Method, examined

What a credible manifestation experiment looks like, and why it has never run

By Andrey Tadevosyan · Published 4 September 2026

For a claim that a hundred years of books insist is a law, manifestation has been tested remarkably little. That is partly because the word bundles several claims, and partly because the one that matters most is easy to state and awkward to run. This page is the two experiments that would settle most of the argument, the ways a study fools itself, and the version a single person can do at a kitchen table.

How would you scientifically test manifestation?

With two separate experiments, because ordinary goal studies cannot distinguish cosmic causation from behaviour. A field trial compares outcome-only visualisation, a process-and-plan protocol, and a neutral control on objective logged outcomes over twelve weeks. A direct-attraction test forms an intention before a secure random process generates the outcome, with preregistration, blinding and independent replication. The first has partial answers; the second has never come out in favour.

Why one experiment is not enough

A study that hands people a manifestation routine and measures whether their lives improve cannot tell you whether the universe helped. If the routine works, it may have worked through attention, planning and persistence, which are ordinary and which the attraction essay lays out rung by rung. To test the direct claim, intention reaching an event, you need an outcome that no behaviour can touch. So there are two experiments, and they answer different questions.

Experiment one: the field trial

Recruit people with a real goal in one domain: a business, a job search, a health behaviour, a savings target, a social life. Randomly assign them to four conditions. A neutral control that journals about something unrelated. An outcome-manifestation condition: visualise the result, feel it as done, affirm it. An agency protocol: picture the process, name the obstacle, write if-then plans, act the same day. And the agency protocol plus a personally meaningful talisman used as a daily cue. Run it for twelve weeks. Measure objective, logged things: applications sent, conversations started, workouts done, money moved. Not feelings, not signs.

Three comparisons carry the weight. Agency versus outcome-manifestation asks whether process-and-plan beats the popular method. Talisman versus agency asks whether the object adds anything. Outcome-manifestation versus control asks whether the popular method helps, does nothing, or backfires. The existing evidence makes a prediction for each: the visualization essay for the first, the lucky charm essay for the second, and the fantasy research for the third, which points at backfire.

The techniques in the agency arm are not guesses. The pooled evidence on mental simulation and on contrasting-plus-planning is what the arm is built from.

Experiment two: the direct test

This one is smaller and stricter. A participant forms an intention toward one of two harmless future outcomes. After the intention is recorded, a secure, independently audited random process generates the outcome. The participant cannot see or touch the generator. The hypothesis, the stopping rule, the exclusions and the analysis are preregistered. The sample is large enough to detect a small deviation. Every trial is reported. And before anyone interprets a positive result, a separate laboratory runs it again.

That design removes every rung of the ladder. No attention, no warmth, no extra attempts can reach a random number generated afterwards. If intention moves the result, it moved it directly, and that would be a genuine anomaly worth the world's attention. Versions of this experiment have been run in adjacent fields for decades. None has produced a result that survived independent replication. That sentence is the entire empirical status of the fourth tier, and it is why this site calls it faith.

Ten ways a study fools itself

Reverse causality: successful people adopt the belief afterwards. Confounding: hope, resources, education and support drive both belief and outcome. Selective reporting: fulfilled intentions get told and failures do not. Regression to the mean: rituals start at low points, and low points are followed by better ones regardless. Base-rate neglect: an ordinary event feels significant because an intention preceded it. Behavioural leakage: the person nudges the “external” outcome without noticing. Multiple comparisons: enough intentions and enough symbols will eventually produce a coincidence. Outcome flexibility: the wish is redefined after the result. Demand: participants behave as the researcher hopes. And commercial incentive: whoever tells the story profits from it.

Every one of those applies to the testimonial you read last week. A study is just a testimonial with those ten holes closed, and the reason manifestation research has been so thin is that closing them is work.

The version you can run alone

You cannot randomise yourself, but you can close most of the holes. Write the criterion before you start, specific enough to fail. Fix the period; thirty days is enough to see something and short enough to finish. Log attempts, not just results, so exposure is visible. Log misses as well as hits, so selective memory has nothing to work with. Choose an objective outcome, something a stranger could verify, rather than a feeling. And decide in advance what would count as it not working.

That is exactly what the free 30-day experiment on this site is, and the experiment designer writes the preregistration for you from four answers. It will not tell you whether the universe listens. It will tell you, after thirty days, which of the five meanings of attraction was at work in your own case, which is the only version of the question that is yours to answer.