Can the Future Change Your Present Reactions
Can the future influence your reactions right now?
Attempt to replicate 'future influence' on reaction times failed to reproduce original results.
In 2010, psychologist Daryl Bem sparked controversy with studies suggesting people could react faster to words they hadn't seen yet. Four years later, French researcher Thomas Rabeyron tried to strengthen this 'retro-priming' effect by bringing back the strongest performers from his earlier experiment for a second round of testing.
Key Findings
- The retro-priming effect disappeared completely.
- Instead of reacting faster to 'future' primes, participants actually responded slightly slower than chance would predict.
- Only 10 of the 28 'high-scorers' showed positive results across both studies.
- However, the standard priming task worked normally—proving the equipment and methods were functioning.
- The researchers found no psychological traits that distinguished the few successful participants from the rest.
What Is This About?
The researchers invited 28 people back to the lab who had shown the strongest 'retro-priming' effects in a previous study with 162 participants. In the computer task, participants saw words flash on screen and had to categorize them as positive or negative as quickly as possible. The twist: sometimes the words were 'primed' by other words that appeared AFTER them in the sequence—an impossible causal influence if time only flows forward. The researchers also ran a standard priming test where earlier words helped identify later ones, and asked participants about their mental health, trauma history, and belief in paranormal experiences.
Test-retest design selecting high-scorers from a previous retro-priming study to attempt replication, comparing retro-priming with standard forward priming
Negative results for retro-priming (effect size = -0.25, non-significant), significant results for standard priming (effect size = 0.63, p < 0.1), with only 10 of 28 participants showing consistent positive results
How Good Is the Evidence?
Only 36% of previously high-scoring participants (10 out of 28) showed positive results again—roughly equivalent to flipping a coin twice and getting heads both times, suggesting the original 'high scores' may have been statistical flukes rather than genuine precognition.
Proponents argue that selecting high-scorers is a legitimate way to amplify weak psi signals that might be drowned out by noise in general population studies, and that 10 consistent performers out of 28 warrants further investigation. Skeptics counter that this is cherry-picking that ignores regression to the mean, and that the overall negative effect size (-0.25) suggests the original findings were likely false positives or statistical noise.
Mainstream: The null result confirms that retro-causal effects are not real and Bem's original findings were statistical artifacts. Moderate: Psi effects may exist but are too unreliable or context-dependent to replicate consistently, requiring new methodological approaches. Frontier: The negative result reflects 'psi-mediated experimenter effects' where the researcher's expectations suppress anomalous cognition, or that high-scorer selection requires different statistical thresholds.
Many assume that if a psi study fails to replicate, it proves the phenomenon doesn't exist. Actually, this study illustrates 'regression to the mean'—a statistical tendency where extreme initial results naturally become more average upon retesting, which is why independent replications with new participants are more informative than retesting the same high-scorers.
To establish retro-priming as a real phenomenon, researchers would need multiple independent laboratories to replicate positive results using pre-registered protocols (publicly registered analysis plans), with effect sizes consistently exceeding statistical noise, and successful prediction of which participants will score high before testing begins. This study attempted replication but produced negative results, failing to meet these criteria.
The results, for the whole group, on the retro-priming task, were negative and non-significant (es = -0.25, ns)
Stance: Skeptical
What Does It Mean?
Imagine taking a practice test, then studying the material, then taking the real test—and somehow doing better on the real test because of the practice you did after it. This study tested whether such 'reverse studying' effects hold up under careful scrutiny.
This study demonstrates 'regression to the mean'—the statistical phenomenon where extreme measurements naturally become more average upon retesting, which is why selecting only high-scorers from an initial study often leads to disappointment in follow-ups.
Understanding Terms
What This Study Claims
Findings
The retro-priming effect was negative and non-significant for the entire sample (effect size = -0.25), failing to replicate the original finding
weakOnly 10 of the 28 high-scoring participants showed positive retro-priming results across both the initial and follow-up studies
weakStandard (forward) priming produced significant results with an effect size of 0.63 (p < 0.1), indicating the experimental apparatus and methods were functioning normally
moderateNo specific psychological variables (anomalous experiences, mental health, mental boundaries, trauma, negative life events) distinguished the high-scoring participants
inconclusiveMethodology
The study employed a test-retest design selecting participants with the strongest retro-priming effects from a previous sample of 162 to test for replication
strongThis 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.