img: Kaan Durmuş
in this episode we find out how motivational talk makes you run faster & i ponder on the history of science & the death of ideas long, long after their time

episode #38, 5th April 2024 #progress #idea #kuhn #death #revolution   image source

  ⇦ listen here ⏱15:14

  ⇨ transcript here

our main brain stories this episode...


the media claims that a study of football coaching motivation shows that a mere instruction could be the difference between scoring a goal or being tackled. it doesn't, it didn't, & the original study was careful to say so.   [read more...]

this story was in episode #38. original article: Moran et al., 2023 (Journal of Sports Sciences), reported in: The Times by Rhys Blakely on 3rd March 2024 #football #motivation #coaching #run #jump



for the avoidance of doubt, the provocative title of this episode relates to scientific ideas & not to scientists.

i've been thinking a lot about death recently. not just of people, but of scientific theories & ideas too. in this third essay, i wonder whether, as scientists doing science at universities, we're really making any real progress at all, or if this is all, in the large part, just a scientific game, a pleasant way to pass one's career while teaching students & spending tax pounds.

in thinking about this alog over the last month, i've had ringing in my ears Plank's principle, which is the quip that 'Science progresses one funeral at a time.'

which scientific ideas live forever, & which ones die an early death? which scientific ideas gain common acceptance, & which are quickly discarded? can we even tell which science is good & which bad? can scientists, as a whole, discriminate?

if you read the news media - as i've been doing with less & less pleasure each month over the last four years - you'd be forgiven for believing that major, important scientific advances occur very regularly; that each one provides a fundamental new truth about its object of study; & that science progresses, steadily, inevitably & in always the direction of greater enlightenment.

but if you spend any time on scientific social media, or if you read the best-available critical papers, reviews, or meta-analyses, or if you read books about the history & philosophy of science, then you'd - also - be forgiven for believing that most science makes very little progress at all.

before asking why this is & what we might do about it, i want to use an example from my own career.

my phd was on several topics related to the body & how the brain processes information from the body & the space around the body. one central theoretical concept was the 'body schema'. during my phd i read as much as i could about it, i tracked the references back to their earliest sources & i tried to pin down what - exactly - the body schema was.

i was unsuccessful.

the concept of the body schema can be traced back to about 1890, when Hermann Munk wrote a book - in German - about the brain. somewhere in that book he used the word 'Körperschema', which i translate as 'body schema'. fifteen years later, in 1905, a Frenchman called Pierre Bonnier used the word 'aschematie' to refer to how people can lose their sense of self after brain damage. then, in 1911, two Englishmen named Henry Head & Gordon Holmes - no relation - wrote about the 'schemata' in the brain that are disorganised following brain damage. the legend of the body schema was born.

the body schema, then, was supposed to be a structure, or set of structures in the brain, which records information about the body & allows you to perceive & use your body.

in my readings between about 2001 & 2012, most scientists i read referred to Head & Holmes as the originator of the body schema concept. it doesn't matter too much that they didn't actually originate it, & it's likely an English language bias effect, that the English scientists are referred to much more than the German & French scientists who came before them. more important here - & during my phd - was the question: well, what *is* the body schema & how do we measure it?

during my phd & beyond, after reading perhaps a thousand papers that used the term body schema, i remain unsure about what it is & how we measure it. typically, researchers would say that the body schema is a representation of the body in the brain that is used for movement, & particularly for the sensory control of movement. they might further say that it can be distinguished from the 'body image', which is a more cognitive, experiential, & visual representation of the body in the brain.

so far, so good. body schema is for sensory guidance of movement, & body image is for one's thoughts & experience of the body. so, what has neuroscience taught us, in one hundred and twenty four years, about the body schema? what specific functions does it perform? where is it in the brain? what happens if we damage it?

after perhaps a decade studying these topics, i can say that the answer to all three questions is: i don't know. from what i do know, my best guess is that movement is the specific function that the body schema performs, that the body schema is sort of distributed across virtually every part of the brain & the body as well, & that when it is damaged we lose some aspects of the sensory guidance of movement.

to summarise my view, then, the body schema is simply those parts of the brain & the body that subserve movement. & since this is a monumentally vague statement, likely the main function of the body schema concept is actually a narrative one: it helps scientists to frame their ideas, to generate hypotheses, & write papers - thousands & thousands of papers over a hundred & twenty four years. but there is nothing in the brain that corresponds directly with the body schema - it is just a narrative tool.

this conclusion might cause concern for a lot of body schema researchers - & i'm sorry about that - but science should be concerning, it should raise problems, & it should kill off ideas that serve little more than narrative functions. assuming that my view on the body schema is fair & the evidence does not indeed support this concept, then why hasn't the idea just died already?

& it's not like no-one has tried to kill it.

thirty years after Head & Holmes' paper, in 1942, Oldfield & Zangwill, two English neurologists, wrote four papers critically reviewing Head's concept of the schema, & placing it in historical context. they wrote that it is "legitimate to raise the question of how we are to try to envisage, in terms of actual cortical processes, the formation, continued existence, & functioning of the schema." (Part II, page 60). ouch.

thirty more years of body schema research passed by, until Poeck & Orgass in 1971 introduced their critical review with the phrase "the concept of the body schema is still basically undisputed, although this concept is ill defined and in its traditional form difficult to reconcile with modern theories of central nervous functions." (p254). double ouch.

thirty five more years of body schema research passed until i wrote my own critical review of the body schema, or rather, i avoided reviewing this "slippery issue" & instead just talked about experiments.

do we need to wait another thirty years for the body schema concept to die, or can we just kill it now?

to paraphrase Professor Dorothy Bishop on her blog last month: can we please just make the body schema stop now? more research is not required.

this neuroscientific idea has no known neuroscientific basis & serves only a narrative function in a small field of science. so perhaps we can let it live in that field, in a protected area, a once thought-to-be-extinct rare animal with no known surviving conspecifics, living on its own nature reserve, a bit like the tortoise, Lonesome George, in the Galapagos.

but the body schema is just one example of a scientific idea that won't die. there are many other scientific ideas that should be put out of their misery. ideas on which the evidence has long been damning, the vital signs have long been weak, & no further amount of research will resuscitate. another idea that i've tried to kill in my career is the 'principle of inverse effectiveness', which is mostly generated as a post-hoc statistical artefact, but which still appears on the Wikipedia page as one of the three principles of multisensory integration - how the brain combines information from different senses. it's probably still one of the three principles because two principles don't make a dogma.

another close related idea which just needs to die already is regression to the mean & other forms of 'double dipping' in data analysis. first bunked - & then de-bunked - by Sir Francis Galton in the 1870s & the 1880s, this is perhaps the most resilient of the zombie scientific ideas. statistical artefacts relating to regression to the mean & other forms of double dipping appear, in my estimates, in about five percent of all published scientific papers to this day. i will return to this in a later episode.

this month, i've been reading a collection of Thomas Kuhn's work. Kuhn's most famous book - & one of my favourites - is The Structure of Scientific Revolutions, in which Kuhn describes how scientists tend to work during normal periods & during revolutionary periods of science, & how scientific culture, language, & behaviour changes over time. i thoroughly recommend it: reading this book will likely be the best three dollars you've spent - i've put a link where you can buy the second handbook episode transcript on the error bar website.

in one of these essays, then, on the history of science published in 1971, Kuhn wrote that, "When a new concept or theory is successfully deployed in a science, some previously ignored precedent is usually discovered in the earlier literature of the field. It is natural to wonder whether atten[p120]tion to history might not have accelerated the innovation. Almost certainly, however, the answer is no. The quantity of material to be searched, the absence of appropriate indexing categories, and the subtle but usually vast differences between the anticipation and the effective innovation, all combine to suggest that reinvention rather than rediscovery will remain the most efficient source of scientific novelty." pp120-121.

Kuhn concluded that scientists are doomed to reinvent their ideas over & over & over again, perhaps until they finally gain acceptance under a new scientific paradigm. on this point alone, & for some scientific ideas, i hope that Kuhn was wrong.