Social Parapsychology
Each of us occasionally makes a decision that, in hindsight, turns out to be remarkably clever or remarkably foolish. Amplify this decision by a few million people, and the resulting collective decisions can be incredibly intelligent or stupendously stupid. These extremes are discussed in two recent books, The Wisdom of Crowds by James Surowiecki, and Collapse: How Societies Choose to Fail or Succeed by Jared Diamond. Surowiecki explores the fact that groups can sometimes be smarter than the most brilliant individual in the group--better at problem-solving, complex cooperation, decision-making, or predicting the future. Diamond explores how collective behavior can sometimes lead to breathtaking stupidity, resulting in the collapse of whole civilizations.
The concept of collective amplification of brilliance or madness prompted me to ponder one of the newest horizons in psi research--social parapsychology. Whereas parapsychology has traditionally focused on individual psychic experiences, social parapsychology investigates collective psychic experiences. Until recently, social parapsychology was an interesting idea, but very difficult to study. With the advent of the Internet, research is now becoming feasible.
As readers of Shift know, the Institute of Noetic Sciences has been at the forefront of this new domain through our participation in the Global Consciousness Project, headed by Princeton University psychologist Roger Nelson, and through our online psi tests. These programs are collecting massive amounts of data that can be used to investigate collective psi. As of Spring 2005, the psi games at www.gotpsi.org (hosted by the Boundary Institute) and at www.psiarcade.com (hosted by IONS) have generated over 70 million individual trials from nearly 300, 000 people worldwide.
One of the games at www.gotpsi.org assesses precognitive ability. It tests how well a user can describe a photo that the computer will randomly select after the user's predictive description of the unknown photo is entered. Because this test asks people to imagine a visual scene they've yet to see, I thought it would be interesting to see whether premonitions might enter into the imagination. I examined the words that people used to describe their imagery from September 9, 2001, through the morning of September 11, 2001. This included a set of nearly 900 trials and just over 2, 500 words.
On the morning of September 9, a user nicknamed sean wrote the following words in a series of three successive trials:
airliner (seen from left-rear) against stormy cloud backdrop, flashes of streaky cloud, ovoids, two persons firstly a dragonfly? then a log [or] branch suggestive of Everglades, then a fast dynamic scene of falling between two tall buildings, past checkered patterns of windows first tall structure like an industrial chimney, then flashes of rounded crenulated form--peacock-like headdress of American Indian woman? then surface like volcanic ash plume or cauliflower
The user's precognitive descriptions didn't match the photos subsequently selected at random by the computer, but they do provide a rather startling impressionistic sense of the terrorist acts in New York City on the tragic day now known as 9/11. The next afternoon, a user named shakey wrote these words in two successive trials:
it is of something falling; it will be a chaotic scene
Again these words were a poor description of the targets, but they are meaningful in the context of 9/11. A half hour later, a different user, nicknamed justatest, wrote in four successive trials:
Intense . . . too hot to handle; blasting; is the coast clear? they were checking the coast!!!
The following morning, September 11, about an hour before the first airplane crashed into the World Trade Center, user xixi wrote the following words in a series of 11 trials:
White House; gone in the blink of an eye; scald; man's folly; band red; surging; palace; not easily conned; US power base; flexing muscles; surprise.
Are these genuine premonitions of 9/11? The ideas suggested by these words seemed to be unusual in the context of this experiment, as most of the photos used in the test are of benign landscapes, people, animals, and other scenes with neutral content. Still, this is just a handful of potentially interesting matches out of 900 trials, and they might have been purely due to coincidence. So I devised a way to judge whether the words used prior to 9/11 were in fact unusual from the point of view of collective performance.
I considered all online precognition trials that included word descriptions contributed from September 2000 through June 2003; this amounted to 256, 000 trials containing 841, 000 words. For each trial, I matched the words each user entered against a set of nine concepts capturing the turmoil of 9/11: airplane, falling, explode, fire, attack, terror, disaster, pentagon, and smoke. The idea was to see how closely the words used by each person each day matched these concepts. Counting only exact word matches wouldn't be sufficient because someone might have used a synonym that an exact word-match would overlook. So I developed an automated concept-matching technique to form an average score indicating how closely the words used on a daily basis contained terrorismrelated concepts.
At first glance, these scores appeared to fluctuate randomly from one day to the next. On closer inspection, however, I found something rather unusual. Rather than increase in value, as might be predicted if lots of people were suddenly having premonitions of disaster, a few days before 9/11 the scores dropped to their lowest point in the three years of data examined.
To see how meaningful this was, I compared the observed drop against similar values based on randomly scrambled datasets. I found that the joint odds against chance of obtaining a "terrorism concept" score as low as the observed minimum, and falling as close or closer to 9/11 as observed, was 3, 300 to 1. This means people were significantly avoiding concepts associated with terrorism just prior to the catastrophe.
As it turns out, I found a virtually identical result in a different gotpsi.org game (an ESP card-guessing game). Collective drops in psi performance in the days before 9/11 suggest that many people unconsciously began to sense trouble brewing, but there was no context for those feelings so they were actively repressed. Only the rare individual can avoid personally identifying with negative thoughts, and fewer still are willing to publicly admit that they're having such thoughts. You can't ask people to report premonitions that they've repressed, but interesting patterns do come into focus when daily collective psi performance is analyzed.
I believe that interest in social parapsychology will undoubtedly continue to grow as we find new ways of using the Internet and the media to study the nature and possible applications of collective psi.
DEAN RADIN, Ph.D. is IONS' Senior Scientist. Dr. Radin earned a Masters degree in electrical engineering and a Ph.D. in psychology from the University of Illinois, Champaign. He has held research appointments at Princeton University, Edinburgh University, University of Nevada, and several Silicon Valley industrial research labs, where he has conducted basic research on exceptional human capacities, including psychic phenomena. His books include The Conscious Universe and the upcoming Entangled Minds. Dean's website: www.deanradin.com 
