A meta-analysis of 90 experiments from 33 laboratories in 14 countries found evidence for a small but statistically significant precognition effect—the ability to be influenced by a randomly selected stimulus before it occurs. The overall effect size (Hedges' g) was 0.09, with a combined z-score of 6.40 and a Bayes Factor of 5.1 × 10⁹, exceeding the threshold for decisive evidence. When the original author's experiments were excluded, the effect size for independent replications was 0.06, with z = 4.16. Additional analyses indicated the database was not significantly compromised by selection bias or p-hacking. The authors discuss the controversial status of precognition and related psi phenomena.
Two international experiments attempted to replicate a controversial precognition study by Daryl Bem. The experiments reversed the usual cause-effect sequence: participants responded to stimuli before a prime appeared, testing whether response times would be longer for incongruent stimuli. The preregistered confirmatory hypothesis was not supported in either experiment. Exploratory analyses in Experiment 1 found a significant effect only among English-language participants, not those using translations.
A meta-analysis of 90 experiments from 33 laboratories in 14 countries found evidence for precognition—the ability to be influenced by a future random event before it occurs. The overall effect was highly significant (z = 6.40, p = 1.2 × 10⁻¹⁰) with a small effect size (Hedges’ g = 0.09). When the original author's experiments were excluded, the effect remained significant (g = 0.06, z = 4.16, p = 1.1 × 10⁻⁵). Bayesian analysis provided decisive evidence for the effect. Statistical tests indicated the database was not significantly affected by selection bias or p-hacking. The authors discuss the controversial status of precognition and related psi phenomena.