Two very different articles I read recently made me think – both essentially question the fundamental approaches of current neuroscience. One is a preprint on biorxiv by Eric Jonas and Konrad Kording called ‘Could a neuroscientist understand a microprocessor?‘, and the other is an essay by Robert Epstein entitled ‘The empty brain‘.
The first paper‘s title is a reminder of another classic: ‘Can a biologist fix a radio?‘ by Yuri Lazebnik, which is both instructive and entertaining (‘Indeed, the insides of even my simple radio would overwhelm an average biologist (this notion has been proven experimentally)’). Jonas and Kording apply established neuroscientific techniques such as lesioning, measuring tuning properties, recording local field potentials and ‘whole-brain imaging’ to understand a microprocessor. They conclude that while they generate lots of data and some knowledge, the resulting understanding is fairly superficial. (Digression: What does it mean to understand the brain? And how can we know before we do?) If you want to know more about it but cannot be bothered to read their paper, there is an excellent article by Ed Yong in the Atlantic, where you can also see in the comment section that the paper started quite some debate (all the more reason for preprints!)
Epstein’s essay is even more scorching – he critizises the pervasive use of computer jargon to describe the brain (think of memory storage, encoding, algorithms). IT speak seems completely normal as well as state of the art, but he puts it into historical perspective: the brain has always been described in analogy to the most recent technological breakthrough, whether that was hydraulics or a computer. While analogies can be useful, maybe it should make us think that we seem unable to describe the brain in different terms.
While all of these articles are controversial (just check out the comments sections), I really enjoyed both of them as they made me think deeply about my assumptions and as well as the validity of my approaches. Can you describe what the brain does without using any computer metaphor? I don’t think I can (yet).
2 thoughts on “Neuroscientists – take this!”
In the memory section, yes it is. But in the emotion section, still few computer metaphor.
“Reading what Machines Think”
The full history …