Skip to main content

Research Glossary

A comprehensive reference of key terms, protocols, and statistical concepts used throughout the Psi Research Library.

24 Terms Defined
B
1 term

Bayes Factor

#
Statistics
Read Definition
A ratio expressing how much more likely the observed data is under one hypothesis (e.g., psi exists) versus another (e.g., chance). BF > 3 is considered moderate evidence; BF > 10 is strong evidence.
C
2 terms

Cohen's d

#
Statistics
Read Definition
A standardized effect size measuring the difference between two group means in units of standard deviation. Values of 0.2, 0.5, and 0.8 are conventionally considered small, medium, and large effects. In psi research, effect sizes are typically small (d = 0.1–0.3), which requires large sample sizes to detect reliably.

Correlation Coefficient (r)

#
Statistics
Read Definition
A measure of the strength and direction of a linear relationship between two variables, ranging from βˆ’1 (perfect negative correlation) to +1 (perfect positive correlation), with 0 meaning no relationship. In psi research, r is commonly used as an effect size metric β€” for example, ganzfeld hit rates are often converted to r values for meta-analysis. By convention, r = 0.1 is small, r = 0.3 is medium, and r = 0.5 is large. Most psi effects fall in the small range (r < 0.1).
D
2 terms

DMILS

#
Protocols
Read Definition
Direct Mental Interaction with Living Systems. Experiments testing whether one person's intention can influence another person's physiology (e.g., skin conductance) at a distance without sensory cues.

Double-Blind

#
Methodology
Read Definition
An experimental design where neither the participant nor the experimenter knows the condition assignment during data collection. Prevents unconscious bias (experimenter effect) from influencing results.
E
1 term

Effect Size

#
Statistics
Read Definition
A quantitative measure of the magnitude of an experimental effect. Common metrics include Cohen's d, Hedges' g, and correlation coefficient r. In psi research, effect sizes are typically small (r < 0.1) but statistically significant across large datasets.
F
1 term

File Drawer Problem

#
Methodology
Read Definition
The bias introduced when studies with null results go unpublished ('filed away'), making the published literature appear more supportive of an effect than the full evidence warrants. Meta-analyses attempt to correct for this.
G
1 term

Ganzfeld

#
Protocols
Read Definition
A sensory deprivation protocol used in telepathy experiments. The receiver relaxes in a mild sensory isolation environment (often using halved ping-pong balls over eyes and white noise) while a sender attempts to transmit an image or video clip.
H
1 term

Hedges' g

#
Statistics
Read Definition
A standardized effect size similar to Cohen's d but with a correction for small-sample bias. Hedges' g multiplies Cohen's d by a factor (approximately 1 βˆ’ 3/(4nβˆ’9)) that shrinks the estimate slightly when sample sizes are small. In psi meta-analyses, Hedges' g is often preferred over Cohen's d because many studies have modest sample sizes where the bias correction matters.
M
1 term

Meta-analysis

#
Statistics
Read Definition
A statistical technique that combines results from multiple independent studies to estimate an overall effect size. Addresses the 'file drawer problem' by aggregating evidence across experiments to find patterns invisible in single studies.
N
2 terms

NDE (Near-Death Experience)

#
Phenomena
Read Definition
Profound experiences reported by people who have been close to death, often including out-of-body perceptions, tunnel vision, life reviews, and encounters with deceased persons or beings of light.

Null Hypothesis

#
Statistics
Read Definition
The default assumption that there is no effect or no difference β€” in psi research, that any deviation from chance is due to random variation rather than an anomalous process. Statistical tests calculate the probability (the p-value) of observing the data if the null hypothesis were true. A low p-value suggests the null is unlikely.
O
1 term

OBE (Out-of-Body Experience)

#
Phenomena
Read Definition
The sensation of perceiving the world from a vantage point outside one's physical body. Sometimes reported during NDEs, sleep paralysis, or induced through specific meditation practices and sensory deprivation.
P
5 terms

p-value

#
Statistics
Read Definition
The probability of obtaining test results at least as extreme as the results actually observed, under the assumption that the null hypothesis is correct. In psi, p < 0.05 is the standard threshold for significance.

PPV (Positive Predictive Value)

#
Statistics
Read Definition
The probability that a statistically significant finding reflects a true effect rather than a false positive. PPV depends on statistical power, the prior probability of the hypothesis, and bias. Ioannidis (2005) showed that when effects are small, samples are small, and researcher degrees of freedom are high, PPV can fall below 50% β€” meaning most 'significant' findings are more likely false than true.

Pre-registration

#
Methodology
Read Definition
The practice of publicly registering hypotheses, methods, and analysis plans before data collection begins. Prevents post-hoc hypothesis revision (HARKing) and p-hacking, increasing the credibility of positive findings.

Presentiment

#
Phenomena
Read Definition
The hypothesized ability of the body to anticipate future emotional events before they occur, measured via physiological responses like skin conductance, heart rate, or pupil dilation seconds before a stimulus is shown.

Psi

#
General
Read Definition
An umbrella term for anomalous mental phenomena including telepathy, clairvoyance, precognition, and psychokinesis. From the Greek letter ψ (psi), chosen by J.B. Rhine to represent the unknown factor.
R
3 terms

Remote Viewing

#
Protocols
Read Definition
A protocol where a 'viewer' attempts to describe a distant or hidden target location using only mental impressions. Originally studied under the US government's STAR GATE program using coordinate-based targeting.

Replication

#
Methodology
Read Definition
An attempt to reproduce the findings of an earlier study using the same or similar methods. Successful replications strengthen confidence; failures weaken it. The 'replication crisis' affects all sciences, including psi.

RNG (Random Number Generator)

#
Equipment
Read Definition
A hardware or software device that produces random sequences, used in psychokinesis (PK) experiments. Participants attempt to mentally influence the output distribution toward a target value (e.g., more 1s than 0s).
S
2 terms

Sample Size

#
Statistics
Read Definition
The number of participants or trials (n) in a study. Small sample sizes reduce statistical power β€” the ability to detect a real effect β€” and increase the risk that a significant finding is a false positive. For small psi effects (d β‰ˆ 0.2), adequate power (β‰₯ 0.80) typically requires n > 400 per group.

Standard Deviation

#
Statistics
Read Definition
A measure of how spread out values are from the mean in a dataset, often represented by the Greek letter sigma (Οƒ). A small standard deviation means data points cluster tightly around the average; a large one means they are more dispersed. When a result is described as '6-sigma' (6Οƒ), it means the observed value is 6 standard deviations from the expected mean β€” an extremely unlikely outcome under the null hypothesis. In psi research, standard deviation is used to standardize effect sizes β€” for example, Cohen's d expresses group differences in standard deviation units, allowing comparison across studies that use different measurement scales.
Z
1 term

z-score

#
Statistics
Read Definition
A z-score (or z-value) measures how many standard deviations a result is from the expected mean under the null hypothesis. Higher absolute z-scores indicate stronger evidence against chance. For example, z = 2.0 corresponds to roughly p = 0.05 (two-tailed), while z = 8.31 indicates an astronomically unlikely result under the null. In psi meta-analyses, cumulative z-scores are used to summarize the overall strength of evidence across many studies.

Missing a key concept?

Suggest a term for the glossary