In the early days of Remote Viewing research in 1972 at SRI-International (back then Stanford Research Institute), an unstructured methodo­logy and a form of free response interview were used. A similar approach was adopted at the mili­tary Remote Viewing unit located at Ft. Meade under the direction of F. Holmes “Skip” Atwater, where some of today’s best known remote viewers worked during the period of the “Grill Flame” pro­ject (1977–1983). Preparations and viewer selection were carried out in cooperation with the SRI team. From 1979 onward, the military ran its own operational remote viewing effort.

During this period, the term Extended Remote Viewing (ERV) came into existence, coined by Skip for the military variant of the method, which was based on elements of the original SRI approach (“freeform” RV or Generic Remote Viewing, GRV) incorporating elements and insights from Bob Monroe’s research work.

Ten years after the initial start of Remote Viewing research, in 1981, the way of working changed to a new level, and the CRV protocol known today was developed [Figure below]. Key players in this development and testing effort were, on the SRI side, “inventors” Hal Puthoff and Ingo Swann, and on the military side, Rob Cowart [1] and Tom McNear, who was the first “official military guinea pig” in training with this new protocol. Beginning in 1984, a group of military personnel from Ft. Meade, now in the Center Lane program, trained in the new methodology under the guidance of Ingo Swann. CRV—the acronym then stood for “Coordinate Remote Viewing,” later redesignated as “Controlled Remote Viewing”[2]—was, in contrast to the methodology previously used, a structured methodological approach based on a multi-stage protocol sequence.

The CRV methodology was established in the early 1980s in response to a directive from the commissioning authorities, tasking the researchers with the development of a trainable approach to remote viewing. This mandate aimed to extend the application of remote viewing beyond natural psychics, enabling the training of regular military personnel in the utilization of this technique.

The operational use of remote viewing to obtain information is, by definition, accompanied by the fact that complete information about the target of the session is not known in advance. The protocol emphasizes that the viewer operates under blind conditions, enabling subsequent analysts to search for elements of information that are already known.

In an interview conducted in 2021, Hal Puthoff explained that during the developmental phase of CRV, a 4-step evaluation matrix was employed to assess the accuracy of the diverse approaches being tested at that time (as shown in the graphic below). Given the extensive volume of thousands of individual tests conducted, this simplified approach sufficed for the primary objective of identifying the most successful method from a range of variants. Nonetheless, the evaluations conducted during that period revealed a significant improvement in the hit rate with regard to the target when the remote viewer employed the Stages approach.

Accuracy Rating of two samples (97 & 93 trials), for the same remote viewer before and after receiving CRV training. The latter series results in an average score of 1.9 (63.4%), which is the base for our assumption of at least 50–80% correctness as mentioned in hypothesis 1. Source: Puthoff, H. E. (SRI International, 1984). RV Reliability, Enhancement, and Evaluation (U), p. 11.


[1] Rob Cowart had to leave the program after a short time due to health issues.

[2] The change of the term was initiated by Ingo Swann himself in the period after disclosure of the program (1995). In all official documents and reports from the previous period, which are available through the CIA library, also in the relevant manuals, the term Coordinate Remote Viewing is used.