Connecting Global Consciousness

We measure the effects of human consciousness via a globally distributed network of physical devices that produce random numbers. These devices are called random number generators (RNGs). Our hypothesis is that shared consciousness can cause the network to stop behaving in a random fashion. This will occur either when large numbers of people share a focused attention towards the same thing, such as a global event that draws out compassion, or smaller numbers of people are in a more coherent state and hold a collective intention. In other words, our collective consciousness can change the physical world.

This effect is not limited to a network of unusual devices. We understand the devices to be detectors for a much broader effect. If human consciousness can affect these electronic devices, then it would affect other physical systems. Other experiments, such as those using organic RNGs, suggest that these effects extend to a broader range, including people, water, plants, trees, weather, and nature in general. The effects can be healing or harmful depending on the intention. However, consciousness effects are not readily observed in all objects because there are other competing influences on their behavior. That is why we use RNGs to detect this effect, as they are free from other influences.

The original Global Consciousness Project (now called GCP 1) was created in 1997 at Princeton University by Dr. Roger Nelson and a group of researchers working at the boundary areas of physics and consciousness. We are honored that in 2020, Dr Nelson asked the HeartMath Institute to become the new home base for the Global Consciousness Project. Over the course of 20+ years, many events were analyzed, including celebrations of New Years, large-scale meditations and religious events and shocking events like terrorist attacks and natural tragedies. The composite statistic for the project shows a 7-sigma departure from expectation, indicating a probability on the order of 1 in a trillion that the correlation of the GCP data with global events is merely a chance fluctuation. The evidence clearly suggests that focused emotional energy and attention can interact with and affect the physical world.

Expanding on the groundbreaking findings of GCP 1, the GCP 2.0 network is much larger. We designed a new device with 4 RNGs per device and plan to build 1000 of these for a total of 4000 RNGs, the largest RNG network on the planet at the time of its design. This is more than 50 times as many RNGs as GCP 1 at its peak of about 60 RNGs. It also uses the latest technology and cryptographic standards for randomness. We are tracking signals at every stage of the process including analysis of the quantum noise sources used in the process of producing a random number, so that we can now dig into the effect of consciousness on the underlying electrons and voltages. Finally, due to HeartMath's extensive collaborations and network of citizen scientists and certified trainers, GCP 2.0 has a broader reach and an explicit mission to engage as many people globally as possible.

There are several advantages to this. It is an international collaboration of citizen scientists all across the planet, encouraging global engagement in the project and feeding the consciousness fields with more coherent emotions such as love and compassion. Additionally, it has been demonstrated that having more devices in the network may result in a more sensitive detection system, but more importantly a larger network will allow us to examine topological effects in how consciousness-related effects distribute around the world when localized events take place.

This should allow us to show unprecedented striking evidence of the effects of human consciousness on the physical world.

The more devices there are, the more sensitive the network is to consciousness effects.

The RNGs are designed to generate truly random numbers. We have cleaned (whitened) out any other influences or biases coming from environmental effects such as the temperature of the room. That way, when we do see non-random effects, we attribute it to unconventional causes like collective consciousness. For more details on how we ensure the numbers are random, see the RNG technology section.

One approach is to generate hypotheses before running our analysis, which is the standard way to perform rigorous statistical testing of an idea. We choose events to analyze based on our understanding of collective consciousness effects, adding them to our list called the hypothesis registry. Then, if we run the analysis and find non-random results, we have validated our hypothesis; we know our reasoning was correct. We have to make sure not to cheat: once we choose an event and analyze it, we must keep the result even if it is random and does not prove our point. We also test against controls, noticing that if we just choose events at random without looking for collective consciousness, the analysis yields randomness and nothing unusual.

Another approach is not to involve an analyst's choice at all, but to observe correlations between the network output and other measures of collective consciousness, such as sentiment measured from news sources. By correlating to several such measures, we gain confidence in consciousness as the source.

Although there are multiple ways to measure this behavior, typically we perform a network variance (NetVar) calculation. This measure adds together the correlated behavior occurring across devices in the network. Significant cross-correlations occur during times of shared human attention and emotionality (consciousness). Each RNG generates many random numbers every second, to which the following procedure is applied:

1. The random numbers in general follow a normal (bell curve) distribution and are standardized into Z-scores. This highlights how much each number generated by an RNG strays from what we expected from pure randomness.
2. All of the Z-scores within and across all of the devices in the network (or a subgroup) are added together via a Stouffer sum to yield the network Z-score.
3. This score is squared to provide the NetVar per second. We compare it to the chi-squared distribution, which shows us the expected behavior for purely random numbers. When the NetVar is improbably high or low, it falls outside the confidence intervals of this distribution.
4. Subtract one from the NetVar so that it is centered at zero for easier chart plotting.
5. Aggregate across the period of time during which there is coherent consciousness to smooth out random behavior while compounding effects into a statistically significant result.