Precognition
Knowledge of future events before they occur — from Daryl Bem's controversial 'feeling the future' experiments to presentiment physiology and the physics of retrocausality.
About Precognition
Precognition — the acquisition of information about future events that cannot be inferred from presently available data — challenges the most fundamental assumption of our ordinary experience of time: that causation flows in one direction, from past to future, and that the future is undetermined until it occurs. If precognition is real, then either the future is in some sense already determined (and consciousness can access it), or consciousness can receive information that propagates backward through time (retrocausality), or our understanding of time, causation, and the relationship between consciousness and information requires fundamental revision. Any of these conclusions would be transformative for physics, neuroscience, and philosophy of mind.
Reports of precognition — through dreams, visions, intuitive feelings, or deliberate divination — are found in virtually every culture throughout recorded history. The Oracle at Delphi, the prophetic books of the Hebrew Bible, the dream incubation temples of ancient Greece and Egypt, the I Ching of Chinese tradition, the nadi shastra palm-leaf manuscripts of India, and the dream interpretation practices of indigenous peoples worldwide all testify to a persistent human conviction that the future can be known before it arrives. The question for science is whether this conviction reflects a genuine capacity of consciousness or a universal cognitive illusion generated by pattern-seeking, selective memory, post-hoc interpretation, and the statistical certainty that, among billions of dreams and intuitions, some will coincidentally match future events.
The laboratory investigation of precognition has produced a body of evidence that is, depending on one's assessment, either the strongest statistical case for any paranormal phenomenon or a cautionary tale about the limits of statistical methodology when applied to extraordinary claims. The modern experimental study of precognition can be traced to J.B. Rhine's Duke University laboratory in the 1930s, where subjects attempted to predict the order of Zener card sequences that had not yet been determined. Rhine reported hit rates significantly above the 20% expected by chance, though his early protocols were criticized for potential methodological flaws. The field advanced significantly with the development of more rigorous experimental designs, culminating in two major research programs that have defined the modern debate: Daryl Bem's 'feeling the future' experiments and the presentiment (physiological precognition) research program.
Daryl Bem — a social psychologist at Cornell University with a distinguished career in mainstream psychology — published 'Feeling the Future: Experimental Evidence for Anomalous Retroactive Influences on Cognition and Affect' in the Journal of Personality and Social Psychology (JAPS) in 2011. The paper reported nine experiments involving over 1,000 participants, in which future events appeared to influence present behavior. The most provocative experiment was a time-reversed version of a standard priming study: instead of showing a priming stimulus before a target response (standard priming), Bem showed the priming stimulus after the response and found that it still affected the response — as if the future priming event was reaching backward in time to influence the present behavior. Eight of nine experiments showed effects in the predicted direction, with a combined p-value of 1.34 x 10^-11 (approximately 1 in 74 billion odds against chance). The paper was published in one of psychology's most prestigious journals after routine peer review, a decision that provoked an explosion of debate, methodological critique, and soul-searching within the field about the relationship between statistical significance and belief.
The presentiment research program represents a different and in some respects more compelling approach. Rather than asking participants to consciously predict future events, presentiment studies measure unconscious physiological responses — skin conductance (galvanic skin response), heart rate, pupil dilation, brain activity (EEG, fMRI) — to randomly selected future stimuli. The basic protocol: a participant sits in front of a computer screen. The computer randomly selects an image from a pool of calm and emotionally arousing images (the selection occurs after the physiological measurement). The participant's physiological response is recorded during the seconds before the image is displayed. If precognition exists at a physiological level, arousing images should produce elevated physiological activation in the seconds before they appear — before the participant or the computer 'knows' which image will be selected.
The results have been remarkably consistent. Dean Radin (Institute of Noetic Sciences) published the first major presentiment study in 1997, finding that skin conductance was significantly elevated 2-5 seconds before emotionally arousing images compared to calm images — before the image was randomly selected. Radin replicated this finding across multiple studies (1997, 2004), and independent replications have been reported by Bierman and Scholte (2002, University of Amsterdam), McCraty et al. (2004, HeartMath Institute, using cardiac measures), Tressoldi et al. (2009, University of Padova), and Mossbridge et al. (2012, Northwestern University). Mossbridge, Tressoldi, and Utts published a meta-analysis in Frontiers in Psychology (2012) analyzing 26 presentiment studies from seven independent laboratories, finding a small but highly significant effect (d = 0.21, p = 5.7 x 10^-8) that was not explained by methodological quality, selective reporting, or other artifacts. Julia Mossbridge (then at Northwestern, a mainstream research university) has been particularly important in bringing presentiment research toward mainstream scientific attention.
The physics of retrocausality — the possibility that causation can operate backward in time — provides a potential theoretical framework. In quantum mechanics, several interpretations allow for retrocausal effects. The Wheeler-Feynman absorber theory (1945) treats advanced waves (propagating backward in time) as physically real. The transactional interpretation of quantum mechanics (Cramer, 1986) proposes that quantum events involve both 'offer waves' traveling forward in time and 'confirmation waves' traveling backward. Costa de Beauregard (1977) proposed that quantum mechanical correlations might serve as the physical basis for precognition. Huw Price (Cambridge philosopher of physics) has argued that the assumption of a 'causal arrow of time' is not required by the fundamental laws of physics, which are time-symmetric — suggesting that retrocausation is not forbidden by physics, even if it is forbidden by common sense. These theoretical proposals remain speculative and are not accepted by most physicists, but they demonstrate that precognition is not ruled out by the fundamental laws of physics as currently understood.
Methodology
Time-reversed psychological paradigms (Bem). Bem's methodology involved taking well-established psychological protocols and reversing the temporal order of cause and effect. In a standard priming study, the prime (a word or image) is shown before the target to influence the response; Bem showed the prime after the response and measured whether it still influenced performance. In a standard habituation study, repeated exposure to a stimulus reduces the novelty response; Bem tested whether future exposure would reduce the present novelty response. This 'time-reversal' approach is methodologically elegant because it uses established paradigms with well-characterized effect sizes, making any anomalous result directly comparable to the known forward-in-time effect. The key controls: all stimulus selection was by hardware random number generator, all timing was computer-controlled, and all data recording was automatic.
Presentiment physiological paradigm. The standard presentiment protocol: (1) Baseline period — participant sits quietly, physiological measures (skin conductance, heart rate, EEG) are recorded for 3-5 seconds; (2) Pre-stimulus period — the seconds immediately before image display, during which the computer has not yet randomly selected the image; physiological data continues to be recorded; (3) Random selection — a hardware random number generator selects the image category (calm or emotionally arousing) and specific image; (4) Stimulus display — the image is shown for 3-5 seconds; (5) Recovery period — the screen goes blank; physiological measures return to baseline. The dependent variable is the difference in physiological response during the pre-stimulus period between trials that will show arousing versus calm images. The logic: if the body responds to the emotional content of the image before the image is selected, then information from the future is influencing the present.
Bayesian versus frequentist analysis. The Bem experiments catalyzed a methodological debate about how to evaluate evidence for extraordinary claims. Frequentist analysis (the traditional p-value approach) assesses how surprising the data would be if the null hypothesis (no precognition) were true — and Bem's data were very surprising (p approximately 10^-11). Bayesian analysis, advocated by Wagenmakers et al. (2011), incorporates a prior probability for the hypothesis: given that precognition violates known physics, the prior probability is extremely low, and even very surprising data may not be sufficient to overcome that low prior. The debate is fundamentally about the appropriate role of prior belief in evaluating evidence — a question that extends far beyond precognition to the foundations of scientific reasoning.
Pre-registration and adversarial collaboration. In response to criticism, Bem and several skeptics have engaged in adversarial collaborations — pre-registered studies conducted jointly by proponents and critics, with agreed-upon protocols and statistical analyses. Bem et al. (2015) reported that pre-registered replications by proponents produced significant results while those by skeptics did not — a pattern consistent with either experimenter effects (the experimenter's belief influences the outcome) or with proponents being more skilled at implementing the specific experimental conditions that produce the effect. This pattern itself has become a subject of study.
Retrocausality in physics. The theoretical framework for precognition draws on interpretations of quantum mechanics that permit retrocausal influences. The Transactional Interpretation (Cramer, 1986) proposes that quantum events involve both 'offer waves' (propagating forward in time from the source) and 'confirmation waves' (propagating backward in time from the absorber), with the quantum event being a 'handshake' between past and future. The Two-State Vector Formalism (Aharonov, Bergmann, and Lebowitz, 1964) describes quantum systems in terms of both forward-evolving and backward-evolving state vectors, with complete physical description requiring information from both the past and the future. These frameworks are not fringe — they are published in mainstream physics journals and debated at physics conferences — though their application to macroscopic consciousness remains speculative.
Evidence
Bem's 'Feeling the Future' (2011). The nine experiments in Bem's JAPS paper tested time-reversed versions of established psychological phenomena. Experiment 1 (precognitive approach/avoidance): participants chose between two curtains on a screen; the computer selected the 'target' curtain after the choice, and participants chose the future target significantly above chance (53.1% vs 50%, p = 0.01) specifically for erotic images — suggesting emotional arousal enhances precognitive detection. Experiment 2 (precognitive priming): participants classified words as pleasant/unpleasant faster when the priming word (shown after the response) was congruent — a time-reversed priming effect. Experiments 3-4 (precognitive habituation): participants showed familiarity preferences for images they would be exposed to in the future. Experiments 5-7 (retroactive facilitation of recall): studying word lists after a recall test improved recall performance — as if future studying reached backward to enhance present memory. Experiments 8-9 (retroactive facilitation): replications with different stimuli. The combined evidence across all nine studies had a p-value of approximately 10^-11.
Replication attempts. Bem's publication triggered over 90 replication attempts worldwide. The results have been mixed and debated. Galak et al. (2012) reported a series of direct replications that failed to find the effect, publishing in the same journal (JAPS). However, Bem, Tressoldi, Rabeyron, and Duggan (2015) published a meta-analysis of 90 experiments from 33 laboratories, reporting a significant overall effect (p = 1.2 x 10^-10) with a small effect size (d = 0.09). The divergent conclusions partly reflect different analytical choices: Bayesian versus frequentist analysis, inclusion/exclusion criteria for studies, and the prior probability assigned to the phenomenon. The debate has not resolved, and the replication evidence is genuinely ambiguous — consistent with either a real but tiny effect or with persistent statistical artifacts.
Presentiment meta-analysis (Mossbridge et al., 2012). The most rigorous statistical assessment of precognition evidence is the meta-analysis by Julia Mossbridge, Patrizio Tressoldi, and Jessica Utts, published in Frontiers in Psychology. The analysis covered 26 studies from 7 independent laboratories, measuring physiological responses (skin conductance, heart rate, pupil dilation, brain activity) before randomly selected future stimuli. The combined effect size was d = 0.21 (small but consistent), with p = 5.7 x 10^-8. The analysis assessed and rejected multiple potential artifacts: expectation effects (the computer's random selection occurs after the measurement period), sensory cues (the participant has no access to the stimulus before it is displayed), selective reporting (the fail-safe N exceeded 87 studies), and methodological quality (effect size was not correlated with quality ratings). The authors explicitly stated that they could not identify a conventional explanation for the results.
Dean Radin's presentiment studies. Radin's series of presentiment experiments (1997, 2004, 2008) established the basic protocol and demonstrated the effect across multiple modalities. His 1997 study found that skin conductance was elevated approximately 3-5 seconds before the random selection and display of emotionally arousing images compared to calm images. His 2004 study replicated this with fMRI, finding anticipatory activation in brain regions associated with emotional processing. Radin has also conducted 'double-slit' experiments testing whether consciousness can influence quantum events, finding small but significant effects (published in Physics Essays, 2012) that bear on the retrocausality question.
Precognitive dreams. The historical and anecdotal evidence for precognitive dreams is extensive. Dunne (1927) published An Experiment with Time, documenting his systematic recording of dreams and their subsequent correspondence with future events. The Premonitions Bureau (documented by Sam Knight, 2022), established after the Aberfan disaster of 1966, collected and evaluated hundreds of reported precognitive dreams and visions. Rhine collected over 7,000 reported precognitive dream experiences in her spontaneous psi case collection. Controlled laboratory studies of precognitive dreaming (Krippner et al., 1972, at Maimonides Medical Center) produced significant results using protocols in which randomly selected target images were shown to the dreamer the morning after the dream, and blind judges matched dream reports to targets. However, independent replication of the Maimonides dream studies has been inconsistent.
Skeptical assessment. The skeptical critique of precognition evidence is substantive. Wagenmakers et al. (2011) published a Bayesian re-analysis of Bem's data in JAPS, concluding that the evidence was insufficient from a Bayesian perspective because the prior probability of precognition is extremely low. Alcock (2011) argued that methodological issues (subtle experimenter influence, post-hoc analytical flexibility) could account for the results. The failure of several high-profile direct replications (Galak et al., 2012; Ritchie et al., 2012) weakened confidence in the Bem findings specifically, though the presentiment meta-analysis presents a harder target for skeptical critique. The honest assessment: the presentiment evidence is the strongest, the Bem evidence is ambiguous, and the question remains genuinely open.
Practices
Precognitive dream journaling. The most accessible practice for exploring precognition is systematic dream recording with subsequent comparison to waking events. The protocol: keep a notebook by the bed; immediately upon waking, record all dream content in as much detail as possible; date each entry; periodically (weekly or monthly) review past entries and note any correspondences with subsequent events. The key discipline is recording dreams before knowing which events might correspond — eliminating post-hoc rationalization. J.W. Dunne's protocol (An Experiment with Time, 1927) added a systematic structure: recording dreams immediately, noting specific details and predicted timeframe, and blind-rating the correspondence between dreams and subsequent events. The challenge is distinguishing genuine precognition from coincidence, symbolic interpretation, and confirmation bias — which requires large sample sizes and ideally blind judging by someone who did not have the dream.
Associative Remote Viewing for prediction. ARV protocols (developed within the remote viewing community) apply precognitive ability to binary predictions. Two possible future outcomes (stock market up/down, team A wins/team B wins) are associated with two different photographs. A remote viewing session targets 'the photograph I will be shown as feedback' — the feedback photograph corresponding to the correct outcome. The viewer's description is matched to one of the two photographs, generating a prediction. Published ARV studies have shown hit rates significantly above the 50% chance expectation, and private investment groups have reported profitable results, though the evidence base is insufficient for strong conclusions.
Presentiment self-experimentation. The presentiment effect can be explored personally using software that presents random images (alternating calm and emotionally arousing) while the user monitors their own physiological and emotional state in the seconds before each image appears. Several open-source and commercial applications implement this protocol. The effect size in laboratory studies is small (d = 0.21), meaning it is unlikely to be subjectively detectable in individual trials — but systematic recording across many trials, combined with honest assessment of one's emotional/physical state in the pre-stimulus interval, can provide personal data.
Contemplative approaches to time perception. Multiple contemplative traditions describe practices that alter the experience of time in ways that may bear on precognition. Deep meditation frequently produces the experience of temporal expansion, simultaneity, or timelessness. The yogic concept of samyama on change (parinamah) — directing sustained contemplative attention to the nature of transformation itself — is described as the practice that develops knowledge of past and future. The Buddhist practice of vipassana (insight meditation), which involves observing the moment-to-moment arising and passing of sensory experiences, is understood to develop a refined perception of causal sequences that extends, in advanced practice, to the perception of causal sequences that have not yet manifested in ordinary time.
Traditional divination systems. Divination systems — the I Ching, Tarot, plant medicine divination, astrology, geomancy — represent humanity's oldest and most widespread technology for accessing future information. Whether these systems work through genuine precognition, through activation of unconscious pattern recognition (Jungian synchronicity), or through the imposition of narrative structure on random events is debated. The I Ching in particular has attracted scientific interest because it employs a genuine random process (coin tossing or yarrow stalk manipulation) to select from a set of 64 possible hexagrams, each with specific predictive content — a protocol that can in principle be subjected to statistical testing.
Risks & Considerations
Cognitive and interpretive risks. The most significant practical risk of precognition research is not physical but cognitive: the human mind is exceptionally prone to perceiving precognition where none exists. Confirmation bias (remembering the hits, forgetting the misses), post-hoc rationalization (reinterpreting vague premonitions to fit specific events), and the base rate neglect (ignoring the vast number of intuitions and dreams that do not correspond to future events) can create a compelling but illusory sense of precognitive ability. This is not a trivial concern — many people make significant life decisions based on perceived premonitions, and false confidence in precognitive ability can lead to poor judgment.
Anxiety and fatalism. Belief in precognition can generate anxiety (if the future is knowable, are negative future events unavoidable?) and fatalistic thinking (if the future is already determined, does present action matter?). Contemplative traditions that acknowledge precognition typically address this through a nuanced understanding of probability, free will, and the difference between perceiving tendencies and perceiving fixed outcomes. Without this framework, precognitive experiences — whether genuine or misinterpreted — can be psychologically destabilizing.
Exploitation by psychic fraud. Claims of precognitive ability are commercially exploited by fortune tellers, psychic hotlines, gambling system sellers, and stock market prediction services. The base rate of accuracy for such services is indistinguishable from chance (or from cold reading and public information analysis), but the psychological appeal of future knowledge creates a persistent market for fraudulent claims.
Methodological contagion. The Bem experiments contributed to the replication crisis in psychology by demonstrating that standard statistical methodology could produce compelling evidence for an extraordinary claim. This raised legitimate concerns about the entire body of social psychology findings that rely on the same methods. While the methodological reforms catalyzed by this concern have been positive, the process has been painful for the field and illustrates how anomalous claims, even if ultimately incorrect, can disrupt scientific confidence.
Theoretical vertigo. Serious engagement with the precognition evidence can produce a form of intellectual disorientation — a destabilization of assumptions about the nature of time, causation, and the relationship between consciousness and the physical world. For some researchers and practitioners, this destabilization is productive and intellectually stimulating; for others, it is distressing and destabilizing.
Significance
Precognition is among the most consequential topics in consciousness research because it challenges not merely our model of consciousness but our model of time — and with it, the foundational assumptions of causality that underpin all of science.
If the presentiment evidence is valid — if the human body physiologically responds to randomly determined future events before those events occur — then information is propagating backward through time at the biological level. This would require either a revision of physics to accommodate retrocausality at the macroscopic level, or a revision of our understanding of consciousness to allow it to operate outside the temporal constraints that bind physical matter. Either revision would be among the most significant scientific developments in history.
The Bem experiments carry additional significance because of their methodological context. Bem deliberately used standard experimental designs from mainstream social psychology — priming studies, habituation studies, approach-avoidance studies — and reversed the temporal order of cause and effect. The resulting data met the same statistical standards (p < 0.05) used to evaluate thousands of published social psychology findings. The implication is uncomfortable: if Bem's statistics are insufficient to demonstrate precognition, then the same statistical standards are insufficient to demonstrate the conventional findings they routinely validate. This forced a reckoning within psychology about the adequacy of null-hypothesis significance testing as a criterion for truth — contributing directly to the 'replication crisis' in social psychology and to the broader adoption of pre-registration, Bayesian analysis, and other methodological reforms. Whether or not Bem demonstrated precognition, his work catalyzed a positive transformation in scientific methodology.
The presentiment data is harder to dismiss than the Bem data because it is physiological rather than behavioral, automated rather than experimenter-dependent, and has been independently replicated across seven laboratories. The Mossbridge et al. meta-analysis (2012) — published in a mainstream neuroscience journal by a Northwestern University researcher, with Jessica Utts (president of the American Statistical Association) as co-author — represents the most authoritative statistical assessment of presentiment evidence. The effect is small (d = 0.21) but highly significant, and the analysis found no evidence of publication bias or methodological artifacts.
For contemplative traditions, precognition is not an anomaly but an expected capacity. The Yoga Sutras of Patanjali describe knowledge of past and future (atita-anagata-jnanam, Sutra III.16) as arising through samyama on the nature of change. The Buddhist tradition includes knowledge of future events among the abhinna (higher knowledges). The Sufi tradition describes kashf (spiritual unveiling) as including knowledge of events before they occur. The I Ching and other divination systems are built on the premise that the future can be accessed through specific procedures. These traditions typically do not describe precognition as seeing a fixed, determined future, but rather as perceiving the tendencies, probabilities, and patterns that are crystallizing in the present — a description that interestingly parallels the 'weak' interpretation of presentiment data, in which the body responds to the statistical distribution of upcoming stimuli rather than to a specific predetermined event.
Connections
Telepathy research overlaps significantly with precognition — some researchers argue that what appears to be telepathy in Ganzfeld experiments may be the receiver precognitively accessing the future reveal of the target rather than reading the sender's mind in real time. This 'precognition hypothesis' for Ganzfeld results illustrates the difficulty of distinguishing between different forms of anomalous cognition.
Remote viewing includes precognitive protocols (viewing targets that have not yet been selected), and the Princeton PEAR Laboratory found that precognitive remote viewing was as accurate as real-time viewing — a finding with profound implications for the nature of time in consciousness.
Psychokinesis research connects through the question of temporal directionality: if consciousness can influence random events (as PK research suggests), and if this influence operates retrocausally (influencing events that have already occurred but whose outcomes have not yet been observed), then PK and precognition may be aspects of the same underlying phenomenon.
Near-death experiences frequently include precognitive elements — experiencers report being shown future events during their NDEs that subsequently occur. These reports are anecdotal and difficult to verify but add to the convergent phenomenological evidence from multiple altered states that temporal constraints on consciousness may be less rigid than ordinary experience suggests.
Meditation and contemplative neuroscience connects through the yogic description of precognition as a siddhi arising through samyama — sustained contemplative attention to the nature of change. The Buddhist concept of dependent origination (pratityasamutpada) describes a web of causation that may not be strictly linear, and advanced meditation practice is traditionally associated with refined perception of causal sequences that extends to future events.
The Yoga Sutras provide the most systematic traditional framework, describing knowledge of past and future as arising through samyama on the three forms of change (dharma-parinama, lakshana-parinama, avastha-parinama) — a practice that develops the capacity to perceive the trajectory of transformation at a level deeper than ordinary temporal sequence.
Further Reading
- Bem, Daryl J. 'Feeling the Future: Experimental Evidence for Anomalous Retroactive Influences on Cognition and Affect.' Journal of Personality and Social Psychology 100(3), 2011.
- Mossbridge, Julia, Patrizio Tressoldi, and Jessica Utts. 'Predictive Physiological Anticipation Preceding Seemingly Unpredictable Stimuli: A Meta-Analysis.' Frontiers in Psychology 3, 2012.
- Radin, Dean. The Conscious Universe: The Scientific Truth of Psychic Phenomena. HarperEdge, 1997.
- Dunne, J.W. An Experiment with Time. Faber and Faber, 1927.
- Bem, Daryl J., Patrizio Tressoldi, Thomas Rabeyron, and Michael Duggan. 'Feeling the Future: A Meta-Analysis of 90 Experiments on the Anomalous Anticipation of Random Future Events.' F1000Research 4, 2015.
- Wagenmakers, Eric-Jan, et al. 'Why Psychologists Must Change the Way They Analyze Their Data: The Case of Psi.' Journal of Personality and Social Psychology 100(3), 2011.
- Cramer, John G. 'The Transactional Interpretation of Quantum Mechanics.' Reviews of Modern Physics 58(3), 1986.
- Knight, Sam. The Premonitions Bureau. Penguin Press, 2022.
- Radin, Dean. 'Electrodermal Presentiments of Future Emotions.' Journal of Scientific Exploration 18(2), 2004.
- Mossbridge, Julia, and Dean Radin. Transcendent Mind: Rethinking the Science of Consciousness. American Psychological Association, 2017.
Frequently Asked Questions
What is the strongest scientific evidence for precognition?
The presentiment (physiological anticipation) meta-analysis by Mossbridge, Tressoldi, and Utts (2012) provides the strongest evidence. Across 26 studies from 7 independent laboratories, the human body showed elevated physiological arousal (skin conductance, heart rate changes, brain activation) 2-5 seconds before the random selection and display of emotionally arousing versus calm images. The combined effect size was d = 0.21 with p = 5.7 x 10^-8. The analysis found no evidence of publication bias, selective reporting, or methodological artifacts. The key strength: the arousing/calm selection occurs after the physiological measurement, eliminating conventional explanations based on sensory cues or expectation. Daryl Bem's behavioral experiments (2011) show a larger combined effect (p approximately 10^-11) but have been more contested due to replication failures in some laboratories.
How did Daryl Bem's 'Feeling the Future' experiments work, and were they replicated?
Bem took standard psychological paradigms (priming, habituation, recall facilitation) and reversed the temporal order of cause and effect. For example, in a standard priming study, showing a word before a response speeds the response; Bem showed the word after the response and found it still influenced response speed — as if the future event reached backward to affect present behavior. Eight of nine experiments showed significant effects, with combined odds against chance of approximately 74 billion to 1. Replication has been mixed: a meta-analysis of 90 attempts (Bem et al., 2015) found an overall significant effect, but several prominent direct replications failed (Galak et al., 2012). The divergence partly reflects different analytical approaches (frequentist vs. Bayesian), different experimenter characteristics, and genuine debate about what constitutes adequate replication. The evidence is ambiguous rather than conclusively positive or negative.
Does physics actually allow for information to travel backward in time?
Several legitimate interpretations of quantum mechanics allow for retrocausal effects. The Transactional Interpretation (Cramer, 1986) proposes that quantum events involve waves traveling both forward and backward in time. The Two-State Vector Formalism (Aharonov et al., 1964) describes quantum systems using both past and future boundary conditions. Physicist Huw Price has argued that the fundamental laws of physics are time-symmetric — they work equally well run forward or backward — and that the 'arrow of time' is a macroscopic statistical phenomenon, not a fundamental constraint. These are published in mainstream physics journals and represent legitimate scientific positions. However, most physicists maintain that quantum-level retrocausality does not translate to macroscopic information transfer (precognition), and no widely accepted physical theory predicts precognition as currently understood.
How can I distinguish genuine precognitive experiences from coincidence?
The key principles are: (1) Record predictions before the events they concern — writing down dreams, intuitions, or premonitions with dates and specific details before knowing what will happen eliminates post-hoc rationalization. (2) Be specific — 'something bad will happen' is too vague to evaluate; 'I dreamed a red truck hit a white car at an intersection near my office' is specific enough to assess. (3) Track your base rate — record all predictions, not just the ones that come true, so you can calculate your actual hit rate versus chance expectation. (4) Consider alternative explanations — could the 'prediction' have been based on unconscious inference from available information, logical deduction, knowledge of someone's habits, or anxiety about a likely event? (5) Use large samples — any single hit, no matter how specific, could be coincidence; genuine precognition would show a pattern across many recorded predictions that statistically exceeds chance. The honest practice is systematic recording and honest assessment, not confirmation of a desired conclusion.