Future Feelings: 90 Trials, Zero Proof
Can your body sense future events before they happen?
Imagine sitting at your computer, making choices about images you're about to see—except the computer hasn't randomly selected those images yet. That's essentially what happened in labs across 14 countries, where researchers tested whether people could somehow 'feel' future events before they occurred. When psychologist Daryl Bem first published these controversial experiments in 2011, the scientific community demanded replications. Now, after analyzing 90 experiments from 33 independent laboratories, the data shows a persistent statistical signal that challenges our understanding of time and consciousness.
Large-scale analysis finds evidence that people unconsciously anticipate random future events.
In 2011, psychologist Daryl Bem published controversial experiments suggesting people could unconsciously respond to future events. The findings sparked intense debate and calls for replication. This meta-analysis examines 90 follow-up experiments from 33 laboratories across 14 countries to test whether the effect holds up.
A large-scale analysis of 90 experiments suggests humans may show measurable responses to future events, though the effect is small and the mechanisms remain completely unknown.
Key Findings
- The combined results showed a small but statistically robust effect suggesting people do unconsciously anticipate future random events.
- The effect remained significant even when the original controversial studies were excluded, indicating independent laboratories could replicate the findings.
- However, the effect size was small (0.09), meaning the phenomenon, while detectable, is subtle.
What Is This About?
Researchers collected data from 90 experiments testing 'presentiment' - the idea that people can unconsciously sense future events. In typical experiments, participants viewed randomly selected images while their physiological responses were measured. The key question: did their bodies react differently before seeing emotional versus neutral images, even though the computer hadn't yet randomly selected which type to show? The researchers combined all results using meta-analysis techniques to look for overall patterns.
Meta-analysis combining 90 experiments testing whether people can unconsciously respond to future random events before they occur.
Strong statistical evidence for presentiment effects across multiple independent laboratories, with effect remaining significant even when original studies excluded.
How Good Is the Evidence?
Effect size of 0.09 - considered small in psychology research, comparable to the effect of aspirin on heart attack prevention. The statistical significance (p = 1.2 × 10-10) is extremely strong, much stronger than the typical p < 0.05 threshold used in psychology.
Supporters argue the statistical evidence is overwhelming and the effect has been independently replicated across multiple laboratories and countries. Skeptics contend that despite statistical significance, the effect size is so small it could result from subtle methodological artifacts, publication bias, or statistical anomalies rather than genuine precognition. Both sides agree more research with improved controls is needed.
Mainstream: Statistical artifacts or methodological flaws likely explain the small effects, not genuine precognition. Moderate: The consistent replication across laboratories suggests something interesting is happening, but the mechanism remains unclear and may not involve precognition. Frontier: The evidence supports genuine unconscious anticipation of future events, challenging conventional understanding of time and causality.
This isn't about conscious psychic predictions or fortune telling. The research focuses on unconscious physiological responses that occur before random events, measured through subtle changes in heart rate, skin conductance, or pupil dilation that participants aren't aware of.
To settle this question would require large-scale, pre-registered studies with rigorous controls for potential artifacts, real-time data monitoring, and independent verification of results. This meta-analysis meets the replication criterion by combining multiple independent studies, but individual studies would benefit from pre-registration and stronger methodological controls.
We here report a meta-analysis of 90 experiments from 33 laboratories in 14 countries which yielded an overall effect greater than 6 sigma, z = 6.40, p = 1.2 × 10-10 with an effect size (Hedges' g) of 0.09.
Stance: Supportive
What Does It Mean?
The idea that your brain might be subtly responding to events that haven't happened yet challenges everything we think we know about cause and effect. With 90 experiments showing consistent results across different cultures and laboratories, this represents one of the most systematic challenges to conventional time perception ever documented.
Like when you get a 'gut feeling' something is about to happen, or when you inexplicably feel uneasy moments before receiving bad news - this research tests whether such intuitive anticipation of future events occurs more often than chance would predict.
If these presentiment effects prove genuine, they would suggest that consciousness operates in ways that transcend our current understanding of linear time. This could revolutionize fields from neuroscience to physics, potentially indicating that the brain processes information through mechanisms we haven't yet discovered. Such findings might also validate certain meditation and mindfulness practices that claim to enhance intuitive awareness.
Meta-analyses gain power by combining multiple studies, but they're only as good as the individual studies they include - statistical significance across many small studies doesn't automatically prove the effect is real.
Understanding Terms
What This Study Claims
Findings
Meta-analysis of 90 experiments yielded an overall effect greater than 6 sigma (z = 6.40, p = 1.2 × 10-10) with effect size of 0.09
strongIndependent replications excluding original author's studies still showed significant effects (z = 4.16, p = 1.1 × 10-5)
strongBayesian analysis yielded Bayes Factor of 1.4 × 109, greatly exceeding criterion for 'decisive evidence'
strongMethodology
544 unretrieved experiments would be needed to reduce the overall effect to trivial levels
moderateSeven of eight statistical tests support conclusion that database is not significantly compromised by selection bias or p-hacking
moderateThis summary is for general information about current research. It does not constitute medical advice. The scientific interpretation of these results is debated among researchers. If personally affected, please consult qualified professionals.