apparently inspired to examine Tan by a lecture delivered a few days earlier by Ernest Aubertin, who challenged the audience to find a single case of speech loss without a frontal lesion. And Broca soon found another case, an 84 year-old man named Lelong. Anatomical specialization may thus have been an idea that had been growing in strength, and given a crucial impetus by a couple of striking cases.

As well, Broca may actually have been wrong. He did not dissect either brain, but was satisfied by the rather obvious lesion. The brains were preserved in a museum. Recent imaging studies have shown (Dronkers et al, Brain 2007) that the damage was much more extensive than Broca had supposed. The anatomical specialization of speech is completely accepted now, but it is now generally accepted that this, and many other abilities, relies on the networked participation of many brain areas.

Still, neuroscience one-shot induction invites philosophical examination.

There are many cases of large sample lesion studies in neuroscience. During the Russo-Japanese war of 1905 a new kind of injury appeared: where a bullet had damaged a portion of the brain, but the patient survived. Many of these patients were found to have selective loss of cognitive function. In the 1950s, similarly, the Montreal Neurological Institute studied hundreds of patients who had suffered the removal of large portions of their frontal lobes to relieve epilepsy and certain psychiatric conditions. (This procedure was regularly conducted in those days.) In these cases, however, the damage was not all in the same precise location. Thus, the conclusions drawn from large samples were not as significant as those that Broca drew from a single case. Indeed, many were wildly off-track—in Montreal it was for a time suspected that the frontal lobes were non-functional. Single sample cases have the advantage of pairing localizable damage and precise loss of function.

In 1956, William Beecher Scoville and Brenda Milner reported on a patient, HM, who was profoundly amnesic after removal of large portions of his temporal lobes. HM was unable to form new memories, but Milner demonstrated that he could learn new motor tasks. Memory was thus shown to be specialized, not general purpose.

In 1983, Josef Zihl and colleagues in Munich reported that a visually acute stroke patient, LM, was unable to see things moving. She could not pour liquids into a cup, because she could not see the cup gradually filling up; she had difficulty following people speaking, because she could not see their lips move; she could not cross a street, because cars seemed far away, then suddenly dangerously close. CT scans discovered damage in area V5 in the visual pathways. Later, LM was discovered to suffer from certain other agnosias.

In the early 1990s, David Milner and Mel Goodale came to know a patient DF, who, subsequent to carbon monoxide asphyxiation, lost much of her ability to visually identify objects, but strangely enough retained her ability to visually control her handling of these same objects. When presented with two blocks, she was unable to distinguish them or to guess at their size. Yet, when asked simply to reach out and pick up different blocks, “the aperture between her index finger and thumb changed systematically with the width of the object as the movement unfolded, just as in normal subjects.” Perceptual vision was separated from vision for action.

One-shot induction has changed our view of the mind, it seems. Readers: Tell us how!