Spooky actions cannot be tricked: Exploring the Nature of Micro-Psychokinesis with Higher-Level Analytical Strategies

Objective. Testing specific mind-related effects on reality formation by investigating deviations from quantum randomness during the process of intentional observation, also known as micro-psychokinesis (micro-PK), is one empirical approach to address the mind-matter problem. While meta-analytical evidence over decades of research suggests that observers’ conscious or unconscious motivational states could produce significant micro-PK effects on quantum number generator (QRNG) outcomes, direct replication attempts often failed or the initially strong and apparently stable effects declined over time. Therefore, in addition to a direct replication strategy, alternative confirmatory approaches have been proposed that apply Higher-Level Analytical Strategies (HLASs) to test these effects existence. The present study reports a preregistered replication attempt of a micro-PK effect on QRNG outcomes documented by a specific HLAS variant, the Change of Evidence (CoE) measures, which were performed on a post-hoc basis in an original study.Method. To assess the observers’ unconscious intentional states, in a re-analysis of an existing data set we determined participants’ personality traits (PTs) on Cluster C and analyzed the relation between Cluster C outcome and trait-related sentences selected via QRNG outputs mirroring the respective typical concerns associated with the PTs vs. neutral stimuli. We predicted the presentation of more PT-related stimuli than chance expectation among participants exhibiting high PT assessment scores (PT-high) but not among individuals who exhibited low levels of these characteristics (PT-low). Although in line with this prediction initially strong Bayesian evidence for H1 was found within the PT-high group, this effect unexpectedly declined upon further data collection to indecisive evidence. Three post-hoc CoE tests were then performed comparing the empirical data to a randomly generated set of 10,000 simulations to distinguish a false positive finding i.e., an unsystematic, random Bayesian evidence variation, from a true positive effect that later declined i.e., from a systematic one. These exploratory post-hoc re-analyses of an existing data set rejected the null hypotheses in all CoE tests performed. These findings were then tested for replicability in a preregistered replication study which will be reported here.Results. The replication attempt failed. No Bayesian evidence for micro-PK was found and none of the CoE analyses could reject the null hypothesis in the PT-high group. As expected, no effects were found in the PT-low group either.Conclusion. The findings of the herein presented analyses go along with the results of earlier studies reporting similar experimental approaches which initially revealed strong and supposedly robust evidence for the postulated effects in non-preregistered original experiments but failed in a direct replication. Most central, the present findings indicate, that even CoE measures following the idea of HLASs to circumvent the non-replicability problem cannot provide replicable higher order micro-PK effects. The results obtained are discussed with regard to the Model of Pragmatic Information (MPI).