The instrument cannot have a hypothesis.
Consider what it would mean for a measurement device to believe something. In practice it would mean one of a small number of concrete design decisions: the acquisition parameters were tuned on data from the intervention arm; the output metric was chosen after looking at which of several candidates separated the groups; the processing pipeline includes a step whose parameters were adjusted until the expected pattern appeared. Each of these is a way for a hypothesis to enter the instrument. Each of them destroys the instrument’s usefulness as evidence, because its output is no longer independent of the belief it is being used to test.
This is why the manufacturer of an endpoint is in a structurally different position from the investigator running the study. The investigator has a hypothesis; that is fine, that is what randomization and blinding and pre-registration exist to contain. The endpoint must not have one, because nothing contains it.
So when we say we take no position on whether a given practice produces a physiological effect, we mean something specific: we have no design commitment, no analytic commitment, and no commercial commitment that would be served by the answer coming out one way rather than the other.
Neutrality has to be built, not promised
Anyone can claim to be unbiased. The claim is worth roughly nothing on its own, and readers are right to discount it. What makes agnosticism real is structural — arrangements that would make bias difficult to exercise even if we wanted to. The ones that matter here:
- Identical acquisition across arms. The measurement protocol, device configuration, and session structure are the same for every participant in every condition. The instrument is not told, and cannot infer, what preceded the scan.
- Metrics specified before unblinding. The primary output metric for any study is named in the pre-registration. Selecting a metric after seeing group data is the single easiest way to manufacture a result, and it is not available if the metric is already on the record.
- Analysis by parties without a stake. Where possible, the data should be analyzed by someone who is neither the practitioner nor the manufacturer, working from arm labels they cannot decode until the analysis is locked.
- Nulls are reported. A study that returns no difference between arms is a result, and we will say so. This is the commitment that costs something, and therefore the only one worth much.
- We do not recommend modalities. We do not endorse, rank, or market any therapy, practitioner, or practice. We are not in that business and entering it would compromise the one we are in.
The commercial incentive, stated plainly
Readers should not have to take our neutrality on trust, so it is worth being explicit about where our interests actually lie.
Aurorus is a platform licensing company. Our revenue comes from the use of the measurement instrument by researchers, clinicians, and wellness practitioners. That revenue is the same whether a given study finds an effect or finds nothing — arguably a well-publicized null result serves us better than a positive one, because it demonstrates that the endpoint discriminates rather than merely confirming.
That is a real structural feature, and it is more informative than any assurance we could offer. But it is not a guarantee, and we would not ask anyone to treat it as one. The guarantee, if there is one, comes from the pre-registered endpoints and independent analysis described above, which do not depend on our good intentions.
Four commitments
Measure first. Not persuade, not interpret, not contextualize. Establish what the instrument does, characterize its variability, and only then point it at anything interesting. Most arguments in this field are downstream of the fact that nobody did this.
Pre-specify. The endpoint, the analysis, the sample size, and the stopping rule go on the public record before data collection begins. This is the mechanism that prevents honest people from fooling themselves, which is by far the most common way false findings enter a literature.
Report nulls. A study that finds nothing is a result. We will publish it, and if the instrument turns out not to detect what someone hoped it would detect, that is the finding. This is the commitment that costs something, and it is the one by which the others should be judged.
Take no position. We do not recommend modalities, endorse practitioners, or offer views on whether any given practice works. Our value in this research is precisely that we do not care what the answer is. An instrument with a preference would be a defective instrument.
What this looks like in practice
It looks like a slower and less exciting content program than the alternative. We cannot write the post that says a study proved something, because no study has. We cannot cite a compelling case, because a case is not evidence. We cannot borrow legitimacy from academic centers investigating adjacent questions, beyond noting accurately that the investigation is happening.
What we can do is build the endpoint, specify what a real study would require, be explicit about where our own preliminary work falls short of that standard, and make the resulting numbers available to researchers who are not us.
What agnosticism is not
Three misreadings are worth heading off.
It is not agnosticism about what we measure. We are not uncertain on that point and never have been. The platform is a non-invasive electrophysiological measurement device — it applies a field to tissue and quantifies the resulting optical emission, which is an energy measurement in the same ordinary sense that an ECG or an EEG is one. That is established, it is what the device is cleared as, and it is not the open question. The open question is whether those values shift reliably in response to a given practice. Our neutrality attaches to the answer, not to the instrument.
It is not skepticism in disguise. We are not quietly confident the answer is no and hedging politely. A researcher who is privately certain of a null result and publicly neutral is running the same bias in the opposite direction, and would make the same design errors with the signs reversed. Genuine agnosticism means genuinely not knowing, and finding that state comfortable.
Nor is it credulity in disguise. Declining to dismiss a question is not the same as endorsing an answer to it. The reason we treat these claims as open is not that we find them plausible — several of them we would consider low prior probability, and we say so in the post on intention. It is that plausibility is not how empirical questions get settled, and a measurement company that resolved questions by intuition would have no reason to exist.
Why this is the valuable position
There is no shortage of parties in this space with a view. Advocacy organizations are confident it works. A good deal of mainstream commentary is confident it does not. Both have been confident for decades, and the confidence on both sides has produced almost no data that would move anyone.
The scarce thing is a party with no stake in the outcome holding an instrument that could settle it. That is a narrow role, and it comes with real constraints — we cannot market on efficacy, we cannot tell an appealing story about what the numbers mean, and we have to publish results we would rather not. Those constraints are the product. An endpoint that could be talked out of its answer would not be worth measuring with.
Regulatory note: Aurorus is a non-invasive electrophysiological measurement platform. It was granted FDA De Novo classification (DEN150004) as a Class I non-invasive electrophysiological measurement device. It is not cleared for diagnostic purposes. Nothing in this article claims that any therapy, modality, or practice diagnoses, treats, cures, or prevents any disease.
End · 6 min
Nancy Rizzo Prior to developing the science and technology for Aurorus, Nancy spent 13 years at Mayo Clinic Scottsdale as a board certified radiological therapeutic physicist and cochaired the Complementary and Alternative Medicine program. Nancy is board certified as a Radiological Therapeutic Physicist by the American Board of Radiology (DABR) and holds a Ph D. in Energy Medicine with an emphasis on Biophysics from Holos University.
Follow →