Neuroenhancement for sale: assessing the website claims of neurofeedback providers in the United States

Abstract

Although electroencephalographic (EEG) neurofeedback is a technique that has been in existence for many decades, it has remained controversial, largely due to questions about efficacy. Yet neurofeedback is being widely offered to the public, often at great expense. To date, however, there has not been empirical data on which providers are utilizing neurofeedback, what they are offering it for, and how they are advertising the technique. The present study aimed to fill that gap by systematically analyzing the websites of neurofeedback practitioners in the United States. To that end, we obtained data from four directories of neurofeedback providers, extracting practitioner names, geographical locations, professional training, and website URLs. Only websites offering neurofeedback services (N=371) were included in the next step, wherein two coders independently coded the websites based on a codebook developed from preliminary analyses. We found that nearly all websites (97.0%) contained claims about at least one clinical indication, most commonly anxiety, ADHD/ADD, and depression; however, only 36.0% of providers had either a medical degree (MD) or a doctoral-level degree in psychology. The majority of websites advertised neurofeedback for cognitive (90.0%) or performance (67.9%) enhancement, and roughly three-quarters utilized language related to complementary and alternative medicine (CAM). In sum, there is a considerable divergence between the scientific literature on neurofeedback and the marketing of neurofeedback services to the general public, raising concerns regarding the misrepresentation of services and misleading advertising claims.

Introduction

Although electroencephalographic (EEG) neurofeedback is a technique that has been in existence for many decades (Seifert and Lubar 1975; Sterman and Egner 2006; Kamiya 2011; Thibault et al. 2015; Hamlin 2018), it has remained controversial, largely due to questions about efficacy (Ramsay 2010). Scholars have criticized the poor design of studies, particularly the lack of blinding and sham controls (Arns et al. 2016; Cortese et al. 2016; Thibault and Raz 2016; Schabus et al. 2017; Thibault et al. 2017). Of studies that have utilized double-blind and sham controls, nearly all have found equivalent improvements in both neurofeedback and sham groups (e.g., Logemann et al. 2010; Arnold et al. 2012; Vollebregt et al. 2014) suggesting that the effects of neurofeedback are not neurofeedback-specific but likely due to psychosocial suggestion or placebo (Cortese et al. 2016; Thibault et al. 2016; Arns et al. 2016; Schabus et al. 2017; Thibault and Raz 2017). Indeed, previous work has found that contextual factors, such as patient expectations, healthcare setting, social cues, and treatment modality can all influence treatment outcomes (Di Blasi et al 2001; Miller and Kaptchuk 2008; Rossetini et al. 2018). The lack of effect of neurofeedback beyond a placebo has been underscored by two recent studies—double- and triple-blind, sham-controlled studies of EEG neurofeedback for insomnia and attention deficit hyperactivity disorder (ADHD), respectively—which found equivalent improvements in both real and sham neurofeedback groups (Arns et al. 2016; Cortese et al. 2016; Schabus et al. 2017; Thibault et al. 2016; Schönenberg et al. 2017; Thibault and Raz 2017). In addition, early reports from the much-touted, National Institute of Mental Health-funded, International Collaborative ADHD Neurofeedback (ICAN) study—a multi-site, double-blind, sham-controlled study of neurofeedback for ADHD in children—have indicated that both real and sham neurofeedback groups showed equal improvements (Arnold 2018).

Despite questions about efficacy, EEG neurofeedback is being widely offered to the public, often at great expense. A typical course of neurofeedback consists of up to 40 sessions, each lasting 20–30 minutes, during which a client attempts to self-regulate his or her brainwaves, often via a video game interface (Hamilton 2010; Hammond 2011; Kamiya 2011; Thibault et al. 2015; Hamlin 2018). The total cost of neurofeedback varies by provider and location, and can range from $3,000–10,000 (Ellison 2010; Thibault et al. 2015). In the United States (U.S.), these fees are not regularly reimbursed by insurance, and thus the cost is often passed to the client (Rosenthal 2011; Hamlin 2018). Neurofeedback is currently not recommended as a treatment by any medical association.

EEG neurofeedback received renewed attention in 2017, when media outlets reported that U.S. Education Secretary Betsy DeVos had invested millions in the neurofeedback franchise Neurocore (Fink et al. 2017; Boser 2017). Subsequently, Neurocore’s advertising claims—of 90% improvement rates for indications such as depression, anxiety, and ADHD (Fink et al. 2017)—became the subject of an inquiry from the National Advertising Division (NAD), which found that the company did not have sufficient evidence to substantiate its claims (BBB 2018). Although the company promised to reign in its claims (NARB 2018), in November 2019 the non-profit advertising watchdog Truth in Advertising filed a complaint about Neurocore with the Federal Trade Commission, alleging that the company has continued to make unsubstantiated claims (Mueller 2019; Smith and Patten 2019).

Though the Neurocore case garnered national attention, there are currently hundreds, if not thousands, of independent providers offering neurofeedback services to the general public. As of December 2019, the website for the International Society for Neurofeedback (ISNR) listed 717 members worldwide (ISNR 2019), though the organization estimates that “15,000 clinicians world-wide” are using neurofeedback (ISNR 2017). The prevalence of neurofeedback is particularly concerning from an ethical and legal perspective, given the numerous questions raised in the academic literature about efficacy, as well as conflicts of interest amongst providers (Thibault et al. 2017, 2018; Thibault and Raz, 2017). Though many neurofeedback providers appear to be offering services for a wide array of both clinical and non-clinical (e.g., cognitive enhancement) indications, to date there have not been any empirical data on what kinds of providers are utilizing neurofeedback, what they are offering it for, and how they are advertising the technique.

The present study aimed to fill that gap by systematically analyzing the websites of neurofeedback practitioners in the U.S. In particular, we sought to characterize the training of neurofeedback providers, the clinical and non-clinical indications advertised, the target audiences, and the language used to describe neurofeedback (e.g., hype language, and words related to complementary and alternative medicine). In doing so, we hoped to shed light both on the social phenomenon of neurofeedback, as well as the concomitant ethical and regulatory implications.

Methods

We identified four directories containing listings of neurofeedback providers: two from biofeedback/neurofeedback societies (Association for Applied Psychophysiology and Biofeedback, AAPB; and International Society for Neurofeedback and Research, ISNR); one from a certification-granting body (Biofeedback Certification International Alliance, BCIA); and one from an organization providing neurofeedback educational services (EEG Education and Research Inc., EEGER). Although these directories contain international listings, we chose to focus on providers based in the U.S. in order to constrain our subsequent analysis of ethical and legal issues to a particular sociopolitical context. In February 2019, we extracted provider names, geographic locations, professional training, and website URLs for providers across these four directories (Fig. 1). After removing duplicates, 1,755 providers remained, of whom 694 had listed a website (Fig. 1).

Fig. 1.

Flow diagram depicting the process of identifying websites of neurofeedback providers based in the U.S.

Next, two coders independently assessed all websites. Those that did not offer neurofeedback services were excluded (n=193), along with non-working websites (n=83) and websites that did not have enough information to code (e.g., those that only had listings of practitioner names and/or locations; n=8). Websites for clinics based outside the U.S. were also removed (n=2). This screening process yielded 371 eligible websites, which were included in the final sample for qualitative coding.

Two coders developed a codebook iteratively based on an examination of a subset of 100 websites; all codes were discussed and confirmed with a third researcher. Codes were generated for ten major themes, including clinical and non-clinical indications (e.g., cognitive enhancement); clinic type (individual or multi-provider); target audience; interventions offered; language indicating relationship to complementary and alternative (CAM) medicine; gym language (i.e., “train your brain”); hype language (i.e., “miracle cure”); patient testimonials; and whether the provider was selling a product or device (see supplementary material for full codebook).

Two coders each coded approximately half of the websites. A subset of websites (n=90) was coded by both coders to assess inter-rater reliability, which was found to be 86.6%.

All codes were manually inputted in Microsoft Excel. Quantitative data was analyzed using SPSS Statistics v26 (SPSS Inc., Chicago, IL) to generate variable frequencies and descriptive statistics.

Results

Neurofeedback clinics in our sample were geographically distributed across the U.S., with hotspot clusters in Los Angeles, Dallas, Chicago, and New York (Fig. 2). Most were single-provider clinics, though a subset of websites contained clinics with multiple providers (n=21, 5.6%) and/or locations (n=6, 1.6%). Of the 390 total providers in our sample (19 websites had two providers, two websites had three providers, and four providers had two websites), only 3.9% had a medical degree (MD) and 32.1% had a doctoral-level degree in psychology (Fig. 3; also see supplementary material). Most providers (n=312, 80.0%) had a neurofeedback certificate (Fig. 3). The most common certification was a BCN (board certified in neurofeedback; n=289; 74.1%), which is issued by a biofeedback organization, BCIA. A small number of providers had a QEEG-D (n=15; 3.9%) or QEEG-T (n=6; 1.5%), which are issued by the International QEEG Certification Board (IQCB 2019).

Fig. 2.

Geographic distribution of the neurofeedback clinics in the United States (N=378) that were included in our sample. The total number of clinics depicted is greater than the number of websites in our sample because 5 websites listed two geographic locations and one listed three locations.

Figure 3.

Degrees, licenses, and certifications of neurofeedback (NF) providers in our sample (N=390). Percentages are >100% because some providers listed more than one degree, license, or certification.

Nearly all websites (97.0%) contained claims about at least one clinical indication, most commonly anxiety, ADHD/ADD, and depression (Table 1). The vast majority (90.0%) made statements about the potential of neurofeedback for cognitive enhancement (Table 2), most commonly for improvements in focus and concentration. Other non-clinical indications that providers advertised for included mood, relaxation, and improved performance (Table 2). With regard to claims about performance enhancement, providers often claimed that they could help clients achieve “peak performance.”

Table 1.

Top clinical indications referenced by neurofeedback clinics in our sample (N=371). 97.0% of websites advertised for at least one clinical indication. Clinical indications referenced by <10.0% of websites are not depicted.

	n (%)
Anxiety	340 (91.6%)
Attention-deficit disorder (ADD)/attention-deficit hyperactivity disorder (ADHD)	318 (85.7%)
Depression	303 (81.7%)
Post-traumatic stress disorder (PTSD)/trauma	252 (67.9%)
Head injury/traumatic brain injury (TBI)/concussion	238 (64.2%)
Sleep disorders	223 (60.1%)
Autism	205 (55.5%)
Pain	185 (49.9%)
Learning disorders/dyslexia/dyscalculia	177 (47.7%)
Migraine	177 (47.7%)
Addiction	151 (40.7%)
Epilepsy/seizures	128 (34.5%)
Obsessive-compulsive disorder (OCD)	108 (29.1%)
Stroke	102 (27.5%)
Headache	74 (19.9%)
Asperger syndrome (AS)	69 (18.6%)
Fibromyalgia	64 (17.3%)
Bipolar disorder	58 (15.6%)
Chronic fatigue	55 (14.8%)
Substance use disorder	47 (12.7%)
Tourette syndrome	46 (12.4%)

Table 2.

Top non-clinical indications advertised by neurofeedback clinic websites (N=371). 98.9% of websites advertised for at least one non-clinical indication.

	n (%)
Cognitive enhancement indications
Focus/concentration	284 (76.5%)
General cognitive enhancement	275 (74.1%)
Memory	164 (44.2%)
Learning	128 (34.5%)
Number of websites advertising at least one of the above cognitive enhancement indications		334 (90.0%)
Mood and wellness indications
Mood/emotion	276 (74.4%)
Relaxation/destress	263 (70.9%)
General quality of life	230 (62.0%)
Sleep	167 (45.0%)
Number of websites advertising at least one of the above mood and wellness indications		343 (92.5%)
Performance indications
General performance	239 (64.4%)
Athleticism	146 (39.4%)
Number of websites advertising at least one of the above performance indications		252 (67.9%)

Three-quarters of websites contained specific language targeting parents (e.g. “Regardless of whether your child is on medication, neurofeedback can help”). Other targeted groups included athletes, executives/businesspeople, and artists/musicians—for whom “performance enhancement” was typically offered—as well as the elderly (Fig. 4). Amongst websites targeting veterans, 88.9% offered services for post-traumatic stress disorder (PTSD).

Fig. 4.

Target audience of neurofeedback websites in our sample (N=371). Percentages are >100% because some websites had multiple target audiences. Audiences targeted by <10.0% of websites are not depicted.

Nearly 80% of websites contained language that related the brain to fitness, exercise, or training (Table 3). Approximately three-quarters of websites utilized language related to complementary and alternative medicine, such as “healing,” “holistic,” and “natural.” A quarter of websites utilized hype language, promising exaggerated benefits of their services, while almost half featured patient testimonials (see Table 3 for examples).

Table 3.

Use of gym language, CAM language, hype language, and patient testimonials in our sample (N=371).

Type of Language	n (%)	Example
Gym language	296 (79.8%)	“Think of it as exercise for your brain!” “…it is like taking your brain to the gym and giving it a workout.” “Mentally Fit? When was your last BrainPhysical?”
Complementary and alternative (CAM) language	282 (76.0%)	“fun, pain-free, natural alternative to medications” “…involves a holistic emphasis on body, mind and spirit.” “An integrative approach in medicine/healing…”
Hype language	90 (24.3%)	“Life-Changing Neurofeedback Training” “…our clients begin to live healthy, productive, and more meaningful lives quicker, and sometimes, to them, it seems like magic!” “Discover the miracle of brain wave training; We witness miracles every day and hope you too can find more happiness in your life through neurofeedback!”
Patient testimonials	163 (43.9%)	“This is the most amazing therapy I have ever had the privilege to witness and feel.” “If you’re at all serious about your emotional and spiritual growth, this is going to absolutely blow your mind.” “Neurofeedback is an investment in my health and is giving me my life back.”

Most clinics offered other services in addition to neurofeedback, including psychotherapy (n=266; 71.7%), biofeedback (n=164; 44.2%), heart rate variability training (n=149; 40.2%), and eye-movement desensitization and reprocessing (EMDR; n=70; 18.9%). Some clinics also offered forms of brain stimulation (n=76; 20.5%), such as cranial electrotherapy stimulation (CES) and transcranial direct current stimulation (tDCS). In addition, 11.3% sold products via their websites (e.g., books, at-home brain training devices, heart rate variability monitors), either directly or through an external link.

Discussion

There is a considerable divergence between the scientific literature regarding the efficacy of neurofeedback and the marketing of neurofeedback services to the general public. Though more than 80% of websites in our sample offered neurofeedback for depression, anxiety, and ADHD, there is no rigorous research (i.e., double-blind, sham-controlled studies) that supports the efficacy of neurofeedback beyond a placebo for these indications (Thibault et al., 2015). Furthermore, at least some providers appear to be offering services for indications for which there is even less published data, such as Asperger’s, bipolar disorder, and Tourette’s.

The provision of neurofeedback for clinical indications is even more concerning given that the majority of providers in our sample do not have a medical degree or a doctoral degree in a relevant field such as psychology. While the BCN certification that most providers reported having does require a significant investment of time—a university-level neuroanatomy course, 36 hours of didactic education, 25 hours of mentoring, and 100 neurofeedback sessions—only two hours of training are devoted to the “research evidence base for neurofeedback” (BCIA 2004; BCIA 2019). Furthermore, although the BCIA requires a bachelor’s degree in a “relevant health care field” to be a candidate for BCN certification, it interprets “health care” broadly to include speech pathology, physical therapy, and sports medicine; even those with a master’s degree in music education are considered prospective candidates (BCIA 2017). The requirements for the QEEG certification are even less stringent, requiring only 10 hours of mentoring (IQCB 2019).

From a regulatory perspective, it should be emphasized that the BCN is a voluntary credential that allows an individual to “demonstrate entry level competency” to deliver neurofeedback services; it “provides no legal authority to practice” (BCIA 2015). Indeed, certification differs from licensing, which in the U.S. is a legal requirement for certain professions (e.g., physicians, psychologists, and nurse practitioners) that defines which activities are considered to be within one’s professional “scope of practice.” For example, a speech pathologist advertising neurofeedback services for the treatment of depression would likely be in violation of state law for operating outside her scope of practice and for practicing medicine without a license (see, e.g., Cohen 1998; 2014). Although a detailed analysis of individual providers is outside the scope of this paper, it is concerning that 97.0% of websites in our sample advertised for at least one clinical indication while only 36.0% of providers had either an MD or a doctoral-level degree in psychology.

Notably, the legal issues regarding the practice of medicine do not apply to providers who are merely offering “enhancement” services and not claiming to treat a disease or condition. However, the provision of neurofeedback for cognitive and performance enhancement raises ethical concerns, given that the evidence is mixed or lacking. Although one multi-part review found support for the use of EEG neurofeedback for enhancement (Gruzelier 2014a, 2014b), more recent work has criticized the variability in methodology and the lack of placebo controls (Dessy et al. 2018), raising questions regarding whether neurofeedback can truly enhance performance (Mirifar et al. 2017, 2018; Xiang et al. 2018). While it is not unethical per se to offer the public an experimental treatment, the provision of such services requires informing clients of the mixed evidence and of the experimental nature of the procedure (at minimum). Given that a quarter of websites in our sample utilized hype language, and that 43.9% made use of patient testimonials, it seems likely that at least some of these providers are not accurately representing the current state of the science regarding enhancement. Still, given that neurofeedback may result in behavioral improvement—even though such effects may be due to placebo—more fraught ethical questions center on the morality of offering a placebo directly to the public (Miller et al. 2004; Miller & Colloca 2009; Foddy 2009).

In practice, the ethical issues regarding neurofeedback parallel those surrounding the provision of other complementary and alternative medicine (CAM) services, in which providers offer therapies for which there is little or mixed evidence (Ernst et al. 2004; Racine et al. 2016; Murdoch et al. 2018). Although major neurofeedback societies such as ISNR and AAPB do not explicitly refer to neurofeedback as CAM—both highlight linkages to medicine, psychology, and nursing (AAPB 2011; ISNR n.d.)—three-quarters of providers in our sample seemed to align themselves with CAM, as evidenced by the frequent use of terms such as “holistic,” “natural,” and “drug-free.” Indeed, many websites portrayed neurofeedback an alternative to medication or pharmaceuticals, as an effective procedure that “our doctors aren’t telling us about,” (Michigan Brain Health n.d.). The linkage between neurofeedback and CAM was also evidenced by the provision of other therapies that are not commonly used in mainstream medicine—such as heart-rate variability training and brain light helmets (Center for Brain 2019)—as well as the sale of pulsed electromagnetic stimulation devices (to “reestablish healthy electromagnetic exchanges”; Krieger 2016), healing floor mats (to “help remove toxins that may contribute to neurological illnesses,” (Thompson Neurofeedback n.d.), and supplements that purportedly boost brain health (Better Brain Balance 2015). One website in our sample even advertised the use of neurofeedback on animals, describing how the procedure has been applied to horses, dogs, a cat, and a rooster (Stone Mountain Center n.d.).

This study has a number of limitations. First, our sampling method did not capture all neurofeedback providers, rather only those that were listed in one of the four directories we identified. However, by casting our initial net widely—and later excluding websites if they did not offer neurofeedback services—we aimed to be as broadly inclusive as possible. Second, given that we extracted provider information from existing directories, it is possible that some data was missing or outdated. Third, on some websites, multiple services were offered and it was sometimes unclear whether providers were claiming that neurofeedback specifically could treat those indications, or whether the provider, using a variety of methods including neurofeedback, was claiming to be able to treat those indications. Both coders independently estimated that this uncertainty was present in approximately one-fifth of websites. Still, given that we found overwhelmingly high numbers (>70%) for many advertised clinical and non-clinical indications, and given that neurofeedback was often the primary technique offered on these websites, our results are likely representative of advertised indications amongst neurofeedback providers.

While there has been much debate over the efficacy of neurofeedback in laboratory settings, our paper provides the first empirical examination of how it is currently being offered to the public, at least in the U.S. Although prior work has focused on ethical and regulatory issues regarding the sale of neuroenhancement devices (Maslen et al. 2014; Wexler 2015; 2018; Ienca et al. 2018; Kreitmair 2019; Wexler and Thibault 2018; McCall et al. 2019), this study lends a critical eye to the sale of neuroenhancement services. While in principle, neurofeedback is worthy of receiving additional attention if tested with solid study designs and by independent researchers, given the current state of the science, there remain questions regarding the efficacy beyond a placebo for EEG neurofeedback. Thus, the results of our study—which point to exaggerated claims for indications for which there is little rigorous research—raise concerns regarding the misrepresentation of services and misleading advertising claims. This is especially notable in light of the fact that some clinics appear to be specifically targeting vulnerable populations, such as parents of children with disorders, and given that the financial costs of neurofeedback are not typically covered by insurance. Regulators, scholars, and ethicists should focus additional attention not just on the scientific basis of neurofeedback, but also on how this experimental technique is currently being marketed and sold to the general population.