Dreams come true at uncomfortable rates. The phone rings, and the survivor already knows who it is. The survivor pulls their child off the playground equipment seconds before something breaks. Researchers documented this for decades before it was respectable to discuss in the medical literature, and the data has been quietly accumulating since the 1970s.
Precognition in NDE survivors tends to show up in three forms. Dream precognition: the survivor has a specific dream, and within days the content of the dream happens. The dream is usually short, sometimes only a single image or a single sentence, and it is usually mundane (a friend's car accident, a phone call from an aunt, a job offer). It is not normally apocalyptic or dramatic. It is small, specific, and verifiable after the fact.
Waking knowing: the survivor has a sudden conviction that something is about to happen, and acts on it. The classical version, reported across multiple researcher datasets, is the survivor who suddenly decides not to take their usual route home, and learns the next day there was a serious accident on that route at that time. Smaller versions happen weekly: the survivor changes lanes, gets up to walk the dog, calls a friend, and the change of behavior turns out to have been load-bearing.
Anticipatory awareness: the survivor knows who is on the phone before they look at the screen, knows what their partner is about to say, knows that there is bad news before the messenger has opened their mouth. This is the most common form, and it is the form that family members most often confirm independently.
"I do not know how to talk about this without sounding like I am bragging or losing my mind. So I keep a private log, and I show it to my brother, who keeps the other half of it because he refuses to believe me without the dates. We have, between us, 162 documented hits in three years. The dream of the train going off the tracks in Pennsylvania, six days before it happened. The conviction that my mother needed to go to the emergency room on a Tuesday morning, when she felt fine. The morning the call came from my best friend's wife and I picked up the phone before it rang. My brother is the most skeptical person I know. He believes me now." — experiencer journal, presented in Sutherland, Within the Light (1993)
The methodologically careful studies, especially Sutherland's Australian fieldwork and the parapsychology-adjacent work of Charles Tart and Russell Targ, suggest hit rates well above chance in selected NDE-survivor populations under structured testing conditions. The studies are not numerous, and they have not been replicated at scale, but they exist. The popular press treatment of the topic is generally either dismissive or credulous, and the actual literature is between those positions: the effect is small, it is real in the data that has been collected, and it has not been adequately studied because the topic is professionally costly for researchers.
Pattern recognition at increased resolution. The standard skeptical account, which is also the most generous. The survivor is reading the social and environmental field at higher resolution after the experience (see Heightened Empathy), heightened empathy). They are picking up subtle cues, integrating them unconsciously, and producing predictions that feel like precognition but are actually fast unconscious inference. The objection is that this does not account for the verified dream cases, where the survivor was asleep and was not receiving environmental input.
Statistical inflation by selective memory. The standard skeptical fallback. Survivors remember the hits, forget the misses. The objection is that the notebook studies, including the brother-as-control example above, attempt to correct for this and still show elevated rates.
Nonlocal consciousness. The framework Dean Radin, Pim van Lommel, and Roger Penrose (in different versions) have proposed. Consciousness is not strictly local to the brain, and after the experience, the survivor has more access to nonlocal information. This is the survivor's framing as well. It is unfalsifiable in the strict Popperian sense, though Radin has run laboratory presentiment studies that show small, statistically significant effects in general populations. The NDE-survivor version of those effects, if it were studied at scale, would be expected to be larger.
The retrocausal frame. A subset of physicists (Cramer's transactional interpretation, Aharonov's weak-measurement work) allow for time-symmetric information flow. If consciousness can interact with retrocausal information channels, even weakly, then a small precognitive signal is consistent with established physics. This is fringe and is not the standard textbook view, but it is not nothing.
Long-term experiencers tend to treat the precognitive signal as advisory rather than authoritative. They do not bet money on it. They use it for small decisions, especially decisions about safety and about who to be around. Many keep the logs that the case above describes. The logs serve a private function: they keep the survivor honest about the rate, and they protect against the drift into magical thinking that the daily texture of the phenomenon makes seductive.