The Mind Science of Practice

These are notes kept over eight months while trying to work out whether any of this is doing what people say it does. The short version is that the mechanism, such as it is, is ordinary and measurable, and that the gap between what a practitioner feels and what an instrument records is much wider than most people suspect. Both halves of that sentence are worth sitting with. I am not writing this to settle the question. Nothing in the existing literature settles it, and eight months of a personal log settles nothing either. What follows is a record of what a measurement actually looked like, how big the autonomic change was, where expectancy ran out, and why a particular deck of cards taught me more about my own reasoning than about anything else.

What the measurements actually showed

The most reliable finding over eight months is boring and real. Performing a fixed ritual sequence reliably shifted my own physiological state. Heart rate variability moved in a consistent direction once the routine was established, breathing rate changed during and after, and skin conductance dropped across the second half of a session rather than the first. These are ordinary autonomic responses and they happened reliably enough that I could count on them.

What I could not detect was anything outside myself. I have no way to measure whether a sigil affected a situation I was not present for, and no method I tried produced data of a kind that would let me say. So the honest summary is that the ritual reliably changes the practitioner, that the change is measurable, and that the question of whether the change reaches outward is exactly the question my instruments do not address.

The autonomic change, and how big it is

Size matters more than direction. The changes I recorded were not dramatic. Heart rate variability moved by amounts that a single restless night could also produce. Breathing rate changes were clearer because I was measuring at fixed points, but a change of two breaths per minute is not a transformation. Anyone telling you their practice produces a large physiological shift is describing something other than the standard measurable effect.

The place where the effect is more reliable than the numbers suggest is subjective state. By week six the sequence itself had become reliably settling, in the sense that I felt more settled afterwards, and that effect did not depend on the aim. It was a consequence of doing a predictable thing at a predictable time. That is a genuinely useful thing to have and it is not supernatural, and the two statements are compatible.

Expectancy without belief

The interesting complication in the research is that belief in the treatment is not always required for a measurable physiological effect. Open-label studies, where participants know what they are receiving, have produced effects that the investigators struggle to account for through expectation alone. This complicates the tidy account in which everything is explained by believing you will feel better.

It does not open the door to a mechanism that runs through the imagination. A body responding to a procedure it knows is a procedure is still a body responding to a procedure. The most defensible reading is that some effects are expectancy-driven, some are habit-driven, and some are a mixture of both, and that separating the three requires designs nobody has run on this subject.

Perception as a process that manages hypotheses

There is a well-developed line of reasoning that treats perception not as a window but as an active process of managing hypotheses about what is out there. On this account, expectations do not need to be false to shape experience. They change which hypotheses get maintained, which changes what gets noticed, which changes what gets reported as having happened.

This is the most economical account I have found of my own data, and it is also the most deflating. If my ritual changes my expectations, and my expectations change what I attend to, and attention changes what I remember and report, then I would expect to see results of a certain size and reliability without anything outside the process doing any work. And that is roughly what I see.

The deck that taught me the arithmetic

The most useful object in eight months of this was a deck of cards designed for testing psychic claims, with five symbols, where chance performance is twenty per cent. I used it to test my own impressions, and what it taught me was arithmetic rather than anything about the claims being tested.

The arithmetic is this. With five symbols, a person guessing has a one in five chance of a correct hit. With fifty cards, that is ten correct hits, and on a set of results I find impressive, a competent guesser gets close to the same number. To separate the two you need a hit rate above what guessing produces, on enough trials, and a comparison group. Without those, the number means almost nothing, and it means the same almost-nothing to the enthusiast and to the sceptic, which is why this subject generates so much heat and so little information.

A protocol with a number attached

After about three months I stopped recording impressions and started recording counts. The aim became a countable behaviour, the record became a tally, and the interpretation moved to a separate page written a week later. This change did more for the quality of my data than anything else I tried, and it is the single recommendation I would make to someone starting.

The reason is structural. Impressions are produced by the same process that is being evaluated, so recording them is circular. A tally is not. Once the record is a number that could have come out the other way, the practice acquires the property that matters, which is the possibility of being wrong. Everything else in this area is downstream of that.

The month the effect stopped

In the fifth month the settling effect largely disappeared, and the reason is instructive. What had changed was that the sequence had become routine in a bad sense. I was performing it while doing something else, I had stopped noticing the state shift, and I had started skipping the charge when the fortnight was busy. The ritual had become a formality that I was maintaining rather than a procedure I was undergoing.

The recovery was to make it inconvenient again: change the time, drop two elements, do it in a different room. Within a week the effect returned at roughly its previous size. This is worth recording because it is a warning about a broader pattern. Almost every practice I have looked at degrades not through failure of belief but through the absorption of the procedure into ordinary activity.

Where the mechanism claim outruns the data

The gap is between the internal claim, which is modest and mostly defensible, and the external claim, which is not. A practitioner can say with some confidence that a fixed sequence changes their own state. The same practitioner, with the same evidence, will often also say the sequence affected a situation at a distance, and the second claim has no support whatsoever in anything I produced or could produce.

The reason this is common is that the internal claim is experienced as evidence for the external one. Something happened, therefore something can happen further out. The inference is invalid but it is psychologically very natural, because the internal change is real and the mind does not like to conclude that a real effect stops at the skin. Naming the invalid step is more useful than arguing about the external claim, which cannot be resolved with the evidence available.

What to record so the numbers mean something

Five things, and none of them require equipment beyond a phone camera and a notebook. The aim, written before. The count of completed actions, which is the only outcome measure. The time of day, because the effect I measured was strongly time-dependent. Whether the charge happened, on days it got skipped. And the aim's difficulty rated at the start, because an easy aim completing tells you nothing about the method.

A sixth field is worth adding after a few cycles: what you did that you would have done anyway. This is the column that prevents a good month from being read as a success and a bad month as a failure. Without it you will attribute outcomes to the practice that were caused by the circumstances, and you will do it more often when things go well than when they go badly.

The rituals that produced nothing measurable

Not all of it worked on me and it is worth being specific about which parts. Complex layered invocations produced nothing beyond the baseline settling effect. Elaborate correspondences and planetary timings produced nothing at all, including nothing subjective, which I found surprising and slightly deflating. The one-sentence aim with a plain mark and a fixed time performed best, and the difference was not subtle.

The interesting pattern is that the practices that added complexity added nothing to the outcome while adding a great deal to the experience of having done them. That gap is where a lot of occult tradition lives, and I do not think it is a sign of fraud. Elaborate ritual is doing something for people. It is just not doing the thing it is usually described as doing, and those are two separate questions that the literature tends to run together.

Keeping the practice, dropping the claim

I kept the routine and dropped most of the interpretive frame, and I have not missed the frame. What remains is a reliable settling procedure, a commitment device that works, a recording habit that has made me more honest about my own follow-through, and a genuine interest in a subject with a poor evidence base. That is a reasonable position to be in after eight months.

The dropped part was the sense that these practices reach outward. I do not have anything to offer on that question that would satisfy anybody, including me. What I do have is the observation that the internal effects are real, consistent, and entirely explicable, and that this combination is exactly what a working commitment device looks like when someone describes it in the vocabulary of power.

What I would tell a sceptic

I would say that the measurable effects are of two kinds and I can only measure one. I can see reliable changes in my own autonomic state and in my own follow-through rates, and I can give you numbers for both. I cannot see anything external, I have not tried to design a test that could, and I would not accept my own results as evidence of it.

I would also say that the literature in this area is unusually poor, and that a serious attempt at a controlled design in this literature is something nobody seems to want to fund or publish. So the state of the evidence is not a finding about the practice. It is a finding about the incentives, and I would be careful about reading it in either direction.

Frequently Asked Questions

Do rituals physically change you?

There is a measurable effect and it is not a large one. Fixed sequences reliably produce autonomic changes in the practitioner, including shifts in breathing rate, heart rate variability, and skin conductance, and these are documented in the experimental literature. The changes are real, reliable, and modest in size, and the subjective sense of a large transformation is a separate phenomenon with a separate explanation.

Does belief need to be present for the effect?

Probably not in the simple form the popular account assumes. Open-label studies have produced measurable physiological effects in participants who knew exactly what they were receiving, which means belief in the treatment is not always doing the work. Whether the alternative explanation is habit, expectancy of a different kind, or something unmeasured is not settled, and the honest position is that the simple account is incomplete.

How do you test whether something worked?

Turn the aim into a countable behaviour and count it. Impressions are useless as data because they are produced by the same process being evaluated. A tally is not circular, and it can come out the other way, which is the property you need. Add a column for what you would have done anyway, because that column is what stops a good month being read as proof.

What about the card decks used to test psychic ability?

They are excellent for teaching probability and nearly useless for testing the claims they were designed to test. With five symbols, chance performance is twenty per cent, and a competent guesser produces a number that looks impressive on a small set. Separating the two requires enough trials and a comparison group, and without both, the result means the same thing to everyone regardless of what they already believed.

Can I practise this without caring about the evidence?

Yes, and plenty of people do. The internal effects do not require a settled view about mechanism, and the method works as a commitment device under any interpretation. What the evidence question changes is only whether you should expect the practice to affect things outside yourself, which is the one claim the internal effects do not support.