On Mon, Oct 12, 2026 at 8:51 AM Matt Mahoney <[email protected]> wrote:
> I believe the operational definition of consciousness that James is > referring to is the mental state of wakefulness, > No. I'm referring to what might be referred to as paranormal perception. Perhaps I should make clear a few things that bias my position somewhat: 1) I worked at SAIC during the era in which this work was being conducted and rubbed shoulders with some of the folks there most likely involved with this work while they were doing it. 2) I worked directly with the Boundary Institute circa 2000-2004 while Ed May was there, and I did so because of my support of Tom Etter's work that provided the theoretic basis for the Boundary Institute's investigation of paranormal perception. 3) I have personally experience remote viewing, so it is rather like Jacques Vallee's position regarding "UFOs" -- he is biased having not only observed one but been present when data relevant data was being destroyed precisely because it would provide hard evidence. Having said all that, the SRI protocol was flawed, which allowed the work there to be discounted in Nature. However, the SAIC work is another matter. Grok4.7 on the SAIC refinement: *The SAIC remote-viewing studies (roughly 1992–1994, under Edwin May) addressed the specific early SRI flaw—chronological cues in transcripts plus non-randomized/ordered target lists—by redesigning target selection and judging so that sequence information was no longer available to the judge.* Key changes included: - *Proper randomization of targets and judging sets.* Targets were selected with a tested randomization procedure. For judging, each response was paired with a set of alternatives (typically the correct target plus several decoys) in which every alternative had an equal prior chance of being the actual target. The order of presentation to the judge was randomized, so knowing the sequence of trials no longer mapped onto the list. - *Rank-order judging with a blind judge.* After the session, a judge who did not know the correct target ranked how well the response matched each of the alternatives. No one who knew the specific target had access to the response until after judging was complete. - *Duplicate target materials.* Separate physical sets of targets/photographs were used for the experiment itself and for judging, reducing the chance of physical handling cues. - *Trial independence.* The earlier practice of sampling targets without replacement (which created sequential dependencies) was eliminated, so knowledge of previous trials gave no information about later ones. These steps removed the pathway that had allowed a judge in the early Price/Hammid series to match by chronology alone. In the 1995 American Institutes for Research review, Ray Hyman noted that the SAIC experiments were methodologically superior to the earlier SRI work precisely because they avoided the major flaw of non-independent trials and ensured proper randomization for both target presentation and judging. (Remaining criticisms of the SAIC work focused on other issues, such as the repeated use of the same judge across experiments, but those are separate from the original chronological-cue problem.) > > ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T2784bbfe524b0b93-M2a0bcc562c0f98e7cb1a5a09 Delivery options: https://agi.topicbox.com/groups/agi/subscription
