Discovery Lab Answers

Framing the answer

First, a scope caveat that governs everything below: the sources retrieved for this question are a small, uneven slice — a large replication attempt of one precognition paradigm, a smartphone-based exploratory psi study, and several reflective/theoretical essays. They are not a measured sample of the century-long literature, and the library’s share of the published psi corpus on this “big picture” question has not been measured. So what follows describes what these particular sources show and how ESP-Nexus frames the terrain — not a settled verdict on the field. Treat any generalization as scoped to these holdings.

ESP-Nexus itself takes no position on whether psi is real; its commitment is to reporting research as its authors wrote it and setting critique beside rebuttal.

What the retrieved sources actually show

The pattern across these sources is not a single clean signal — it is exactly the tension a century of work has produced:

• A large preregistered replication failed to reproduce a headline precognition effect. Muhmenthaler, Dubravac, and Meier (2022) ran a large-scale replication of Bem’s (2011) retroactive-priming work and found no evidence for precognition, with only scattered incidental correlations that did not survive alpha-level correction — the authors explicitly frame those as things that would need theory-driven, preregistered, confirmatory follow-up rather than as findings ^[1].
• Contemporary proponents themselves now argue for small, hard-to-pin effects. Mossbridge and Radin’s (2021) smartphone-test work proposes a “SEARCH” strategy whose very first premise is “Small effects” — an explicit acknowledgement from within the research community that forced-choice psi tasks do not produce large or easily commanded results, and that much of the work remains early and exploratory ^[2].
• A reflective assessment of the field’s own reference literature is candid about limited progress. Solfvin (2016), reviewing Parapsychology: A Handbook for the 21st Century, quotes Broughton’s own chapter conclusion that three and a half decades of neurobiology-and-psi research “have not advanced the field any further,” and characterizes the field structurally as having lost the coherent center it once had ^[3].
• The dispute is partly a physics-and-philosophy dispute, not only a data dispute. Williams (2023) engages Sean Carroll’s argument — from The Big Picture — that our success in quantum field theory sets a bar that should discourage taking psi seriously, and tries to build a consciousness-and-causality framework in response ^[4]. This is the “would it even fit our physics?” front rather than the “did the experiment work?” front.

So the honest big-picture summary from these sources is: claimed effects are small, a major replication of a flagship paradigm came back null, proponents increasingly frame the work as early/exploratory with small effect sizes, and a substantial part of the disagreement is theoretical (does psi cohere with known physics?) rather than purely empirical. The ESP-Nexus Precognition page makes the same structural point in plain language: any effect here is near-zero and only becomes detectable when many studies are pooled via meta-analysis — which is also why it is easy to drown in noise or produce with minor methodological flaws.

Skeptical critiques and discussion

What critics argue. Carroll (2016, 2008), as presented by Williams (2023), argues from physics: our theory-building successes (quantum field theory) set a high bar, and psi phenomena as described have no plausible mechanism within that framework, so the prior against them is very strong ^[4]. Separately, the replication critique is embodied directly in the data: Muhmenthaler, Dubravac, and Meier (2022) treated Bem’s precognition claim as a replication target and reported no evidence for it in a large sample ^[1].

What the experimental data show. The data these sources carry are mixed and mostly small or null. The 2022 replication found no confirmatory precognition effect and only fragile incidental correlations that vanished under correction ^[1]. Mossbridge and Radin (2021) report significant regression models within an explicitly exploratory smartphone dataset while conceding effects are small and the work is early-stage ^[2]. Solfvin (2016) relays Broughton’s own judgment that biological-mechanism research has not advanced ^[3].

Analysis. Where one research group (Bem) reported a positive precognition effect, an independent group (Muhmenthaler et al., using a large preregistered design) reported a null — that specific divergence is the concrete state of that paradigm ^[1]. The physics-based objection (Carroll) and the response to it (Williams) sit on a separate, theoretical axis and are not resolved by any experiment in these sources ^[4]. Independent, preregistered, confirmatory replication of the exploratory smartphone findings has not been published in these sources ^[2].

Where to go deeper

For the effect-size-and-meta-analysis framing of one core domain, see Precognition; for how the “reduce sensory noise to detect a weak signal” logic and meta-analytic pooling are treated, see Anomalous Cognition; and for a researcher whose own work centers on when psi effects appear and who has written on replicability across experimenters, see Mario P. Varvoglis.