The Tell-Tale Brain: A Neuroscientist's Quest for What Makes Us Human

17. There has long been a tension in biology between those who advocate a purely functional, or black-box approach, and those who champion reductionism, or understanding how component parts interact to generate complex functions. The two groups are often contemptuous of each other.

     Psychologists often promote black-box functionalism and attack reductionist neuroscience—a syndrome I have dubbed “neuron envy.” The syndrome is partly a legitimate reaction to the fact that most funding from grant-giving agencies tends to be siphoned off, unfairly, by neuroreductionists. Neuroscience also garners the lion’s share of attention from the popular press, partly because people (including scientists) like looking at the results of brain imaging; all those pretty colored dots on pictures of brains. At a recent meeting of the Society for Neuroscience, a colleague approached me to describe an elaborate-brain imaging experiment he had done which used a complex cognitive-perceptual task to explore brain mechanisms. “You will never guess which area of the brain lit up, Dr. Ramachandran,” he said, brimming with enthusiasm. I responded with a sly wink saying, “Was it the anterior cingulate?” The man was astonished, failing to realize that the anterior cingulate lights up on so many of these tasks that the odds were already stacked in my favor, even though I was just guessing.

     But by itself, pure psychology or “black boxology” (which Stuart Sutherland once defined as “the ostentatious display of flow diagrams as a substitute for thought”) is unlikely to generate revolutionary advances in biology, where mapping function onto structure has been the most effective strategy. (And I would consider psychology to be a branch of biology.) I will drive home this point using an analogy from the history of genetics and molecular biology.

     Mendel’s laws of heredity, which established the particulate nature of genes, was an example of the black-box approach. These laws were established by simply studying the patterns of inheritance that resulted from mating different types of pea plants. Mendel derived his laws by simply looking at the surface appearance of hybrids and deducing the existence of genes. But he didn’t know what or where genes were. That became known when Thomas Hunt Morgan zapped the chromosomes of fruit flies with X-rays and found that the heritable changes in appearance that occurred in the flies (mutations) correlated with changes in banding patterns of chromosomes. (This would be analogous to lesion studies in neurology.) This discovery allowed biologists to home in on chromosomes—and the DNA within them—as the carriers of heredity. Which in turn paved the way for decoding DNA’s double helical structure and the genetic code of life. But once the molecular machinery of life was decoded, it not only explained heredity but a great many other previously mysterious biological phenomena as well.

     The key idea came when Crick and Watson saw the analogy between the complementarity of the two strands of DNA and the complementarity between parent and offspring, and recognized that the structural logic of DNA dictates the functional logic of heredity: a high-level phenomenon. That flash of insight gave birth to modern biology. I believe that the same strategy of mapping function onto structure is the key to understanding brain function.

     More relevant to this book is the discovery that damage to the hippocampus leads to anterograde amnesia. This allowed biologists to focus on synapses in the hippocampus, leading to the discovery of LTP (long-term potentiation), the physical basis of memory. Such changes were originally discovered by Eric Kandel in a mollusk named Aplysia.

     In general, the problem with the pure black-box approach (psychology) is that sooner or later you get multiple competing models to explain a small set of phenomena, and the only way to find out which is right is through reductionism—opening the box(es). A second problem is that they very often have an ad hoc “surface level” quality, in that they may partially “explain” a given “high level” or macroscopic phenomenon but don’t explain other macroscopic phenomena and their predictive power is limited. Reductionism, on the other hand, often explains not just the phenomenon in question at a deeper level but often also ends up explaining a number of other phenomena as well.