Discovery Lab Answers

The case for remote viewing

Coverage note: The studies below come from the ESP-Nexus library; the library’s share of the full published literature on remote viewing has not been measured. Treat what follows as a summary of what the library holds, not a settled account of the field as a whole.

Remote viewing (RV) is a structured experimental protocol in which a person attempts to describe a distant or shielded target — a location, object, or event — using no known sensory channel. The case for it rests on several distinct pillars: experimental effect-size data, meta-analytic synthesis, replication across independent laboratories, a documented government-program record, and a body of applied work. The evidence is genuinely mixed, and important null results exist alongside the positive ones.

The experimental and meta-analytic record

The most systematic quantitative case comes from Tressoldi and Katz (2023), a systematic review and meta-analysis covering RV studies from 1974 through 2022. Their frequentist random-effects analysis across 40 studies produced a standardized effect size with a confidence interval well above zero and an associated probability far below conventional significance thresholds. A positive pooled effect was also reported across 13 PEAR laboratory experiments covering a very large number of trials (Dunne, 2005), and in a smaller combined ganzfeld-condition analysis across 3 experiments (Roe, 2020).

Several individual series also reported positive outcomes. Mueller (2021) reported a hit rate across 100 trials. Mueller (2017) found a direct-hit rate across 36 sessions above chance. Targ (2019) reported results from a formal nine-trial series with Pat Price. Dunne (2003) reported a positive correlation across 336 formal binary trials. Dunne (1979) reported a significant sum-of-ranks result across eight trials.

The table below organizes the structured evidence by study:

Important: these rows use different, non-comparable metrics (hit rates, standardized effect sizes, z-scores, p-values, prevalence proportions). They cannot be averaged into a single “overall” figure.

Unsettled signals — what the evidence does not settle

The library’s evidence base contains outcomes pointing in different directions, and intellectual honesty requires surfacing them:

• Null and below-chance results are present. Wahbeh (2021) found no overall effect across 534 participants on a Quick Remote Viewing task. Wahbeh (2022) found a below-chance (psi-missing) result with Hedges’ g = −0.16. Wiseman (2010) found null results across four blind-judging trials. These are not footnotes — they are substantive findings from controlled designs.
• The Tressoldi & Katz (2023) meta-analysis itself reported high between-study heterogeneity (I² ≥ 50%) in its pooled results, meaning the studies are not measuring the same underlying quantity uniformly, and the summary figure should be read with that in mind.
• Escola-Gascon (2023) shows a large belief-moderator effect — the effect size for believers (d = 0.853) is roughly double that for the total sample (d = 0.457), which raises the question of whether attitude or expectation interacts with performance in a way that complicates simple effect-size summaries.
• Small-N series (Targ 1975, Mueller 2021, Dunne 1979, Piao 2023) contribute significant p-values but from very few participants or trials, limiting what can be generalized from any single series.

The government program and applied record

A documented government dimension exists. Sources and describe a multi-decade program — associated with SRI International and later government intelligence agencies — in which remote viewing was used operationally. A Marine Corps War College analysis examined the strategic implications of the program’s performance record and considered what cognitive and temporal effects it might imply. This history is relevant context for understanding why the protocol received sustained funding and how methodology was formalized, though operational claims from classified programs are not subject to the same verification as peer-reviewed experimental data.

Expert practitioner knowledge

A 2026 survey of eleven recognized RV experts — including Katz, Tressoldi, and Akin — gathered consensus recommendations on optimal protocols for real-time RV and precognition tasks. The experts emphasized protocol fidelity, appropriate target design, and pacing of sessions as variables that affect result quality. This is not itself evidence of an effect but informs how the experimental case should be evaluated: the claim is that results are protocol-sensitive, which has implications for why some studies find effects and others do not.

Skeptical critiques

What critics argue. A central line of critique — associated with Marks and Kammann’s work on the foundational Targ-Puthoff studies — holds that early positive results were contaminated by sensory cues embedded in transcripts: orderings, verbal habits, and incidental references that allowed judges to match sessions to targets without genuine anomalous information transfer. On this account, proximity-based study designs (where the outbounder visits targets in a sequence) are especially vulnerable.

What the experimental data show. The transcontinental remote viewing work conducted by Marilyn Schlitz used a design that partially addresses the proximity-cueing problem by separating viewer from target geographically [see Schlitz’s page]. The Tressoldi & Katz (2023) meta-analysis aggregated across studies spanning nearly five decades and multiple independent laboratories, which is relevant because cue-contamination critiques apply most forcefully to specific early designs rather than to the entire corpus uniformly. However, the meta-analysis’s high heterogeneity means it cannot straightforwardly override design-specific critique — what holds for later, better-controlled studies may not apply to the earlier ones that anchor the historical record. The null results from Wahbeh (2021, 2022) and Wiseman (2010) come from designs that post-date the cueing-critique era and still found no effect.

Analysis. The cueing critique has driven methodological improvements — blinded judging, remote targets, multiple-blind protocols — that are now standard. Independent replication outside the original SRI group has occurred (Schlitz, Roe, Wahbeh), with both positive and null outcomes. A program-level operational evaluation (the AIR/Stargate assessment) concluded the program had not demonstrated sufficient operational utility to justify continuation — a finding that is logically separable from the question of whether controlled laboratory experiments show an anomalous signal, but that constrains the broader public-interest case.

For a deeper quantitative treatment, the Remote Viewing Meta-Analysis and Systematic Review page covers the Tressoldi and Katz (2023) work in detail, including the skeptical critique baseline. The Remote Viewing phenomena page provides broader context on protocols and history. Brenda J. Dunne’s experimental and analytical methods are covered at the PEAR remote viewing page.