PEAR and the Tiny Effect

Two chairs

For nearly three decades a Princeton lab ran millions of trials testing whether intention could shift random machines. It reported a tiny effect; a 2000 replication couldn't reproduce it. Does the number still mean something?

The question

In 1979, Princeton engineering dean Robert Jahn founded a lab to test whether human intention could subtly bias machines generating random numbers. Over nearly thirty years and millions of trials, it reported a tiny statistical deviation from chance. In 2000, a strict three-lab replication using the same equipment failed to reproduce it. Who reads the number right?

Round 1

The chair that trusts the effectAcross millions of trials over decades, the combined deviation from chance was small but consistent — not the kind of thing that should show up by accident.

The chair that trusts the replicationIn 2000, Princeton, Freiburg and Giessen ran the exact same protocol on the exact same machines, and none of the three found a significant effect.

Round 2

The chair that trusts the effectRobert Jahn, the founding engineer, argued that dismissing a tiny, repeated effect just because it's tiny is its own kind of bias.

The chair that trusts the replicationAcross millions of trials, even a microscopic bias in equipment or analysis can look like a signal. That's exactly the range PEAR's effect lived in.

Round 3

The chair that trusts the effectA null replication in three labs doesn't erase twenty-eight years of data — it just means the effect, if real, is fragile and hard to pin down.

The chair that trusts the replicationTwenty-eight years of trials is a long record, not a replication. Only independent labs finding the same effect could confirm it, and in 2000 they didn’t.

Where are you sitting?

The chair that trusts the effect The chair that trusts the replication Neither — I’m still listening

Held open

PEAR ran nearly three decades of trials and reported a small, consistent statistical deviation. A 2000 replication across three independent labs, using identical equipment, found no significant effect in any of them. Most physicists never accepted the original result.

A tiny number, defended and doubted, for twenty-eight years.

scroll to hear both

1 / 7