Plain-language answer
This site’s working position is that no paranormal mechanism (psychic ability, precognition, an “acausal connecting principle”) has been established by the strongest available evidence to date — not because such claims are dismissed without a hearing, but because the single most rigorous, mainstream-published attempt to demonstrate one, subjected to the scientific process’s own follow-up scrutiny, did not hold up. This page walks through that case in detail, deliberately, because it is the fairest available test of the claim, not the weakest one.
Why it matters
Coincidence stories are frequently offered as evidence for a paranormal connection between events. Rather than surveying every such claim in the abstract, this page focuses on the strongest, most carefully executed scientific test of a closely related claim — precognition — because a methodologically serious attempt that still failed to replicate is far more informative than dismissing an anecdote out of hand or citing only weak studies as if they represented the field’s best work.
The strongest case: Bem’s precognition experiments
In 2011, psychologist Daryl Bem published nine experiments in a mainstream, peer-reviewed psychology journal reporting statistically significant evidence that ordinary participants could sense future, randomly determined events slightly better than chance would predict — for example, showing improved recall for word lists they would later be asked to practice, as if the future practice were influencing past memory.(Bem, 2011) This was a serious, carefully designed set of studies, using standard statistical methods, that passed the same peer-review process as any other psychology paper of its era. Its publication is precisely the kind of result — extraordinary, but produced through ordinary scientific channels — that this site’s evaluation method says deserves careful, good-faith follow-up rather than reflexive dismissal.
The follow-up: what happened when others tried to replicate it
That follow-up came quickly and rigorously. A team of researchers ran seven pre-registered experiments, involving over 3,000 participants combined, specifically designed to replicate two of Bem’s key effects. Across all seven studies, the combined result showed no detectable effect — statistically indistinguishable from zero, not merely a weaker version of the original finding.(Galak, LeBoeuf, Nelson & Simmons, 2012) Bem’s original result did not hold up under the scientific process’s central tool for checking extraordinary claims: independent replication.
What this does, and doesn’t, establish
This does not prove precognition is impossible, and this page does not claim that. What it shows is that the single most rigorous, best-published attempt to demonstrate a paranormal effect under controlled conditions failed the direct replication test that the scientific process itself calls for — which is a meaningfully stronger and more specific statement than “no evidence has ever been offered.” Anyone encountering a paranormal explanation for a coincidence is entitled to ask the same question this case answers: has the claim been tested under conditions designed to rule out chance and bias, and if so, did the result hold up when independently repeated?
Common misconception
“Scientists refuse to take psychic claims seriously” does not match this case. A mainstream journal published Bem’s experiments after peer review, and other researchers invested significant resources testing the claim directly rather than ignoring it. The claim was taken seriously, tested rigorously, and did not survive that testing — which is a different, more informative outcome than simple refusal to engage.
Limits and open questions
A failed replication of one specific set of experiments does not settle every claim ever made about paranormal phenomena, many of which have never received comparably rigorous testing in either direction. This page presents Bem’s case specifically because it is unusually well documented on both sides (a serious original claim and a serious, large-scale replication attempt) — it should not be read as a complete survey of parapsychological research, which remains an active if minority field of study with its own internal debates about method and interpretation.
Related
- Replication and publication bias covers why a single published result, however striking, needs independent replication before being treated as established.
- How science tests unusual claims covers the general evidentiary standard this case illustrates directly.
Key takeaways
- The most rigorously published, mainstream-peer-reviewed attempt to demonstrate a precognition effect (Bem, 2011) did not survive a large, pre-registered replication attempt.
- This case shows the scientific process engaging seriously with an extraordinary claim, not refusing to consider it — publication, then rigorous follow-up testing, then a null result.
- A failed replication of one specific claim does not settle every paranormal claim, but it is a fair, well-documented illustration of the evidentiary bar this site applies to all of them.