Remote viewing is the claimed ability to describe a hidden or distant target without ordinary access to it. It has been investigated, including in US government programs, but a striking match or an enjoyable exercise does not establish a reliable psychic ability.
What the historical record shows
The CIA’s account describes an independent review in 1995. Its summary reports some results beyond what it attributed to chance, while concluding that remote viewing was too unreliable and inconsistent for intelligence use. The agency did not restore the program. Government interest in a claim is not the same as confirmation that it works reliably.
Read the CIA’s explanation of its psychic research for that institutional history. This is a brief overview, not a complete review of the experimental literature.
Why casual matches are hard to interpret
A description such as “something bright near water” can fit many pictures. Seeing the target first and then deciding which details count makes the result difficult to evaluate. Cues from a helper or knowledge of the available pictures can also influence the task.
Recording misses is as important as recording apparent hits. A fair exercise needs rules set before the answer is known.
An optional blinded picture exercise
- Have a helper prepare several distinct, ordinary pictures and keep them out of your sight. Avoid disturbing images or private material.
- Agree beforehand on the number of trials and how choices will be scored. For a simple version, each trial ends with choosing one of four pictures.
- The helper randomly selects one target for that trial and stores the selection. Avoid interacting with the helper while making notes.
- Write and timestamp any impressions before seeing the four options. Do not revise the notes afterward.
- View the four options without knowing the selected target, and make one final choice.
- Reveal the target and record the outcome, including misses. Keep all trials rather than only the most interesting ones.
With four equally likely targets, a random choice has a one-in-four chance of being correct on an individual trial. A few correct choices can occur by chance. This home exercise is not a validated test and cannot establish a mechanism or dependable ability.
A worked record: keep the choice separate from the impression
Before starting, assign the four pictures the labels A, B, C and D. Keep the helper’s randomly selected target hidden until after you lock in one choice. The helper should prepare the options and the target record before receiving your notes, so the pictures cannot be picked to fit what you wrote.
The following four trials are invented to show how a record works; they are not results from a real participant. An exact label match is the only success criterion in this example. A colorful description does not turn an incorrect final choice into a hit.
| Trial | Choice locked before reveal | Target revealed | Recorded result |
|---|---|---|---|
| 1 | C | A | Miss |
| 2 | D | D | Hit |
| 3 | A | B | Miss |
| 4 | B | C | Miss |
This made-up record contains one correct choice out of four trials: 25 percent. That describes these four rows only. It is not an estimate of someone’s ability, and the one-in-four chance on each equally likely selection does not require every group of four attempts to contain exactly one hit.
If a cue gets through: retain the row and note what happened, such as “the target name was visible before I chose.” Follow the rule you agreed in advance for an interrupted trial. Do not quietly drop a miss or retry until a hit appears.
Next step: copy the four column headings into a notebook, add a date and time for each locked choice, and store the original impressions separately. Compare the final choice with the target using the rule you set before the reveal.
Keep the exercise low stakes
Do not use impressions to locate missing people, accuse someone, assess illness, make financial decisions or override privacy. Use ordinary consenting participants and harmless materials, and stop if it becomes distressing or compulsive.
What you can learn without claiming proof
You may notice how you generate images, how broad descriptions feel specific after feedback or how your confidence differs from results. Those observations can be interesting in their own right.
For another example of distinguishing meaning from evidence, see patterns and angel-number interpretations. Keep curiosity alongside honest records and willingness to be wrong.