Since the days of John Henry, humans have been competing with machines. These days the battle rages metaphorically, as some of us examine the brain as computer metaphor for the limits of its usefulness. And now here is a NYT essay entitled, “Face It, Your Brain is a Computer,” that, as I see it, has missed the issue entirely. Link: http://www.nytimes.com/2015/06/28/opinion/sunday/face-it-your-brain-is-a-computer.html?mabReward=CTM&action=click&pgtype=Homepage®ion=CColumn&module=Recommendation&src=rechp&WT.nav=RecEngine&_r=0
Gary Marcus, a psychologist-neuroscientist, bemoans the loss of focus by some neuroscientists on our brain as a computer (oh, like Gerald Edelman did, who rejected the metaphor outright even as he used them to model functions?) because they miss the big picture and then refutes the arguments against, such as the brain is analog & the computer digital or brains are parallel and computers serial. He leaves out for some reason the old two approaches to study the brain, wet (neurotransmitters, proteins holding memory, hormones etc) vs dry (electricity). Even his organizing trope, that we use machine metaphors to help us understand, is misused, as he says the comparisons to hydraulic pumps, holograms, and steam engines have “failed.” Excuse me, but these are heuristics and heuristics only fail if we fail to learn from or through them. Is he saying that Sigmund Freud and Karl Pribram failed to advance our learning? Of course, Dr. Marcus holds up the computer metaphor as an important heuristic and I can agree it is, so long as we remember that it is an heuristic and that we are seeking to understand the nature of the brain (to see the object whole with all its facets), which happens, as it turns out, to be biological (surprising, yes, I know).
The computer heuristic clouds over an important aspect here, that which Susanne Langer called autogenic action, the organism’s activity, endogenous and autonomous and not contingent upon environmental impacts (Langer’s other class of organic activity). The brain is always active to some degree: neurons fire without input, metabolism continues, the processing of information, real or imagined, current or otherwise, continues. Understanding this is what made Dr. Marder’s essay a few posts back so important to me: that we understand enough about the brain from the perspective of the connectome to understand that it is very “difficult to predict the outcome of a specific pattern of activity or understand the results of a perturbation.” (The perturbation is an impact on the ongoing activity of the brain).
This more than hints at the truth that the organism determines what to make of information, if anything, and organisms function according to their own purpose, if to any purpose at all (besides gene replication). As Langer put it: the environment determines what is given, the organism what is taken [and what is to be made of it all]. Here, I think, John Henry does win.
Turn this around, the brain is a more important heuristic for understanding and developing computers than vice versa, and these tools then help us analyze data and model functions, but they soon fail as an heuristic to understand the actual nature of the brain. Let me end by offering another metaphor, quite apt in its way, though not as useful an heuristic, because it is so very organic and complex in its structure (much like the brain). Our brains are like estuaries, river deltas accrued over the ages with diverse structures and flows. A cat brain is like the delta of the Columbia; the human brain is like the delta of the MIssissippi or even better, Mother Ganges. That conveys the historical complexity of brain evolution and current functioning; when we understand how our brains are like that, computers will be old hat.