Overview

The phrase “anomalous cognition” was coined at Stanford Research Institute (SRI) in the early 1970s. Hal Puthoff and Russell Targ were running experiments in which volunteers tried to describe distant locations they had never visited. The results looked promising enough that the CIA began funding the work. Calling it “ESP” or “psychic perception” would have invited ridicule in government circles. So the researchers settled on a clinical alternative. They defined anomalous cognition as information transfer in which all known sensory channels are absent.⁴⁵ The term stuck. It now appears throughout the academic literature as a near-synonym for ESP, remote viewing, and related phenomena.

The central question is simple: can a person accurately describe a target, such as a distant location, a hidden object, or a future event, without receiving any normal sensory information about it? Proponents say yes, backed by decades of data. Skeptics say the experiments have not been clean enough to rule out ordinary explanations.

How the Experiments Work

Most AC experiments use a free-response design. Instead of guessing a playing card from a fixed set, the participant describes whatever impressions come to mind about an unknown target. The target might be a photograph, a geographic location, or a short video clip. This open-ended approach tends to produce richer data and larger effects than card-guessing tasks.¹²

To make sure results can be evaluated fairly, researchers use blinding. The person running the session does not know which target was selected. The judge does not know either. This removes the chance that subtle cues, a raised eyebrow or a hesitation in speech, could leak information to the participant. When neither the experimenter nor the judge knows the answer during the session, the design is called double-blind.

After the session, an independent judge compares the participant’s description against a small set of possible targets, usually four or five. The judge ranks them for similarity to the description. The actual target is then revealed. If the description matches the real target better than the decoys, that counts as a hit. Repeating this across many trials lets researchers calculate whether the hit rate is higher than random guessing would produce.

A related design is the ganzfeld procedure. “Ganzfeld” is German for “whole field.” The participant lies in a reclining chair, wears translucent goggles bathed in red light, and listens to white noise through headphones. This uniform sensory environment quiets the ordinary chatter of perception. The idea is that psi signals, if they exist, are weak. Reducing background noise should make them easier to detect, just as you can hear a whisper better in a quiet room than at a party. A sender in a separate room concentrates on a randomly chosen target while the participant speaks their impressions aloud.¹⁴

What the Evidence Shows

Any single experiment gives an ambiguous result. Small samples produce noisy data, and a lucky run of hits could happen by chance. To get a clearer picture, researchers combine many studies using meta-analysis. The logic is straightforward: if a real effect exists, it should show up consistently across many independent experiments. When you pool that much data, even a small genuine effect becomes statistically detectable.

A 2021 umbrella review pulled together 11 separate meta-analyses covering roughly 928 studies from more than 80 years of research. It found that free-response AC protocols consistently produced effects above chance. Remote viewing showed the largest average effect sizes. Dream-based ESP studies came second. The ganzfeld design, with the largest evidence base, produced a smaller but still significant effect. Altered states of consciousness combined with free-response methods reliably produced the strongest results.⁷

Remote viewing studies from 1974 to 2022 show an average hit rate 19.3 percent above chance (95% CI: 13.6% to 25%), substantially higher than ganzfeld protocols, with no signs of publication bias after correction.¹¹

The ganzfeld literature is the most thoroughly scrutinized. A meta-analysis of 59 free-response studies from 1992 to 2008 found that ganzfeld studies produced a hit rate of about 32 percent where 25 percent was expected by chance. That gap is small in absolute terms, but it was highly consistent across studies and labs. Standard free-response studies without the sensory-reduction component showed no significant effect in the same analysis. That supports the idea that reducing sensory noise genuinely helps.¹⁴

Factors That Seem to Modulate Performance

One striking finding in the AC literature is that performance is not constant. Several factors seem to push it up or down. Altered states of consciousness are the most consistently supported moderator. Participants who are relaxed, meditating, or dreaming tend to score better than those in ordinary waking states. This fits the noise-reduction model: if psi is a weak signal, anything that quiets mental chatter should help.⁷

Two more unusual moderators have attracted attention: local geomagnetic conditions and local sidereal time. Local sidereal time is not the same as clock time. It measures the Earth’s orientation relative to the fixed stars rather than the Sun. Researchers noticed that AC effect sizes in a large database of trials were dramatically higher when sessions occurred near a specific sidereal time, around 13.5 hours. The effect appeared in an original dataset and then replicated in an independent set of studies.² A follow-up study suggested that specific geomagnetic pulsations in a particular frequency band may be the underlying driver of both associations.⁵ These findings are intriguing but have not yet been widely replicated by independent groups.

Participant selection also matters. Across multiple meta-analyses, people who believe they have psychic ability, who meditate regularly, or who are pre-screened for high performance tend to score better than unselected volunteers. This effect is especially pronounced in ganzfeld studies.¹⁴ A survey of experienced remote viewers found that most report subjective markers, such as vivid or surprising imagery, that they associate with accurate target hits. These markers were developed through practice rather than formal instruction.¹³

Where Things Stand Today

After more than 50 years of lab research, anomalous cognition sits in an uncomfortable middle ground. The statistical evidence, particularly from remote viewing and ganzfeld studies, is more substantial than most scientists outside the field realize. Multiple independent meta-analyses, using different datasets and different statistical methods, converge on small but consistent effects. Those effects are extremely unlikely to be chance alone.¹² At the same time, the effects are small, the mechanisms are unknown, and replication is uneven enough that mainstream science has not accepted the phenomenon.

Current research is moving in two directions. One is better methods: tighter blinding, automated judging using natural language processing tools, brain imaging to identify states linked to high performance, and pre-registered designs that prevent selective reporting.²⁷ The other is theory: researchers are asking whether AC is best understood as information transfer, as a form of unconscious precognition, or as something that requires a fundamentally different model of how minds relate to time and space.⁹⁵²

A 2018 analysis put it well: the evidence is sufficient to justify investigating anomalous cognition seriously, but not yet sufficient to endorse it as an established fact. Those two thresholds are different. Conflating them has produced much of the heat in this debate.⁴⁸