Interview for the Global Intelligencer |
|
In early March 2007, Cate Montana, the publisher of The Global Intelligencer wrote to propose an interview.
She sent a list of good questions of the sort we might address in the FAQ, and they are presented here in that format. If, as you point out on the website, "Single events are not generally capable of showing effects reliably; this requires 30 to 50 events." what events have been capable of showing effects reliably? The statement is general. No single event can be counted on to show a significant departure from expectation. Those with lots of people engaged have a tendency to larger effects, and I believe a good case can be made for events that embody or evoke compassion tending to show deviation. But the formal result is strong only because there is an accumulation across many tests. If single events are generally incapable of showing effects, doesn't this obviate your research? No. The research is exactly like any scientific study of subtle phenomena. It requires patient replication of conceptually similar experiments -- whose separate results can be concatenated to yield a reliable statistical assessment of the question of interest. This is similar to meta-analysis in medicine, where many separate experiments are combined to determine, e.g., whether a drug is effective (and not unacceptably harmful.) So it is possible to learn how something like meditation, prayer, or Deeksha is seen by the EGG network only by patient repetition of well-designed tests of the hypothesis that there will be correlation. Are meditation effects significantly different from other effects? Is it possible that repeated tests designed to find a correlation fabricate a correlation out of that intent? Is this one of the reasons why statistical averaging is necessary.. you screen this factor out? Yes, we can learn something if we ask good questions. We don't know yet whether meditation effects are significantly different from other effects, but if that question is formulated intelligently, it will be possible in principle to find out. The question about "repeated tests designed to find a correlation fabricat[ing] a correlation" is another that if intelligently formulated as a hypothesis test might be answerable. To the extent I understand your interest, this is similar to the question of anomalous experimenter effects, or "observer effects" which are the subject of some long-term study by my group and others. As for screening out such effects, we're open for propositions, but it is hard to imagine an experiment without an experimenter, or a question which nobody asks. The Global Deeksha event November 23rd 2006. How many participated? I don't know. Our basic hypothesis test does not require that information, although we do secondary analyses on the effect of numbers of people, using best estimates (usually just large numbers vs small). In the global deeksha event the results was: The modest deviation with Chisquare 3613.3 on 3600 df, a p-value of 0.435 and Z = 0.164 . Could you explain these results in layman's terms? This is a classic null result, and an excellent example of the rule that we cannot learn much from single events. The deviation is positive, but small. Only with many repetitions of a small result like this can the signal emerge from the background noise, assuming there is any signal.
In the September 11 2001: Exploratory and Contextual Analyses you write: I
want to acknowledge that I like the notion of Global Consciousness, but that
this idea is really an aesthetic speculation. I don't think we have real
grounds to claim that the statistics and graphs representing the data prove
the existence of a global consciousness. On the other hand, we do have strong
evidence of anomalous structure in what should be random data, and clear
correlations of these unexplained departures from expectation with
well-defined events that are of special importance to people. The events
share a common feature, namely, that they engage our attention, and draw us
into a common focus.
It is speculation, and I think it is aesthetically pleasing, so: "aesthetic speculation". The statement was written years ago, and in the intervening time, evidence that the effects may represent a suitably defined "global consciousness" has accumulated more persuasively. - If you do have "strong evidence of anomalous structure in what should be random data," what other possible explanations are being proposed than global consciousness effects? The list could be long, but a few examples are: calculation errors, inappropriate statistics, inadequate data specification, physical fields (like EM, cosmic flux, temperature changes), fraud, misunderstanding, chance fluctuation, coincidence, etc. Scientific research is very much about eliminating alternatives like these, and that's where we spend a large share of our design and analysis energies. - Do you find there is a specific "emotional" trigger that is more powerful than others? For example, does grief tend to create a greater effect than say, excitement? Or in your speculation, do the effects result simply from large numbers of people having a concentrated focus on one event at one point in time? The closest to that is compassion, as I mentioned earlier. Large numbers are necessary, but not sufficient. World Cup Soccer does not produce a reliable deviation. But damaging earthquakes on land do. What are your normal baseline readings with the eggs actually picking up? Do you take into consideration such large potential factors as picking up information from the zero point energy field? Analysis of the whole database, without identifying special times, shows random behavior according to the design specifications and statistical expectations. Nobody knows how to "pick up information from the zero point field". That's a speculative notion that might be a starting point for theory, but it is not empirically accessible. With September 11th you write "The final probability for the formal hypothesis test was 0.028, which is equivalent to an odds ratio of 35 to one against chance." That doesn't seem very high considering the enormity of the event. Please give some kind of context for this figure. This is another instance of the small effects we deal with. Though many people think 9/11 is special and ought to show a big effect "considering the enormity of the event" there is no a priori justification for this. But to your point: we have looked seriously and carefully at the context for this pre-defined formal event. It was specified as a little over 4 hours beginning close to the first crash, and gives the p-value you quote. When we look at what the data do in the days and weeks surrounding this event we discover that: 1. The trend that begins near the first crash continues for 50 hours, and if we had had the foresight to predict that (instead of 4 hours) the p-value would be perhaps 1/3000. 2. The variance (an independent statistic) fluctuates over an enormous range (7 sigma) on 9/11, and if we had had the astounding prescience to be able to make a prediction about that, the p-value would be 1/100000000000 (one in a trillion). 3. The data are linked across time (we call it autocorrelation) over lags up to 2 hours. One of the defining characteristics of random sequences is independence (no autocorrelation). If we had known to predict this, we would have another independent indication of a situation that could occur by chance only once in a million tests like this. Was 911 the single most significant event registered so far? No -- but once more: The single event outcomes are not reliable statistics. They are correct, of course, but the signal-to-noise ratio is huge. Real deviations attributable to the events may be buried in noise (like a needle is buried in a haystack) and on the other hand, the normal statistical fluctuations (noise) can also masquerade as signal. This can be dealt with only by patient repetition or replication of the scientific tests. How are the "egg" sites chosen? Opportunistically. When someone volunteers to host a site and has adequate skills, equipment, and interest, and lives somewhere we would like to have an egg, we set it up. Are there any eggs on satellites? It seems possible that an "egg array" further outside other possible geological/physical/electromagnetic interference effects might register events differently. No. If the opportunity were to arise, we would happily set up an egg on a satellite. Not to avoid interference like you suggest (the data are fully protected from it already), but to have such a cool site, and, more seriously, to add to the distance array we have, and to be able to assess the effects, if any, of a dynamic location. How is the GCP results being received and interpreted by the scientific community? Not much notice is taken, except in the community of scholars who are interested in anomalies and consciousness research. We have had some publications in mainstream physics journals, but most are in specialty journals such as the Journal of Scientific Exploration. After all these years of research, what do the cumulative findings say to you? There is solid evidence of structure where there should be none -- in otherwise demonstrably random data. The departures from expectation we find are correlated with events of importance to humans. The findings thus indicate and provide substance for deeper questions in physics, and in sociocultural and psychological domains. Personally, I think the data say we are interconnected, and though we cannot perceive it because the phenomenon is subtle, this means we are directly linked to each other and to the world in which we live. |