What Ganzfeld Telepathy Experiments Found

Some ganzfeld telepathy experiments have been reported to show a small surplus of correct matches over the one-in-four chance rate built into the design. Meta-analyses disagree because they include different experiments, apply different quality filters, and start from different ideas of what counts as a replication.

Sit with that mixed record instead of forcing a yes or a no out of it.

What is a ganzfeld telepathy experiment?

If you have heard of ganzfeld, you have probably pictured halved ping-pong balls, red light, and white noise. That picture is accurate enough.

Photograph courtesy of Nealparr / Wikimedia Commons, Public domain

A receiver sits with even, featureless stimulation: translucent hemispheres over the eyes, a red wash of light, and a steady hiss in the headphones. The point is to quiet ordinary sight and sound so a faint impression has more room to appear.

A sender, in a separate room, views a randomly chosen target, often a still image or a short clip, and tries to convey it. After the session, the receiver sees four options, only one of them the true target, and makes a choice or a ranking.

Chance, by design, is one in four. The test asks whether the receiver's choice lines up with the hidden target more often than that baseline.

Follow it as a lab attempt to give a claimed mind-to-mind signal a fair, boring trial, whether you believe in telepathy or not.

What did the experiments find?

Through the 1970s and 1980s, Charles Honorton and colleagues reported ganzfeld series in which receivers picked the correct target more often than chance. Ray Hyman argued that problems in randomization, possible sensory leakage, and incomplete reporting could account for the surplus (Hyman, 1985).

Photograph courtesy of Rouven Schäfer, CC BY-SA 3.0

Honorton, answering in the same journal, said the effect survived those criticisms (Honorton, 1985).

In 1994, Daryl Bem and Charles Honorton published a review in Psychological Bulletin of later autoganzfeld work, which used computers to tighten randomization and target handling. They concluded that the results pointed to an anomalous process of information transfer (Bem & Honorton, 1994).

That paper is why you still hear ganzfeld named as the main laboratory case for telepathy. Later work kept arguing with the claim.

Julie Milton and Richard Wiseman reviewed a newer batch of studies in the same journal in 1999 and reported that the overall effect did not come through (Milton & Wiseman, 1999). Storm, Tressoldi, and Di Risio later found a significant effect in a free-response meta-analysis that included ganzfeld (Storm, Tressoldi, & Di Risio, 2010). Hyman answered that the analysis concealed more than it revealed (Hyman, 2010).

Across decades, a statistical tilt keeps appearing in some of these databases, the methods fight is real, and independent reviewers still do not agree that the tilt is telepathy.

Why do the meta-analyses disagree?

They disagree because they are not scoring the same exam.

A meta-analysis follows the studies it admits, the quality cuts it applies, and the question it poses. Ganzfeld reviews have split on all three.

Some analyses keep older sessions and count a hit as a hit. Others drop work with weak randomization, possible sensory cues, or flexible stopping rules, and the surplus shrinks or disappears. Bem and Honorton saw a signal when the tighter autoganzfeld studies were the story. Milton and Wiseman did not, once later failures to replicate were in the mix.

Controversial fields also have a file-drawer problem: striking sessions are easier to publish than quiet ones. Labs that expect a hit and labs that expect a miss do not always return the same pattern. Storm and colleagues cast a wider free-response net than a ganzfeld-only review, which is part of why they and Hyman talked past each other in 2010.

A meta-analysis is a pile of human decisions about evidence, which is why two of them can look at ganzfeld and walk away with different numbers.

Hyman and Honorton did publish a joint statement on the safeguards future ganzfeld tests should meet (Hyman & Honorton, 1986). They could agree on method and still disagree on what the numbers meant.

What should you do with a mixed record?

You will not get a clean yes or no out of this literature this week.

Read one primary paper instead of a recap. Bem and Honorton (1994) and Milton and Wiseman (1999) are both short, both in Psychological Bulletin, and they disagree in language you can follow. Sit with both. Notice where hope or dismissal fills a gap the data do not cover.

If you want a structured way to look at your own inner pattern after that, you can Get your Soul Blueprint. Use it as self-reflection rather than a lab result, a diagnosis, or a forecast.

Classical psi stays interesting because the studies never went quiet and never settled, so keep the interest and skip the theater.