Nowadays the BBC Science Focus site contains many misleading stories. This year a story on the BBC Science Focus site had this utterly bogus Fake News headline: "We finally know how life on Earth started, staggering new asteroid discovery suggests."
Fake News, BBC-Style
The study discussed provided not the slightest warrant for such a claim, and the article referred to a study written by authors who did not make such a claim. The study merely claimed to have found the tiniest trace amounts (roughly 1 part in a billion) of some chemicals used by living things, in a sample taken from an asteroid. As I discuss at length in my post here, the claims made in the study are not reliable, because the amounts reported are such negligible trace amounts that we can have no confidence that the reported chemicals came from the asteroid, rather than from earthly contamination. The BBC article on this study has quite a few statements as bogus and untrue as the article's headline.
On the same day as the bogus headline shown above appeared at the BBC Science Focus site, we had this equally bogus headline: "Alien life on Mars: ancient beach discovery may offer clearest proof yet." The article discussed nothing that was any evidence of life on Mars, where scientists have failed to even find any of the building components (amino acids) of the building components (proteins) of the simplest living things. This BBC Science Focus site these days is a shameless spreader of bogus science-related stories. This year it had a phony headline claiming "Our chances of finding alien just skyrocketed." It is referring to research that did nothing to make extraterrestrial life seem more probable, for reasons discussed here.
A recent article at the BBC Science Focus on so-called super-agers did not have a misleading headline. But it did contain very misleading statements by both the article writer and a scientist quoted. The article referred to "super-agers" defined as people who are very old but still have very good memory, and perform on memory tests as well as people decades younger. The article stated this:
"These super agers also had a youthful brain structure. Older brains generally have thinner cortexes – the outer layer of the brain – than younger brains, but these super-agers’ cortexes had not thinned."
The statement is not supported by any strong evidence. The 2012 scientific paper "Superior Memory and Higher Cortical Volumes in Unusually Successful Cognitive Aging" examined this question, looking at previously published papers. The paper stated this:
"Six articles [11, 14, 16–19] did not report statistical differences between superagers and age-matched controls concerning the whole-brain cortical thickness (mean wholebrain cortical thickness was 2.29–2.34 mm vs 2.27–2.32 mm, respectively). Only Harrison et al. [8] demonstrated a higher whole-brain cortical volume in superagers compared with age-matched controls and a similar volume to middle-aged control subjects (superagers vs. middle-aged controls vs. age-matched controls was 288.05 mm3 vs. 306.43 mm3 vs. 244.13 mm3 , respectively).....With the exception of the Lin et al. study [14] that has a high risk of bias, the included studies showed a weak correlation between levels of amyloid deposition and cognition among superagers and age-matched controls in a cross-section analysis [15, 17–19] following previous works in unimpaired adults [34–36]. In addition, histological studies demonstrated abundant deposition of neurofibrillary tangles in the hippocampalentorhinal complex [23] and in the posterior cingulate cortex of superagers [10, 21, 22]."
So according to the paper above, 6 out of 7 scientific papers on the topic failed to find any evidence that super-agers have more thick cortex regions of the brain than average people of their age.
The BBC Science Focus article quotes Sandra Weintraub, lead author of a year 2025 paper "The first 25 years of the Northwestern University SuperAging Program." That paper states this, using as a reference the outlier Harrison paper mentioned in the quote above:
"We found, as expected, that neurotypical seniors≥80 years of age have
significant and widespread cortical thinning compared to neurotypical 50- to 60-year-olds.19 Superagers, on the other hand, showed no
cortical thinning compared to the younger controls."
The reference (19) is to the Harrison paper mentioned in the quote above. But that paper is an outlier. As the quote above states, six other studies found the opposite, because they "did not report statistical differences between superagers and age-matched controls concerning the whole-brain cortical thickness."
The Weintraub paper states, "In
a preliminary study over 18 months, overall cortical thickness was
reduced by 1.06% in superagers compared to 2.24% in neurotypical
peers, a difference that was statistically significant." This is a low-quality study failing to make minimal standards for a correlation study. The study in question ("
Rates of Cortical Atrophy in Adults 80 Years and Older With Superior vs Average Episodic Memory") compared a super-ager group of 20 subjects to a normal control group of only 12 subjects. But such a control group of 12 subjects was too small for any reliable result to be claimed. A good rule of thumb regarding neuroscience research is that no study should be taken seriously unless it uses a minimum of 15 or 20 subjects per study group, and such a requirement includes the control group.
The paper "Prevalence of Mixed-methods Sampling Designs in Social Science Research" has a Table 2 giving recommendations for minimum study group sizes for different types of research. According to the paper, the minimum number of subjects for an experimental study are 21 subjects per study group. The same table lists 61 subjects per study group as a minimum for a "correlational" study. A study trying to establish a correlation between super-aging performance and cortical thickness is an example of a "correlational" study, so it requires more than 50 subjects per study group to be good evidence.
The Weintraub paper then claims this:
"The most surprising finding was the identification of an anterior
cingulate region in which superagers had greater cortical thickness
than even neurotypical participants 50 to 60 years of age (Figure 2A).
This finding subsequently has been confirmed in other studies."
We have a reference to four different papers, none of which provide robust evidence for such a claim. The first is the paper here. It had a study group size of only 17 super-agers, not large enough for a convincing result to be claimed. A study like this cannot be taken seriously unless it followed a blinding protocol, in which analysts analyzing brain scans were blind as to whether a brain scan being analyzed did or did not come from one of the super-agers. The paper makes no claim to have followed any blinding protocol. The reported greater thickness in the anterior cingulate region was so small (much less than 1 millimeter) than we can have no confidence in the claimed greater thickness, particularly given the lack of a blinding protocol. The second paper mentioned is the one here, which also has a too-small sample size of super-agers (only 19), and also fails to follow any blinding protocol. The reported greater thickness in the anterior cingulate region was so small (much less than 1 millimeter) that we can have no confidence in the claimed greater thickness, particularly given the lack of a blinding protocol. The third paper has a much better sample size of super-agers, but also failed to use any blinding protocol. So we should have no confidence in its claim of "greater thickness" in the anterior cingulate region of super-agers, which is not backed up by any specific numbers. The fourth paper uses a probably-not-big enough study group size of 26 super-agers, and also fails to follow any blinding protocol. So its vague unquantified assertion of greater thickness in the anterior cingulate region of super-agers does not qualify as convincing evidence.
The Weintraub paper then claims "the anterior cingulate gyrus of superagers also contains more von
Economo neurons compared not only to neurotypical peers but also to
much younger persons." We have a reference to three papers. The first paper used study group sizes (such as 21 and 18) too small for a reliable result to be claimed for a study of this type. We have only the scantiest claim about blinding, the claim that "the three cingulate subregions were traced...and
analyzed...by an individual blinded to group affiliation." Claims about blinding are never credible unless a paper discusses a detailed blinding protocol; and we get no such discussion in this paper. There are innumerable ways for a supposedly "blinded" analyst to actually know which study group some brain tissue belongs to, and we can only believe claims about blinding if the paper has at least a paragraph discussing how the supposedly blinded analyst really was blinded. The paper claims "there was a higher density of von Economo neurons," but fails to give us any numbers backing up this claim; and the reported statistical significance is an unimpressive "< .05." Also, the paper does not claim that this search for "von Economo neurons" was made by someone blinded as to group affiliation. So we end up with no compelling evidence that super-agers have more "von Economo neurons."
The second paper cited in support of the claim of more "von Economo neurons" is worse than the paper just discussed, as it used only way-too-small study group sizes of five, and has no mention of a detailed blinding protocol, other than the scantiest one-sentence mention, one failing to claim that the analyst looking for "von Economo neurons" was blinded as to group affiliation. The third paper mentioned is just as bad, involving way-too-small study group sizes of only five subjects per group. Contrary to one of the claims made above, the paper found "Preliminary stereological analysis found no significant differences in total neuronal counts (p>0.05) or neuronal size (p>0.05) among
the three groups across the cingulate." Those three groups included a super-ager group and normal-performing people of the same age.
I can hardly over-emphasize how important a detailed blinding protocol is for any analysis of this type to be credible. The type of questions addressed by studies such as this involve analysis that is extremely easy to do incorrectly whenever there is any motivation in an analyst to perform the analysis in some particular way. Doing something such as analyzing whether one group of subjects has an average cortical thickness half a millimeter greater (about the size of the period at the end of this sentence) is just the type of hard-to-get-right analysis task where bias (or belief in a desired result) might make analyst judgments unreliable. Trying to judge whether a subject has more neurons of some particular type is also the type of hard-to-get-right affair in which potential analyst bias is enough to invalidate any reported results. For there to be credibility of any claims made of an analysis of such marginal, hard-to-observe and hard-to-measure things, a paper must state a detailed blinding protocol explaining exactly why we should not believe that the analysts were just reporting a result that was longed for by their superiors. None of the discussed papers state such a protocol. Most of them fail to even use the word "blind" or "blinding." In the few cases when we find the word "blind" or "blinding" in such papers, it is the skimpiest mention, with a paper referring to blinding only in a single sentence, and mentioning only some limited type of blinding, not the following of a detailed blinding protocol. None of these papers did blinding right, and that is a major reason for dismissing all of them.
With a wink and a nod
The fact is that we have no solid evidence that the brains of super-agers with superior memories tend to be different from the brains of typical people of the same age. Analysis of super-ager brains does nothing to support claims that brains store memories.
No comments:
Post a Comment