For decades neuroscientists have taught or attempted to suggest the unwarranted claim that brain scans suggest the human brain is the source of human cognition. Given the great numbers of such claims that have been made over the past several decades, a reader may be surprised to read that a new scientific paper states, "We report the first systematic review that assesses how information from structural and functional neuroimaging methods can be integrated to investigate the brain substrates of cognition." This should raise our suspicions that brain imaging can help establish that cognition is produced by brains. If such a claim is true, why is it only now (in the year 2022) that we would be seeing "the first systematic review that assesses how information from structural and functional neuroimaging methods can be integrated to investigate the brain substrates of cognition"? Such reviews should have been done before any claims were made that brain imaging helps show the brain is the source of human cognition, not after such claims had been made for decades.
We seem to have here in such "jumping the gun" behavior another example of what biologists have very long been guilty of: claiming some triumph has been achieved before they have achieved necessary prerequisites of such a claim, like some mother claiming that her small baby can run before the baby has even learned how to crawl or walk. Something similar went on in discussing human origins, with biologists boasting for many decades that they had successfully explained the origin of the human race, when they had not achieved some of the most important prerequisites that should have been achieved before making such a claim (such as understanding the molecular nature and organization of protein molecules, the structural organization of cells, and the still not-understood riddles of the origin of language and the morphogenesis origin of an immensely organized adult human body from the million-times simpler simplicity of a single fertilized ovum).
After its abstract the new paper ("Relating cognition to both brain structure and function: A systematic review of methods") starts out by giving us a classic example of what occurs so very often in neuroscience papers: an invalid citation in which some dubious claim is followed by a reference to a paper that did not establish such a claim. We read, "Cognitive function and adaptive behaviour rely on structure and dynamics of largescale neural networks (Friston, 2002)." The citation is to a paper that merely states, "We try to show that learning can be implemented in a biologically plausible way."
After getting 1923 papers from database searches, and removing 251 duplicates, the new study had 1673 papers. An assessment "for eligibility" reduced this total to only 159 papers, and a further application of quality criteria reduced this number to only 102. Such extremely high rates of exclusion should raise our suspicions. Why did only 102 out of 1923 papers meet the study's eligibility and quality criteria standards? This is what we might expect if the vast majority of experimental neuroscience papers are using faulty methods or making invalid claims.
It seems that the number of papers that should have been excluded would have much higher. The new paper lists four reasons why it excluded papers. None of the reasons listed are one of the top problems with experimental neuroscience papers these days. Such reasons include the following:
(1) A failure to do a sample size calculation needed to determine the number of subjects needed for a robust result.
(2) The use of too-small study group sizes, in which the number of subjects is less than the minimum (about 15) needed for a moderately persuasive result.
(3) A failure to declare and implement a thorough blinding protocol to prevent experimental bias in which the experimenters tend to find whatever result they were hoping to find.
(4) A failure to pre-register a detailed plan for gathering and analyzing data, leaving researchers with freedom to run a "fishing expedition" kind of study in which they can "slice and dice" data in countless different ways until they find a result they were hoping to get.
How many of the 102 papers would have survived a quality check excluding papers with such methodology flaws? Probably very few, because the prevalence of poor methodology is epidemic in neuroscience these days, with most experimental studies being guility of two or more of these failures. Then there is the fact that according to a pie chart in the new paper, only about one quarter of the 102 papers used "direct inference," with about half using some "indirect inference" method that is less reliable than direct inference.
In the new paper's summary we have this confession, which should raise further doubt in anyone thinking brain scans have supported claims that brains make minds. We read this:
"First, it became apparent that fMRI protocols have taken clear dominance over other functional imaging techniques in this research field. As mentioned in the introduction of this review, fMRI method suffers from low temporal resolution and is not a direct measure of neural activity."
You should never expect to get in a neuroscience paper a really candid discussion of how the boasts of neuroscience do not match experimental results. Neuroscientists are members of a conformist belief community, and within such a community there are taboos that cannot be violated and speech customs which scientists are pressured into following. But in the last paragraph of the new paper we do get a kind of watered-down confession about the shortfall of neuroscientists in proving their belief dogmas. We read this: "This review demonstrated that the relationship between brain structure and function and cognitive function is still largely underexplored." A more candid statement would have stated, "The relationship between brain structure and function and cognitive function has not yet been established."
The study here gives a rather interesting poll of neuroscientists. There are some surprising answers. Based on their standard claims about brains and minds, we would expect close to 100% of neuroscientists would agree with this statement: "If it were possible to transplant our brain to another body we would still be ourselves, albeit in a new body." The actual percentage of polled neuroscientists who agreed with this statement was only 51%. Could this be because deep down inside a large fraction of our neuroscientists don't really believe some of the things they teach? I don't know.
Also interesting was the fact that only 6% of the polled neuroscientists agreed with the statement that "memory is stored in the brain much like in a computer, that is, each remembrance goes in a tiny piece of the brain." 82% of them disagreed, with 12% saying they did not know. I can understand why so few neuroscientists would have agreed with such a statement. If a neuroscientist claims that each memory is stored in one tiny spot of the brain, this raises the problem of how a brain could be able to instantly read from just the right tiny spot when an instantaneous recollection occurs. For example, if I suddenly hear the phrase "death of Abraham Lincoln" and instantly remember "assassination by John Wilkes Booth at Ford's Theater, April 14, 1965," how was I able to find the exact tiny spot where that information was stored, in a brain without any coordinate system or indexing system or position notation system?
Neuroscientists try to get around this problem by imagining that a memory is stored in not just one tiny spot in the brain but multiple spots in the brain. Unfortunately, this does not make things better for the theory that we retrieve memories from brains; it makes things worse. If, for example, my memory of how Abraham Lincoln died was scattered among several different tiny spots in the brain, then my instant retrieval of that memory from a brain without any coordinate system or addressing system or indexing system would be even more inexplicable than if I were to get the information all from a single spot. I would now have the additional difficulty of explaining how this spatially scattered information was all instantly retrieved from just the right places and also pieced together to make a single seamless integrated memory. No neuroscientist has even given a credible explanation of instant memory recall, and the only credible explanation of such a thing would one abandoning the notion of a neural storage of memories. We do not recall at the speed of brains; we recall at the speed of souls.
No comments:
Post a Comment