The Somewhat Tarnished Gold Standard

This post is exclusively HM’s. It is being introduced here before the final two posts on “The Distracted Mind: Ancient Brains in a High Tech World” by Drs. Adam Gazzaley and Larry Rosen. The final two posts provide guidance on how to cope with the distracted mind. The authors do a well-intentioned but naive review of research and attempt to rank methods with respect to their level of confidence. This post provides some background for understanding research results and conclusions that should be generally valuable.

Randomized, placebo-controlled, double-blinded studies called randomized controlled trials or RCTs are the Gold Standard for research. This is a good standard to employ when it is feasible. But it is not always feasible, and attempts to apply it can lead to erroneous conclusions.

Here is an example where the Gold Standard was applied with no adverse consequences. This was the test of the Salk Vaccine for polio. HM was in the second grade at this time. Some test participants were given the Salk Vaccine and others were given a placebo. HM did not know whether he had been given the vaccine or a placebo, nor did the people administering the vaccine. We were assigned randomly, there were placebo controls, and the people administering did not know if we were being given the vaccine or the placebo (number were assigned to identify the conditions, but the administrators did not know what the numbers meant.

Now suppose participants assigned themselves. Here everything falls apart and valid interpretations are not possible.

Now suppose the Gold Standard was employed, but negative results, no evidence that the vaccine worked, were achieved. Does this allow the conclusion that the vaccine does not work? In statistics, you cannot prove a negative. The procedure is to decide to reject the null hypothesis with a certain degree of confidence. It is conceivable that the dose was too small. Another test might be warranted using larger doses.

Suppose the test involved a medication that was self-administered, and the Gold Standard was rigorously applied. What could possibly confound the results? Well the question is how well did the participants self-administer the medication? Differences in the results could be the result of an artifact caused by their being differences in adherence to self administration in the two groups.

The efficacy of meditation has been tested. HM has been pleasantly surprised by the positive results when the training was short and the training period fairly limited. In a study in which a group instructed to meditate is compared to a group instructed to do something else, there is the following possible problem: if participants have been randomly assigned to the groups, some who have been assigned to the meditation group might not believe in meditation and have a negative attitude to training and the entire project. This is different from RCTs in which the participants are passive and the treatment is administered to them.
When HM was a graduate student there was a hotly contested debate regarding whether humans could learn to control their autonomic nervous systems. HM thought this was ridiculous as there were practitioners of certain religions, Buddhism for example, who were able to control their heart rates and reduce them to frighteningly low levels. So HM thought the issue was resolved. But research was being done at colleges in which students were given biofeedback and examined as to whether they could learn to control their heart rates. Since this research failed, these researchers effectively accepted the null hypothesis, and ignored evidence from the millions of humans who were effective controlling their autonomic nervous systems.

HM is a strong advocate of mindfulness meditation. This increases the control of our attentional processes, which gives us increased control of our mind and emotions. The research question is not whether it works, but how much meditation of different types is useful. There is more than ample research indicating the benefits of the relaxation response discussed in healthy memory blogs.

So for RCTs to yield valid results, the experimental design and sample sizes should be adequate. Research participant compliance is another issue. Moreover, there is a much more important issue to which the research community at large has yet to consider. This issue comes from epigenetics: it is not just genes, but what is read out from the genes that is important. Nurture affects what is read out from the gene, so two individuals with identical genes can differ in how these genes are expressed. So identical twins can differ radically. One outstanding example involved two identical twin sisters. One was popular and a successful student. The other was socially withdrawn and a poor student. These twins were raised in the same family. Medications for people with identical genes could still have different effects. So under what conditions, are RCTs are still applicable? Herbert I. Weinberg has raised this issue in his book, “Willful Ignorance: The Mismeasure of Uncertainty.”

Yet another factor for consideration is the distinction between between near and far conclusions. Practically all results and conclusions should be regarded as near studies. Studies showing the cognitive benefits of games provide a useful example. If cognitive tests reveal a difference between people playing games and people who don’t, one can only make conclusions about these immediate benefits. Conclusions about the far effects of these games, say in the prevention of dementia, are questionable extrapolations. How long do these games have to be played? These conclusions await further research.

Now there are good data (see the healthymemory blog post “Cognitive Activity and the Risk of Alzheimer’s Disease”) indicating that cognitive activity helps build cognitive reserve which reduces the risk of dementia. Now the brain is always active, even when we sleep. So the question is what types of cognitive activity? HM has strongly argued that effortful processing, what Kahneman terms System 2 processing.

HM is mildly depressed when physical activity is emphasized, and cognitive activity relatively ignored. Sure physical activity is beneficial along with living a healthy lifestyle. But a main effect of physical activity is to increase oxygen flow to the brain. However effortful System 2 processing, activates many pathways in the brain and creates new links. Practically all learning initially involves System 2 processing, and as long as different and new ideas are being considered or new material is being learned, more pathways are activated and new links are made. HM argues it is this that enables the overcoming of the amyloid plaque and neurofibrillary tangles that are the defining characteristics of Alzheimer’s.

© Douglas Griffith and healthymemory.wordpress.com, 2017. Unauthorized use and/or duplication of this material without express and written permission from this blog’s author and/or owner is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to Douglas Griffith and healthymemory.wordpress.com with appropriate and specific direction to the original content.

Tags: Adam Gazzeley, Autonomic nervous system, Buddhism, Herbert I. Weinberg, Larry Rosen, meditation, null hypothesis, RCT, Salk Vaccine