The goal of this unit is to introduce you to the basic principles of PEAR's experimental device design, and to explore the various correlations found over the course of more
than 25 years of laboratory experimentation with the electronic REG devices.
Experimental Design
Random Sources
The goal of PEAR's human/machine experimental design was to construct devices that produced truly random data that would allow for the occurrence of anomalous consciousness-related events but could also be reliably used for rigorous statistical analysis. The source of the randomness varied from electronic noise to mechanical, optical, and fluid dynamical systems, but all were designed to have truly random but statistically reliable outputs.
It was necessary for the devices to produce data quickly and to be isolated from known physical influences on their random output. To that end, all of the devices were built with commercially available parts, and with well-known technologies used in other fields. They were also all extensively calibrated and shielded to insure that they would behave nominally in a statistical sense.
Creating devices with truly random outputs that were statistically well behaved was necessary so that the consciousness-influenced data produced during experimental trials could be compared to well known statistical behavior.
Tri-polar Protocol, Differential Logic
Beyond physical shielding and empirical calibration, the PEAR lab also used a tri-polar protocol and differential logic to isolate the primary variable of operator intention. They didn't simply ask the operators to try to get high numbers (for example) and compare that with calibration trials. Rather, while everything else in the experiment remained constant, the operator would change their stated intentions from one experimental run to the next. In one run, an operator would try to influence the random process to produce high numbers, in another, to try to produce low numbers, and in a third, the operator would have no intended influence at all.
Comparing the results of these three separate data streams would allow the experimenter to see if there was a change in the statistical results that corresponded to changes in the operator's intention. This tri-polar protocol would isolate intention as the primary variable.
If intention was the only changing variable, any comparison that revealed an aberration in the chance behavior of the devices would either have to be associated with the intervention of the operator's psyche, or it would have to be an artifact that was smart enough to track what the operator wanted to happen. Either way, this positive result would not be explainable by conventional physics, and would be evidence of a consciousness-related anomaly of some sort.
Electronic Random Event Generators (REGs)
The microelectronic Random Event Generator (REG) was the first device used by the PEAR lab, and it was the device most frequently used in their experimentation. The following video serves to introduce you to the statistical character and the operator experience using an REG.
REG Demonstration (video)
REG Demonstration - 18 min 11 sec
REG Operation
The REG is essentially a rapid electronic coin flipper, designed to quickly produce a random sequence of ones and zeros. Various controls and fail-safes were included in its design to insure a completely random noise signal, independent of environmental artifacts or drifts in the random noise signal.
A single trial in a PEAR experiment usually constituted 200 "coin flips," or 200 bits. A trial would produce a random numerical value (ex. 97, 102, 108, 89, etc.) that was usually within the range of plus or minus 15 from the theoretical statistical mean of 100.
The operator would record their intention before beginning of each run, which in most experiments was a sequence of 50, 100, or 1000 trials generated at a rate of one trial per second. The intention, as discussed above, would be to produce higher numbers, lower numbers, or, in the baseline condition, the operator would not try to influence the device at all. When producing a series of runs, an operator would alternate among the intentions from one run to the next.
Operators were not given instructions on how to try to influence the device, but were encouraged to come up with their own methods of mentally influencing the device that they felt would work for them. They used a variety of techniques, ranging from intense concentration, to meditation, to stating an intention and then only giving it peripheral attention.
The advantage of using the electronic REG rather than flipping real coins was not only that it produced a more reliable 50-50 distribution of "heads" and "tails" (or ones and zeros) than a physical coin flip, but more importantly, it produced the data much more quickly. Producing large quantities of statistical data quickly allowed the PEAR lab to develop very large databases in a short amount of time. Because probabilities are calculated not only based on the size of the effect, but also on the size of the database that shows the effect, these large databases gave the PEAR lab the ability to detect subtle deviations from chance that would be impossible to detect in smaller databases. Even very small correlations between intention and the output of the REG could quickly become very statistically significant if repeated often enough.
Human Machine Interactions Lecture
Lecture Presentation (video)
In the lecture video below, Bob Jahn, the director of PEAR, discusses the laboratory REG experiments and the various correlations seen with operator intention and other parameters.
Bob Jahn - Human-Machine Lecture, Part 1 - 40 min 30 sec
Lecture Summary
Correlates
In the REG experiments the PEAR lab found statistical evidence of very strange consciousness-related, and subjective-correlate-related effects on the electronic REGs.
The most salient variable was the operator's intention; the devices frequently showed small but significant correlations between the operator's stated intention and the output of the device. The effect size wasn't large enough to win at the roulette table, but by the standards of statistical science the probability of this occurrence was astronomically small.
This positive result wasn't simply the case for a select few outlying operators, but rather, the effect seemed to be generalized among the population as a whole.
The data also showed that men and women influence the device in different ways. Men usually tended to have modest effect sizes, but were more likely to get a results in their intended directions. Women, on the other hand, tended to produce asymmetrical results, with all three intentions, including the baselines, going in the high direction. Even though more women got inverse results in the high minus comparisons, as a group, they generated larger effect sizes and larger statistical variances than the men.
When operator efforts were combined in co-operative experiments, the gender of the pairings also made a big difference. When males and females were paired together, their effects were enhanced, whereas same-sex pairings had negative or inverse effects on the REGs. Not only that, but the opposite-sex pairings in which the pair were a "bonded-couple," in a romantic relationship, for example, the effect was magnified from six to seven times what either operator could achieve on their own. Experiments where operators competed with each other, on the other hand, only produced chance results.
There also seemed to be confirmation that there is something true about the common gaming term "beginner's luck." Operators tended to do better when they first tried the experiment than they did during their next series, although the effects usually recurred after the fourth series.
This "series-position" line of analysis also documents a common phenomenon in other areas of anomalies research, popularly called the "decline effect," where experiments first show positive results but then show nothing in later experiments and replications. PEAR's data documents this decline, but in the case of PEAR's experimentation, they produced data long past the decline, and saw a resurgence of the effect size after the decline.
Some of the things that did seem to make a difference in the expression of the statistical mean shift were: operator gender, gender pairing, emotional "resonance" or bondedness, and series position.
But there were perhaps even more perplexing variables that did not seem to matter.
Instructive Non-Correlates.
The finding that distance and time did not correlate with the ouput of the REGs, was one of the most perplexing results the PEAR lab encountered.
When an operator was intending to influence the REG, it did not matter how far away he or she was from the REG. Operators thousands of miles away seemed to have the same effect size as operators sitting in front of the REG in the lab.
Neither did it seem to matter when the operator exerted their effort to influence the REG. Within the span of plus or minus a few days, the PEAR lab found no correlation between the effect size and the variable of time.
Both of these notable non-correlates are very instructive, while also being very challenging for future theoretical modeling. They show us that we need to look beyond our current physical models to account for these anomalous consciousness-related influences. Not only do these effects defy known physical mechanisms, but they do not even fit within some of our most basic expectations of how the world works.
Unit Conclusion
That an REG responds to the conscious intention of a human operator is evidence that there's more going on in the physical world than can be simply explained by our physical models. We are seeing that consciousness plays a central role in defining, not only how these anomalous phenomena may originate, but in the determination of aspects of the physical world itself.
In our investigation of consciousness, not only do we now have to face the idea that what happens in our interior world may have a physical result in the phenomenal world, but we are beginning to see that there is a very complex inter-relationship between the conscious actors and physical manifestations.
It turns out that the nature of that consciousness-related influence is also dependent on numerous subjective factors. We still don't know how to isolate some of these subjective factors, and some of them are surprisingly commonplace.
Thinking for a moment about some of these mundane factors, it seems fairly reasonable that variables such as gender or environmental conditions could affect performance on a given task. But when we consider the context of what this particular task is - expressing an intention - we have to wonder, how is a female stating an intention different from a male stating an intention?
Does this difference suggest that different genders interact with the world in inherently different ways? Or that there is something different about the character of their consciousnesses?
While we give each other the benefit of the doubt that each one of us experiences consciousness, there is little reason to believe that all consciousnesses are the same, or that an individual's consciousness remains constant through time. Consciousness itself could be different for separate individuals, or discontinuous within the same individual.
The work of plumbing the depths of these subjective variables, and what they could mean for our understanding of consciousness has just begun.
Suggested Activity
The researcher Helmut Schmidt developed a very similar experimental protocol as that used by the PEAR lab, and at one point developed Java applets to allow internet users to become operators of his RetroPsychoKinesis Experiment(cache). Although at this point the experiment and website are no longer actively updated, the experiment is still functioning and recording data. Take a look at the website and run a few experiments to get an idea of what the operator experience is like. Although these experiments have different operator feedback displays than the PEAR experiments, the fundamental task and operator experience are very similar.
While you do the experiment online, take the opportunity to observe yourself watching the data emerging. Do you notice how sometimes you seem to perceive a trend, and sometimes that trend seems to disappear as soon as you notice it? Do you wonder what is a real pattern, and what is pure chance? How are you reacting to the experiment? Do you notice your own resonance with it when it’s going in your stated direction? Do you have a desire to pull it, or push it? Do you feel a belief or disbelief in yourself or the result? It’s an interesting game, watching your own mind interacting with the world of random events.
(cache). Although at this point the experiment and website are no longer actively updated, the experiment is still functioning and recording data. Take a look at the website and run a few experiments to get an idea of what the operator experience is like. Although these experiments have different operator feedback displays than the PEAR experiments, the fundamental task and operator experience are very similar.
