The CEO of Brain.fm recently had an Ask Me Anything (AMA) on Reddit, which brought their company and claims to my attention. While trying to present themselves as a respectable science-based company, when you strip it down they appear to be just another snake-oil company selling overhyped claims not backed by science. The CEO introduced his AMA beginning:
Hey Reddit! I’m Daniel Clark, CEO of Brain.fm. We use AI to generate music specifically to help people focus, relax and sleep, backed by science. In particular, our popular Focus music is useful for people who must get through a lot of work without distraction, procrastination, or boredom.
After a series of appropriately skeptical questions (good work, all), he had to add an edit to the intro:
We’re getting a lot of questions around our scientific claims. We wanted to clarify that this AMA is to announce our NSF grant and share our excitement around our ongoing research, rather than announcing the publication of a journal article (which would come at the end of the grant). Also, wanted to clarify we are not trying to replace medication but on the track to potentially becoming an alternative for many people–if you have ADHD, I am not claiming that you should stop your current treatment, or that we have the cure, but we are excited for the future and want to share where we are headed.
This is a pattern all-too-familiar to SBM.
The marketing strategy
The strategy for marketing dubious medical devices, services, or products is now well-established. First, target a disease or disorder that is either not well-treated by standard medicine, or there is the perception it is not well-treated. But, to maximize potential sales, also claim the treatment is good for the general population by targeting common everyday symptoms. Who doesn’t want better sleep or focus?
In this case the company, Brain.fm, markets music designed by artificial intelligence to entrain the brain (more on that below). You can choose what you want to target: “Focus, Meditation, Relaxation, Naps, or Night Time Sleep.” They also promise results in, “Within 15 minutes or less.”
It also helps to instill distrust in current treatment. The CEO stated in his AMA intro:
ADHD affects over 8M adults in the U.S. alone, with current solutions causing near-epidemic growth of prescription stimulants. An additional 4M do not meet all the criteria to be diagnosed with ADHD, so they can’t be prescribed medication and have no convenient alternatives for treatment.
He then had to walk that back, as stated in the edit above. That is also very common – stating that the company is not making any “cure” claims, to consult your doctor, do not stop any standard treatment, etc. This is all plausible deniability. With regard to supplements, such statements are FDA mandated.
For those companies trying to market their products with the patina of science, a core aspect of the marketing strategy is to emphasize the science, even when you don’t have any. There are three main ways to do this. The first is to focus on the basic science. As I have discussed many times in these pages, you cannot base clinical claims on basic science. The vast majority of clinical hypotheses we develop from basic science turn out not to work. Biology is simply too complex to make these kinds of extrapolations. We always need to confirm our hypotheses with clinical trials.
The Brain.fm website has a white paper and science FAQ that does just this – talk about the basic science and make simplistic extrapolations to their clinical claims.
The second strategy for seeming scientific is to do in-house or paid-for studies that are designed to show positive results. These can be open-label (meaning not blinded), or may not even be controlled. Or they are simply looking at some outcome that is likely to represent just placebo effects. Or they are looking at biological markers that are non-specific and cannot be used to gauge clinical outcome.
What you rarely see are appropriately designed rigorous clinical trials that are actually capable of showing if the product works or not.
I don’t see any clinical studies touted on the Brain.fm website, and they state they do not have any – yet. Which brings us to strategy 3: play up the fact that you have ongoing research. That is exactly what they were doing in their AMA, hyping the fact that they have a NSF grant to do research. The abstract of the grant does not go into detail about the methods, but they describe it as a phase I study. Phase I is usually just a safety and proof of concept study, and is almost always open label. They are not, generally, efficacy trials. Phase II are preliminary efficacy trials, and phase III are definitive efficacy trials.
Research genuinely takes a long time, and if your goal is to just keep your product and claims in a perpetual state of being studied for marketing purposes, you can easily do that. At the very least you can get 10-20 years of marketing out of the “actively being researched” phase before evidence of lack of efficacy emerges, and the current fad fades, at which time you can move on the next one.
That is the core problematic feature of this (and many) health products. The company is actively selling a product with specific claims, supported with the appearance of having science backing, and being actively studied. But it does not have clinical trials showing efficacy or backing any of their specific claims. In an ideal world they would not be allowed to do that – to make or even imply clinical claims without evidence.
The science behind Brain.fm is not very compelling, in my opinion. It is essentially just another version of the brainwave entrainment claim. I have been writing about this for over a decade. In their science FAQ, Brain.fm gives a decent summary of brain entrainment:
When a temporal regularity is present in the environment, neuronal oscillations can synchronize (i.e. entrain) to this external rhythmic stream. This is operationalized by appropriately biasing the high excitability phases of ongoing neuronal oscillations so that they align with the events in the stream.
Essentially there is a rhythm to the basic electrical activity of the brain, as measured by electroencephalogram (EEG). What exactly causes these baseline rhythms is still a matter of research, but basically it results from synchronized activity in centers in the brain that act as the brain’s “pacemaker.” This, in turn, seems to serve the function of keeping the brain functioning – it is like an alerting signal that keeps pinging the cortex and keeping the stream of activity going.
Further, these rhythms can be driven to some extent by external stimuli. During a routine EEG, for example, patients are exposed to a strobe light at varying frequencies. Often we will see what is called “photic driving” which means the background EEG rhythm synchronizes with the strobe light frequency.
This is part of the basic science that Brain.fm uses to support their claims, and it is all legitimate. They also refer to Dynamic Attention Theory (DAT). This is the notion that our attention is dynamically regulated by the brain based on realtime events and stimuli. Whether this specific formulation is the final answer or not is still being researched, but something like this is surely going on. Attention is a neurological phenomenon unto itself, and it is both an important and complex phenomenon that is certainly dynamic.
But their own FAQ hits upon the key question when they ask:
Does altering brainwaves really mean we are altering mental state?
Yes, that is the 64-thousand-dollar question. It is the same question that has been asked for several decades, so it is not a new question. Companies selling devices to alter brainwaves may still be asking this question in 20 or 100 years.
Their answer to their own question I found to be rambling and confused, without giving a definitive answer but throwing a lot of jargon at the reader (which, it seems, is the point). They conclude:
And specific patterns of neuronal oscillations now start to be seen as fingerprints of cognitive processes. Mental states can thus be altered via neuronal oscillations, but such an alteration has to reach a critical level to be consciously perceived. Indeed, neuronal oscillations are in a constantly fluctuating, while we have a subjective impression of mental stability and continuity.
Got that? I don’t think they actually answered the question, just restated the basic science as if it does.
Having followed these claims and the science for years, I think the answer is – probably not. The essence of the “brainwave entrainment” product claims has always been, that since different mental states correlate with different EEG patterns, if we can force those EEG patterns we will put the brain into the associated state. This has never been a plausible hypothesis, as it is attempting to reverse the arrow of causation in a dubious way.
Looking specifically at attention – it is a complex dynamic process. The brain uses many complex evolved processes to determine how to focus our attention, which is related to how it decides what information to filter out, and what information to bring to our conscious awareness. This complex process is partly responding to external stimuli (of course). So, it is plausible that external stimuli can affect our attention.
Playing music, for example, can be relaxing or distracting. There seems to be great individual variability here as well, as people have different preferences. Further, there are several variables, including type of music, volume, and the specific task one is attempting to engage in.
The notion that some algorithmically designed music can force our brains into a sustainable attentive state enabling us to focus on whatever task we have is implausible to say the least. Individual trial and error would probably be a vastly superior method for determining what kind of music is best to listen to with various tasks.
Further, brain entrainment probably has absolutely nothing to do with the effects of various kinds of music on attention, relaxation, sleep, and other tasks. I have not seen any evidence that external brain entrainment has any exploitable clinical effects. In fact, if there is any attentional affect at all (which I doubt) it would be a coin flip whether it would be a help or hindrance.
This entire approach, which is a “brain hacking” approach, is flawed, in my opinion. There is no simplistic way to hack brain function. You can’t just slap on headphones and improve your attention, precisely because it is a complex evolved function with complex feedback mechanisms. Similarly, you can’t improve your general intelligence by doing a puzzle (even if it is “scientifically designed”). This is the same “brain hacking” fiction that makes for great ad copy, but bad science.
At the very least I see such claims as highly implausible, and not something we can conclude from the existing basic science research. I have no problem with doing clinical research, and maybe we might learn something about how the brain regulates attention that can be exploited. But marketing products in the meantime based on this dubious support is deceptive and amounts to little more than typical pseudoscientific snake oil.
Edit 03/23/2018 – A spokesperson for Brain.fm sent me an e-mail with this notice: “Brain.fm uses AI to generate music that helps people focus, relax and sleep, and so the company does claim to improve concentration. But Brain.fm has not made any claims about ADHD.”
This is exactly the strategy I was detailing above. In the AMA the CEO claims Brain.fm helps people, “focus, relax and sleep, backed by science.” He then has a paragraph about how serious a problem ADHD is, and then another paragraph about their research grant to study its effects, “regardless of whether they meet diagnostic criteria for ADHD.”
In my opinion this is consistent with one deceptive strategy in the marketing of these kinds of treatments – step right up to the line claiming to treat a specific disorder, and create the strong impression that your product can treat the disorder, but never make the claim explicitly for plausible deniability and to be within narrow compliance of FDA regulations.