By Siegfried Othmer, PhD
Science Magazine has just weighed in on the recent sanctioning of Lumosity by the Federal Trade Commission for ostensible over-claiming.(Reference) From our perspective we regard cognitive skills training as a kindred technology to neurofeedback. It’s yet another computer-based method of enhancing brain function. So where do we come down on the issue of the FTC sanction? We come down firmly on both sides of the issue! On the one hand, over-claiming by any of the major players ends up casting a pall on the entire field. And on the other, the FTC may well have exhibited entirely too much regulatory zeal. The review in Science sheds light on the controversy.
First of all the FTC ruling unquestionably found resonance within the neuroscience community. There many are still convinced that the improvement in brain function by such means is not possible. Most have not looked into the matter. This is just what they learned in graduate school, and they don’t take well to being blind-sided and upstaged by some commercial outfit.
Said Michelle Rusk, spokesman at the FTC, about the Lumosity claims: “the most that they have shown is that with enough practice you get better at these games… there’s no evidence that training transfers to any other real-world setting.” Whoa! So the issue is not whether learning actually takes place with the training—-which is acknowledged by the FTC—-but whether that learning generalizes. That is of course a lot harder to demonstrate.
If we are honest, then we recognize that there are other accepted methods in use that also offer only limited generalization. Take ABA, for example, the Applied Behavior Analysis routinely foisted upon unappreciative autistic children. Often the outcome is little more than rote behavior that at best is “emitted” in socially appropriate situations. Even African Grey parrots are capable of such learning. And they are even able to match their verbal repertoire to appropriate social situations. (See Note 1 below.) Often that is the most that can be hoped for in the autistic child. There is no generalization here, and yet people pay a lot of money for even minor gains. No one insisted on a demonstration of generalization before that technique became generally accepted.
In yet other cases, generalization obviously does take place. If one trains for larger digit span, then that capability surely generalizes to other situations that challenge working memory capacity. At a minimum, the burden of proof shifts to those who assert that generalization does not occur, because that would be a bizarre outcome.
And then we come to the final resort of the skeptics, which is to insist that claims must be “supported with statistically significant results from randomized, blinded, placebo-controlled human clinical trials.” As cognitive neuroscientist Daphne Bavelier points out in a response to the FTC, the insistence on blinded designs is “untenable in the case of behavioral interventions—-patients cannot be blind to the game they are asked to play or the exercises they are asked to do.” It doesn’t require a cognitive neuroscientist to call attention to what should be obvious, but it helps.
So much for the blinding. What about the placebo control in general? That’s not possible either. You cannot do placebo cognitive skills training. Actually, the case against a placebo design is even stronger than that. If the baseline position is that cognitive skills training is impossible, then one doesn’t actually have to have a control group at all. One just has to show that good results can be obtained with the method, and that has apparently been done.
Placebo-controlled designs serve the single purpose of proving that the ostensible active ingredient in a drug is in fact an active ingredient. None of this transfers to the situation of cognitive skills training. There is no active ingredient here. We are in the realm of a learning paradigm, and the question to be asked and answered was framed famously by George W. Bush: “Is the children learning?” So it appears that not much critical thinking has gone into this issue at the FTC. (See Note 2)
We have been interested in complementing neurofeedback with cognitive skills training for many years—-going back all the way to the days of Thinkfast and Cognometer. The Thinkfast program was delivered on a floppy disc—-when those were still floppy! There is just no question that these training programs are helpful. Our problem was getting people interested in doing them. That’s because the programs constantly confront you with your own shortcomings. The games furnish very direct evidence on whether performance is improving. This places the user himself in the best position to decide whether to continue with the project. The user does not need the Federal Trade Commission as his shield and protector in this instance.
It’s on the question of generalization into life that formal ‘evidence’ becomes hard to get. It arrives one client at a time. So one is compelled to rely on anecdote. And as we know, the medical field relies on anecdote all the time. The doc asks the patient how he is doing, and when the patient says that he is doing a lot better with his chronic pain, thank you, then that report never fails to be accepted.
What we have, then, is a situation where the FTC is applying an impossible standard—-proving that generalization into life at large occurs without reliance on anecdote. Nothing could have passed their screen. One is reminded of the witches trials in the Middle Ages. Witches were subjected to trial by drowning. If the accused drowned, then God had rendered his judgment and the issue was settled. If the poor woman (and 80% were women) did not drown, then the only way she could possibly have survived the ordeal is if she were actually a witch, so judgment was rendered against her, and she was burned at the stake.
What could be going on here? The best explanation is that both the FDA and the FTC are on the warpath against any non-drug modality that purports to aid brain function. They have become the enforcers for the pharmaceutical industry, plain and simple.
Additionally, there may be an attempt here to bring cognitive function into the domain of medicine by applying research standards that are more appropriate to medical procedures. Cognitive function has not historically been the concern of medicine. There has been very little concern about chemofog over the years, or the loss of mental function among the elderly that occurs with general anesthesia, or the loss of cognitive function that follows minor head injury, or the loss of mental acuity that attends the use of anti-convulsants. The list goes on.
In view of all of the above, let us return to the issue of over-claiming. If over-claiming on the part of Lumosity is indeed a problem, then it is in the same ballpark with toothpaste that promises you white teeth. If the promise is not borne out, that will become obvious soon enough, and all that will have been lost is the price of a tube of toothpaste. This is not where heavy-handed regulation is needed.
Where were the regulators when we needed them, on Vioxx and Rezulin, on duodenoscopes and other hospital-induced infections, on the use of antibiotics with livestock for non-therapeutic purposes, and on the flawed exoneration of added sugars and of aspartame in food? The largest cause of mortality in the US is the medical service delivery system itself. Here the public is nearly totally at the mercy of the professionals. The second largest contributor to mortality and morbidity in the US is the food industry. This is where the FDA and the FTC should be fully occupied.
They should stop flogging the innovators when they are harmlessly engaged at the frontier of enhancing cognitive function. In that domain, matters will be settled perfectly adequately in the marketplace.
Regulators seek to tame brain training’s ‘Wild West.’ Emily Underwood, Science, 353(6270), p 212-3, January 15 (2016)
Note 1. A family in Austin, TX opened their home to the Wednesday evening prayer meeting of their church. Their African Grey parrot sat by in his cage as the minister intoned sagely and sonorously in his baritone voice. At one point, he calls for a moment of silent prayer. As the silent episode unfolds, the parrot seizes the moment: “You’re full of s***!” he says. The parrot had neatly punctured the pomposity, disrupted the solemnity, and also drawn attention to himself.
Note 2. Suppose I claimed a method for aborting migraines within ten minutes. It would be sufficient to demonstrate that the method works systematically. The fact that any other method does not produce such an outcome would not be news to anyone. It would not add to the body of knowledge and therefore does not have to proved all over again. Indeed large claims demand large evidence, but large effects are their own proof. To insist on the ritual dance of a controlled design in such cases is absurd.