Principles of a Post-Materialist Science

What we can call the Age of Materialist Science has given us a scientific academia landscape with the following dysfunctional features:

• People trained as scientists are coerced into accepting or paying lip-service to doubtful belief tenets such as the nonexistence of souls and spirits, the neural origin of all mental effects, and the accidental origin of all biological innovations, contrary to a vast number of facts and observations.

• Much of what is called science consists of the belief dogmas of a belief community, often ideas contradicted by much observational evidence, such as beliefs that human mental phenomena are purely brain effects.

• Scientists act like conformist members of a belief community, afraid to challenge belief dogmas that have become kind of sacred tenets in their community.

• Scientists are effectively encouraged to ignore a vast body of relevant observations of the paranormal, observations conflicting with the materialist belief tenets of the scientist's belief community.

• Much of the activity of scientists consists of doing poorly designed experiments trying to provide evidence for various beliefs prevailing in the scientist belief community.

• A system of peer review exists by which anomalous observations and contrarian analysis and heterodox viewpoints can be prevented from being published.

• Smug achievement legends are constantly repeated, even when they make no sense, such as the claim that a nineteenth century scientist did something to explain protein molecules and super-complex cells that he knew nothing about. 

• Overconfidence, hubris and knowledge overestimation is systematically encouraged, along with absurd claims that scientists “pretty much understand” things that are a thousand miles over their head.

The diagram below contrasts principles of good science with the current tendencies of scientists in universities:

The diagram below gives us a rough sketch of what we have in the Age of Materialist Science:

Someday this very dysfunctional system may be replaced by something better. Below are some very rough thoughts about some of the principles that a reformed, post-materialist science might follow.

• Do not put any previous scientist on any kind of pedestal, or reverently attach special value to his thoughts and theories.

• Recognize the strong possibility of an observer getting novel and currently inexplicable observational results, and instead of ignoring such results, direct funding and attention to follow up on them.

• Make federally funded scientific research freely available to all, rather than hiding scientific work behind paywalls or in expensive journals that make the taxpayer pay again for research he already funded.

• Value all observations by previous careful observers, not dismissing such observations with excuses such as saying people believed the wrong things when the observations were made.

• Discourage inflexible and one-sided printed science textbooks, replacing them with electronic works that allow readers to add links and comments that draw attention to errors and omissions in the works, and draw attention to conflicting evidence and contrary viewpoints.

• Recognize that humans do not understand the deep mysteries of the origin of life, the origin of large organisms and the origin of minds, and be less generous in funding scientists spending most of their careers trying to bolster previous boasts claiming explanations for  such things, while encouraging a critical analysis of their work.

• Do not ignore or dismiss repeatedly reported observational phenomena with a claim that the thing cannot be happening because it is impossible.

• Recognize that science is only one of quite a few important ways of reaching truth, acknowledging the equal importance of other paths such as logic, mathematics, history, scholarship and direct personal experience. 

• Revise psychology textbooks and biology textbooks so that they discuss at length hard-to-explain or inexplicable human observations and anomalous experiences, rather than censoring and suppressing such observational reports. 

• Follow the principle that when reliable observers frequently report specific kinds of observations of the  inexplicable and anomalous, such observations should be given more attention rather than less attention.

• Recognize the high tendency of social structures such as universities and colleges to create conformist belief communities that may have a negative impact on scientific progress, and give rise to speech customs and belief traditions that masquerade as well-established science.

• Create a credential system where anyone who passes a very hard 3-hour test on some scientific subject can be certified as an expert on that topic, even if he has not passed through the conformity-creating system of colleges and universities.

• Create and fund alternative structures and organizations for learning and research separate from universities and colleges that have been so infected by conformist belief traditions, not as a replacement but as a rival and an alternative.

• Respect the observations of people who are not professional scientists, rather than having some snobbish elitist attitude that observations count for little unless they are made by professors or near-professors.

• Revamp the current university and college science instruction system, which suffers from an authoritarian teacher/follower model not varying much from the teaching model of two thousand years ago, replacing it with some model encouraging more dissent and discussion of alternative viewpoints and evidence conflicting with predominant assumptions.

• Revamp or rethink the "one-to-many" model of collegiate classroom instruction which tends too strongly to produce a meek acceptance of ideology or dubious claims passed on by authorities.

• Replace in many university departments the current habit of giving someone with a master's degree a PhD based only on some very narrow research on one tiny topic, and make extremely broad multi-subject cross-discipline study the thing that gets you the PhD. 

• Federally fund independent scientists with worthy research proposals, even those working outside of the university system.

• Have a large fraction of all federally funded scientific research approved by persons outside of academia, to help prevent "echo chamber maintenance" effects whereby some scientist belief community with "purse string" control keeps funding research designed to support its cherished beliefs.

• Create a system in which public comments can appear at the end of every online science paper, allowing the public to dispute poorly designed papers, discuss conflicting evidence, and dispute papers making claims not justified by anything in the paper.

• To discourage studies in which researchers feel free to slice and dice data in innumerable ways until they find something "statistically significant," create a pre-registration system for experimental studies, in which papers must only report on whether the results supported a previously announced hypothesis and whether the data collection and data analysis procedures followed matched a detailed experimental plan published before any data was collected. Also have a red-flag system whereby the reader is warned of the failure of an experimental paper to follow such a standard.

• Create sample-size calculation conventions and study-group-size standards that limit the proliferation of misleading experimental studies in which false alarms (easily explainable by chance effects) are heralded as scientific discoveries.  

• Stop acting as if unintuitive principles not suggested by common experience (such as "everything must be explained by matter and energy" or "there cannot be mysterious unseen influences") should be preferred over intuitive principles suggested by common experience (such as "accidents don't produce complex inventions" and "where there's lot of smoke, there's probably fire," which is a good rule-of-thumb in dealing with whether  observational reports suggest some underlying reality).  

• Stop promoting scientists based on the number of papers they have published, and encourage alternate promotion criteria such as the number of times a scientist has published a paper judged to follow a "Best Practices" standard. 

• End the current secretive peer-review system that acts as an ideological filter preventing dissenting viewpoints, reports of conflicting evidence, and novel observational reports, replacing it with a “let scientists stumble but flag their stumbles” system that will encourage public comments about any mistakes in a scientific paper, along with also a quality grading system whereby inferior papers can be low-graded.

• Stop making dubious claims of a scientific consensus that are not established by secret ballots of scientists, and create some system for secret balloting of scientists that will clarify how much they agree on opinions, a system that always offers a variety of belief answers including “I don't know.”

• Make a large fraction of all scientific funding go to studies that will be guaranteed publication studies in which publication is assured even if a null result is found.

• Make a small fraction of all scientific funding go to groups trying to disprove or falsify prevailing ideas and assumptions.

• Reform the speech and writing habits of scientists, to discourage the continuation of misleading speech practices and misleading visuals that are shockingly common in scientific literature.

• Stop referring to speculative unobserved things such as dark matter, dark energy and accidental macroevolution as "science," and accurately refer to them as "scientist speculations." 

• Reform current profit structures that reward bad science and bad scholarship that ignores important relevant evidence, and create novel profit structures that reward best-practices science and scholarship that takes into account all relevant evidence.

The current profit structure is extremely dysfunctional

Why the "A Memory Is Stored Throughout the Brain" Idea Makes Things Much Worse

No matter what form such an idea takes, the idea that the human brain stores memories creates the most gigantic difficulties, difficulties so bad that we should reject all claims that memories are stored in brains.  Let us look at two different forms of such an idea, and look at some of the difficulties that each form creates. 

The most common form of the idea that brains store memories is the idea that a memory is stored in one particular little spot of the brain, with each memory being stored in a different tiny spot.  Below are some (but not all) of the huge problems that such an idea creates:

(1) The spot selection problem. While computers have operating system algorithms for choosing a random storage spot,  a brain would seem to have no method or capability of choosing one small little storage spot for a memory to be placed. So if, for example,  we imagine that a brain placed a memory in storage spot number 263,432 out of 250,000 storage spots, we have the problem: why would that particular spot have been used to store the memory, and not some other spot? 

(2) The writing and encoding problem.  Once some spot had been selected for a memory to be written, something learned would have to be translated into neural states or synapse states and then written. There is no credible theory of how learned information or episodic memories could be translated into neural states or synapse states. There is no known mechanism in the brain for writing information. A computer has an operating system with formally designed encoding protocols such as the ASCII protocol and a protocol for converting decimal numbers into binary numbers. A brain has no such thing. A computer has a read-write head for writing information. The brain has no such thing. 

(3) The navigation problem.  Humans routinely display the ability to instantly recall learned information, given a name, date or image. So, for example, if you say "death of Lincoln," I will instantly be able to recite various facts about the death of Abraham Lincoln, such as that it occurred because John Wilkes Booth shot Lincoln through the back of his head at Ford's Theater in April, 1865.  If we believe that a memory is stored in some tiny little spot in the brain, such as storage spot 186,395 out of 250,000, then we have the problem: how was the brain able to instantly find that exact tiny spot where the memory was formed? This difficulty is a "show stopper" for all claims that a memory is stored in one exact spot of a brain, an insuperable difficulty.  We cannot get around such a difficulty by imagining that a brain uses the type of things that a book or a computer use to allow instant retrieval.  Books and computers use information addressing and indexes to allow instant access of a particular data item.  The brain has neither addressing nor indexes.  Unlike houses that have street addresses, neurons don't have neuron numbers or any other addressing system. Storing a memory in a brain would be like throwing a little 3" by 5" card into a giant swimming pool filled to the top with a million little 3" by 5" cards.  Just as it should take you ages to find a specific piece of information stored in such a swimming pool, it would take you ages to find in the brain some particular piece of learned information, if it was stored in one tiny spot, like a book stored in one spot on the shelves of a huge library.  

 

(4) The reading and decoding problem.  If a memory was stored in one particular spot, there would be the problem of how a memory could be read from that exact tiny spot. The brain seems to have nothing like a read mechanism.  Nor is there any known mechanism by which information that had been stored as neural states or synapse states could be translated into a thought that would appear in your mind. 

But there is another form of the idea that brains store memories.  There is the idea that the brain stores a memory throughout the brain, rather than writing the memory only in one little spot.  However, the difficulties in this idea are even worse than in the idea that the brain stores each memory in one specific spot. The idea that a memory is stored throughout the brain has the following difficulties:

(1) A greatly worsened writing and encoding problem. The idea that a memory is stored throughout the brain has the same writing and encoding problems mentioned above, except that now the problem is much worse. This is because now rather than just imagining that a memory is written in one tiny spot by a brain without any known writing mechanism, we must now imagine that such a brain manages to write all over itself each time that a memory is stored.  This would take much more time than writing to a single spot in the brain. Humans routinely show the ability to instantly form new memories,  an ability that neuroscientists cannot credibly explain. You only make that problem worse if you imagine that each time a memory is formed, the brain is writing to many different places rather than one. 

(2) A memory disassembly problem. The idea that a memory is stored throughout the brain creates a gigantic new problem that did not exist if you assume a memory is written to only one tiny spot: a memory disassembly problem. If you imagine that a memory is broken up into tiny pieces and stored throughout the brain,  then you have the problem that such a disassembly process would require additional time, making it all the more impossible to explain the wonder of instant memory formation. Similarly, it only takes a second for me to store a piece of paper by opening a book in my library and sticking the page inside the book; but if I have to cut up the page into twenty pieces and store the pieces in twenty books, that takes much longer. 

(3)  A memory reassembly problem. The idea that a memory is stored throughout the brain creates a gigantic new problem related to memory recall: a problem of reassembling the memory that had been stored in scattered pieces throughout the brain. If we imagine a brain with only one memory, such a thing does not seem so hard (the brain could just read throughout itself looking for memory pieces, and read them all up). But if we imagine many, many thousands of memories that had each been stored by storing pieces of individual memories throughout the brain, then such an assembly seems impossible to occur, no matter how long it would take.  

I can give an analogy. Suppose I am storing 1000 family photos through a scattered storage method. I take each of the thousand photos, cut them up into little pieces, and store each by putting them in different pages in the books that make up my large library. Now, suppose my wife comes and asks, "Please get me a picture of our trip to Los Angeles."  Retrieving that photo would be a nightmare.  I couldn't just get all the photo pieces by shaking each book in my library.  That's because the pieces of each photo would be mixed up with all the pieces of 1000 other photos.  Similarly, we can imagine no way in which a brain that has scattered pieces of each memory throughout itself could ever reassemble such pieces to produce a good recall of a particular memory.  And if it ever could do such a thing, such a recall would take very long lengths of time, and a recollection could never occur instantly.   

(4) A greatly worsened reading and decoding problem. The idea that a memory is stored throughout the brain has the same reading and decoding problems mentioned above, except that now the problem is much worse. This is because now rather than just imagining that a memory is read from one tiny spot by a brain without any known writing mechanism, we must now imagine that such a brain manages to read from all over itself each time that a memory is retrieved.  This would take much more time that reading from a single spot in the brain. Humans routinely show the ability to instantly retrieve new memories,  an ability that neuroscientists cannot credibly explain. You only make that problem worse if you imagine that each time a memory is recalled, the brain is reading from many different places rather than one.  

There was recently in the news an MIT press release story making the utterly unfounded claim that some research had shown that "a single memory is stored across many connected brain regions." What we have is another misleading claim about engrams from MIT, which for many years has been a notorious source of unfounded claims about neural memory storage.  In the 2018 post here I took a long look at how MIT memory researchers had repeatedly made grandiose but unfounded claims about memory research.  I showed that MIT researchers had again and again made grandiose claims based on shoddy poorly-designed rodent studies guilty of using way too small sample sizes.  The results proclaimed by such researchers are mainly false alarms, the type of false alarms that are very easy for a researcher to get when he uses fewer than 20 subjects per study group. 

The latest memory research announcement by MIT discusses research guilty of the same old shoddy research practices that MIT memory researchers have been guilty of for so many years.  Once again, when we read the scientific paper (which can be read here) we find that the researchers used way-too-small study group sizes, such as one group of only 7 mice, another group of another 9 mice, and another group of only 10 mice. If the scientists had acted like good experimental scientists and had done what is called a sample size calculation, they would have found out that such tiny study group sizes are utterly inadequate to produce a reliable result. But they did no such calculation. They confess in their paper, "No statistical methods were used to predetermine sample sizes."

The scientists fear-conditioned mice by electrically shocking them (this typically involves getting mice to learn there is one little area of a cage where the shocking will occur). The scientists then measured something in lots of different regions in the brain of a very small number of mice, and the scientists have somehow got the idea that some regions were involved in memory storage.  To test such suspicions they "optogenetically stimulated" mice to try to artificially create fear in the mice, zapping the little regions they thought were involved in storing a memory.  This "optogenetic stimulation" is a method of using light to zap the brain of a mouse. 

The thinking behind such strange zappings of mouse brains is that by zapping some little part of a mouse's brain, you can get a mouse to remember some fear memory formed when a mouse was zapped by stepping on an electrical plate.  The underlying theoretical assumption was wildly implausible. It was the idea that if a mouse has a particular memory stored in many brain regions, then you can get the mouse to re-experience that memory by stimulating only one of those regions.  Such an idea makes no sense. It's kind of like thinking that I would get Tom Brady to throw a pass by sticking a sewing needle in his arm, stimulating one of the many muscles he uses in throwing a pass.  

Conclusions about whether the fear memory was recalled were based on a poor low-reliability technique that neuroscientists have long used: a judgment about whether so-called "freezing behavior" occurred (such behavior being defined as mere inactivity). The underlying assumption is that mice freeze when afraid, and that you can judge if a mouse is recalling a fear memory by looking for an instant of non-movement in which a mouse may be "freezing in fear." Given the start-stop, helter-skelter way in which mice move, any judgment about whether a mouse froze is going to be a subjective, unreliable judgment. So there is too much of a possibility of observational bias here, one in which an observer subjectively reports the effect he is hoping to find. Similarly, you might subjectively report that your goldfish in a goldfish bowl tends to move towards you when you are looking into the bowl, but that would probably tell us more about your desire to see something than about the goldfish. The idea that mice freeze when terrified isn't even a very sound one.  I have seen  dozens of mice flee when scared by a human, but I never once seen a mouse freeze when suddenly scared by the presence of a human. 

There is a very reliable way to measure fear in mice: you measure the mouse's heart rate, which undergoes a very sharp spike in mice when they are afraid. Our neuoroscientists senselessly continue to use unreliable subjective judgments about "freezing behavior" to try to measure fear in rodents, rather than sensibly using reliable measurements of heart rate spikes in rodents.  Being guilty of this flaw, the new MIT study has provided no reliable evidence about whether or not the mice remembered fear when parts of their brains were zapped. 

Moreover, when "freezing" (simple non-movement) occurred in the mice, the "freezing effect" could have been produced not by a recall of fearful memories, but by the very fact the energy was being transmitted into the brain of the mice. Imagine you are running along, and suddenly a scientist switches on some weird thing that causes some energy to pour into your brain. This all by itself might cause you to stop, even if it didn't cause you to recall some memory that caused you to stop. What could have been going on in the mice was just a kind of pausing effect caused by a novel stimulus rather than a recalled fear effect. A science paper says that it is possible to induce freezing in rodents by stimulating a wide variety of regions. It says, "It is possible to induce freezing by activating a variety of brain areas and projections, including the hippocampus (Liu et al., 2012), lateral, basal and central amygdala (Ciocchi et al., 2010); Johansen et al., 2010;  Gore et al., 2015a), periaqueductal gray (Tovote et al., 2016), motor and primary sensory cortices (Kass et al., 2013), prefrontal projections (Rajasethupathy et al., 2015) and retrosplenial cortex (Cowansage et al., 2014).”

Neither the paper nor its supplementary information contains any  mention of a blinding protocol, other than the bare statement that "all behavior experiments were collected and analyzed blind to experimental group."  Unless a paper has a detailed discussion of  how an effective blinding protocol was implemented (one that really achieves a blinding effect to prevent observer bias), we should assume that no effective blinding protocol was implemented.  For example, if you had one group of 7 mice with optogenetic wires attached to their brains, and another group of control mice with no such wires, anyone would be able to tell which group was the group that was hoped would show more "freezing" behavior (even if those judging how much the mice froze were not specifically told which group was which).  So some method that can technically be referred to as "blind" may not be blind at all because of a lack of an effective protocol. Whenever any paper claims a blinding protocol but fails to specify how an effective protocol was achieved,  we should assume that no effective methods of blinding were used (a severe defect in an experiment). 

Being guilty of quite a few serious methodological flaws (primarily the use of way-too-small study group sizes), the new MIT study has produced no robust evidence that memories are stored in the brains of mice, and no robust evidence that a memory is stored in many different regions of the brains rather than in some particular spot. According to the paper here, "Quality research practice requires both testing statistical significance and reporting effect size." But the new MIT paper reports no effect size. That is what goes on when shoddy experimental research practices have been followed, such as using way-too-small study group sizes. 

In this paper here we have a discussion of the absurd technique most commonly used to measure fear in rodents:

"In mice, freezing is a common and easily measured response used as an index of fear conditioning (Blanchard and Blanchard 1988; Graef 1994). Fanselow (1990) and Paylor et al. (1994) define freezing as the absence of any movement except for respiratory-related movements. Freezing behavior is measured by direct observation, scoring an animal as either freezing or active per interval of time, usually every 5–10 sec (Fanselow 1990; Paylor et al. 1994) or measuring freezing duration with a stopwatch (Phillips and Le Doux 1992)."

The technique discussed above measures only mouse inactivity, which will vary randomly. There is no sound basis for calling such a measurement a measurement of "freezing behavior." If I take 10 snapshots of a mouse per minute, that show the mouse not moving in three of those snapshots, that is no reason for thinking that the mouse was afraid when three of those ten snapshots were taken.  What is occurring these days among cognitive neuroscientists is deceptive labeling of mouse inactivity measurement. Graphs that should be labeled "mouse inactivity (%") are being misleadingly labeled "mouse freezing (%)."  The term "freezing" should never be used unless a sudden stopping of traversal was observed. 

Big Study Finds Brain Gray Matter and Cortical Thickness Peak at Age 6 or Earlier, Contradicting Brain Dogmas

A new study published in Nature (with very many listed authors)  has produced a result very relevant to claims that the human mind is produced by the brain.  Entitled "Brain Charts for the Human Lifespan," the paper says, "We aggregated 123,984 MRI scans, across more than 100 primary studies, from 101,457 human participants between 115 days post-conception to 100 years of age."  MRI scans are a type of scan that allow you to see the physical structure of the brain. 

Human experience is that intelligence roughly peaks around age 20, with no major decline before age 40.  The lack of intellectual decline before age 40 is partially why nations generally elect leaders that are 40 years old or older, and it is partially why major corporations generally have as their Chief Executive Officer someone who is age 40 or older.    The claim has often been made that gray matter in the brain is some type of neural matter particularly associated with intelligence. It has often been claimed that you think with the gray matter of your brain.  Given human intelligence peaking around age 20, and not declining much before age 40, such claims predict that gray matter in the brain should peak at around age 20, without much decline before age 40. 

But this is not at all what the "Brain Charts for the Human Lifespan" study found. It found that gray matter in the brain peaks at around age 6, with about a 12% decline by age 20, and about a 20% decline by age 40.  This is shown in a chart from the paper:

From the "Brain Charts for the Human Lifespan" paper

According to this chart:

• Gray matter volume peaks around age 6;
• gray matter volume declines by about 12% by age 20;
• gray matter volume declines by about 20% by age 40;
• cortical thickness peaks by about age 2 or 3;
• cortical thickness declines by about 10% by age 20;
• cortical thickness declines by about 15% by age 40;
• white matter volume peaks at about age 30.

The chart above is a bit hard to read, but at a web site set up by the paper authors, the gray matter volume trend by age is graphed in the easy-to-read graph below:

 Source: https://brainchart.shinyapps.io/brainchart/

The data presented in the new study simply does not match human experience regarding intelligence and age.  I cannot cite any numbers showing a growth of IQ by age, because IQ tests are designed to test differences in either children of one age or adults.  But the following are simple facts of human experience

• Children age 6 have an intelligence that only seems to be 50% or smaller than the intelligence of adults (despite the gray matter volume peaking around age 6);
• adults of age 40 do not have an intelligence noticeably less than those of adults at age 20, and have an intelligence much greater than those age 6 or younger (despite such 40-year-olds having 20% less gray matter volume than those age 6, about 15% less cortical thickness than those age 2, and roughly 10%  less gray matter volume than those age  20);
• adults of age 20 have an intelligence much higher than children of age 6 (despite such 20-year-olds having about 12% less gray matter volume than those age 6, and about 10% less cortical thickness than those age 2 or 3);
• adults of age 30 do not seem any smarter than adults age 20 (despite such 30-year-olds having white matter volume peaking at their age);
• children with an age of about 2 or 3 have an intelligence that only seems to be a small fraction of the intelligence of adults (despite their cortical thickness peaking around this age). 

Once again, the "brains make minds" dogma gigantically flunks an empirical test. But you won't hear about this failure in the mainstream media, which tends to keep scientists and the public in a "filter bubble" that allows them to keep thinking that their cherished dogmas are holding up well, no matter how miserably such dogmas are failing empirical tests. So, for example, a Nature article on the "Brain Charts for the Human Lifespan" study completely fails to mention how dramatically the study's data conflicts with human experience about how intelligence changes with age. 

"Brains Make Minds" Idea Flunks an Audit of a Large Brain Scan Database

For many years neuroscientists have been claiming important results about brains and minds, after doing brain imaging experiments using small sample sizes.  Typically such claims will be based on way-too-small sample sizes smaller than 15.  A new press release from the University of Minnesota Twin Cities announces results which indicate that such small-sample correlation-seeking imaging experiments are utterly unreliable.  The headline of the press release is "Brain studies show thousands of participants are needed for accurate results."

There is a technique to measure the reliability of brain scans when used to make claims about supposed neural signs of cognitive activity.  The technique involves measuring what is called the test-retest reliability of brain scans.  The technique involves trying to determine to what extent some claimed neural sign of cognitive activity shows up both times when two different brain scans are taken of the same person. 

So, to imagine a hypothetical example, suppose some claim is made that the hippocampus of some subject activated more strongly when the subject recalled something. A check can be made as to whether the same thing was seen when the same subject had his brain scanned a second time, doing the same recall task.  If no such increased activation is seen on the second brain scan, we have a good reason for thinking that the claim about the first scan is unwarranted, and that the first scan has simply given a false alarm, a result of random brain fluctuations.  

Conveniently "covering their tracks," the vast majority of neuroscientists fail to do a retest of subjects when doing brain scanning experiments. However, there are some large databases of brain scans that include scanning retests of many subjects. It is therefore possible to judge how well claimed neural correlations of cognitive activity tend to replicate when a second test is done of the same subject. 

One such brain imaging database is the Adolescent Brain Cognitive Development Database. The database includes scans of thousands of subjects doing particular tasks such as a Monetary Incentive Delay task. a Stop Signal task and an n-back or nBack task (as described here). The database includes brain scans of more than 10,000 adolescents, and for more than 7000 of these adolescents a second set of scans were taken two years later, with the subjects performing the same tasks as in the first scan.  Such a database provides an excellent platform to test whether correlations between brain states and mental activity tend to repeat when the same subjects were scanned two years later.  

Such an examination is reported in the scientific paper entitled "Reliability and stability challenges in ABCD task fMRI data" by James T. Kennedy and others, which you can read here or here.  The study used a measure of retest reliability called the intraclass correlation. An intraclass correlation of less than .4 is generally regarded as "poor." In the wikipedia.org article on the intraclass correlation we read the following:

"Cicchetti (1994) gives the following often quoted guidelines for interpretation for kappa or ICC inter-rater agreement measures:

• Less than 0.40—poor.
• Between 0.40 and 0.59—fair.
• Between 0.60 and 0.74—good.
• Between 0.75 and 1.00—excellent.

A different guideline is given by Koo and Li (2016):

• below 0.50: poor
• between 0.50 and 0.75: moderate
• between 0.75 and 0.90: good
• above 0.90: excellent"

The results reported in the scientific paper entitled "Reliability and stability challenges in ABCD task fMRI data" by James T. Kennedy and others were devastatingly negative.  In the paper's abstract we read this: 

"Reliability and stability [quantified via an intraclass correlation (ICC) that focuses on rank consistency] was poor in virtually all brain regions, with an average ICC of .078 and .054 for short (within-session) and long-term (between-session) ICCs, respectively, in regions of interest (ROIs) historically-recruited by the tasks. ICC values in ROIs did not exceed the ‘poor’ cut-off of .4, and in fact rarely exceeded .2 (only 5.9%).... Poor reliability and stability of task-fMRI, particularly in children, diminishes potential utility of fMRI data due to a drastic reduction of effect sizes and, consequently, statistical power for the detection of brain-behavior associations."

What this means is that there was extremely low level of repetition of effects between one scan on a subject and a later scan on the same subject. As mentioned above, an intraclass correlation of less than .4 or .5 is commonly described as "poor." The very low intraclass correlations reported (only .078 and .054) can be described as extremely poor or appallingly poor.  In the quote below, the authors of the study describe their results as a "particularly disappointing outcome," and wonder what factors contributed to so poor an outcome. We read the following: 

"Our main finding was that within-session reliability and longitudinal stability of individual differences in task-related brain activation was consistently poor for all three ABCD tasks. Data cleaning approaches like outlier removal, movement regression, and rank normalization significantly increased reliability and stability, but by a small, seemingly inconsequential amount (average change of less than .025). While the finding of poor within-session reliability and longitudinal stability in the ABCD task fMRI data did not come as a surprise, given the mounting evidence for generally lackluster reliability of task-fMRI in mostly adult samples (Elliott et al., 2020; Herting et al., 2018; Noble et al, 2021), the present estimates are far below the .397 average reliability of task-fMRI activation estimated in the meta-analysis by Elliott et al. (2020). The question then arises, what factors could contribute to this particularly disappointing outcome? "

These results are what we would expect under the idea that the brain is not the source or cause of human mental activity, and not the storage place of memories.  In such a case we would expect that when scientists claimed some correlation between brain activity and mental activity after brain scanning some subjects, they would almost always be finding mere false alarms that would strongly tend to disappear when a second brain scan was made of the same subjects.