Yellow dots are nodes in the network

Preview:

DESCRIPTION

Gathering Global Awareness Are we seeing a Consciousness Field?. Yellow dots are nodes in the network. http://noosphere.princeton.edu Roger Nelson rdnelson@princeton.edu. The EGG Project (aka the Global Consciousness Project). International collaboration 75 Scientists, Artists, Friends, … - PowerPoint PPT Presentation

Citation preview

Yellow dots are nodes in the network

http://noosphere.princeton.eduRoger Nelson rdnelson@princeton.edu

Gathering Global AwarenessAre we seeing a Consciousness Field?

The EGG Project(aka the Global Consciousness

Project)

International collaboration75 Scientists, Artists, Friends, …Network of host sites world wide

The tools: REG technology, Field applicationMake an EEG for the earth, an ElectroGaiaGramIdentify engaging moments of global events

The question: Can we capture a Glimmering of Global Consciousness?

When you put a thing in order, and give it a name, and you are all in accord, it becomes.

- - From the Navajo, Masked Gods, Waters, 1950

The technology is only now availableElectronics, Computers, Networking

REG/RNG devices run continuously

Synchronized computers and software

Internet transfer of data to central server

Automatic archiving, public access

Formal analyses and explorations

Background, methods, poetic history

A Random Event Generator (REG or RNG)

Mindsong REG

Orion RNG

How it works: Here’s 1000 Trials Each trial is the sum of 200 bits

Like flipping 200 coins and counting heads

The binomial distribution of 1000 200-bit trials, compared with

Theoretical normal distribution

100 is expected

What happens in such data over time?Plot cumulative deviation from expectation

A General Case GCP Standard Analysis

Should be a Random Walk (a “Drunkard’s Walk”)

Laboratory Experiments, PEAR:Intention to change the REG behavior

High and Low both depart from expectation

HI

LO

BL

Field REG Experiments: No Intentions We simply collect data in the situationWe find departures from expectation

Queens Chamber

GrandGallery

KingsChamber

Departures from expectation correlate with Coherent or Resonant group consciousness

Deeply engaging ideas and emotions

The next step: A prototype collaborationColleagues in Europe and the US

Collected 12 independent data streams

How About a Global Network?Might we see Effects of Engaging Events?

Natural disastersTerrible accidents

The beginning of warThe Pope’s pilgrimage

Grand celebrations Political excitement

Astrological hot spotsWorld-wide meditations

A network of “eggs” around the world: A computer, special software, an REG/RNG

Internet transfer of data to Princeton Here we see the combined dataFor a whole day, from 48 eggs

We can see better what’s happening by Plotting the cumulative deviations ( df)

Correlation Tilts … Variance Spreads

Combine all eggs and plot cumulative deviation May show a trend if there is a common Influence or correlation among the eggs

( df)

When we make a composite across eggs, the cumulative deviation may show a trend

Maybe a trend in the cumulative Deviation is Meaningful?

So far, sporting events seem Not to produce big deviations

The Pope’s 6-day pilgrimage to the middle eastAn occasion of hope for resolution of differences

Political events, even big ones, are not necessarily of interest to the EGG

Clinton Impeachment Acquittal

We’ll try anything once. Significant correlations with astrologically determined “hot” times

Tearing the Social Fabric

Terrorists Attack Civilians and Diplomats

Nato Bombs Kosovo to End Ethnic Cleansing

Taliban Destroy Ancient Buddhist Treasure

September 11 Enters History of the Earth

Major disasters that engage us powerfullyOften correlate with big deviations

Context explorations: Six hours of dataAround the beginning of bombing in Kosovo

Cumulative deviation of Zs2 or 2

The Buddhas of Bamiyan

Before March 12 2001

After

The destruction of a world treasureApproximate time, noon in Afghanistan

noon

The World Trade Center

Fall, 2001

Carol M. Highsmith Al Dove

The World Trade Center

September 11 2001

Tamara Beckwith

The destruction of the World Trade TowersSept 11 2001

A 50-hour trend followed the attacks

Sept 11 prediction: Inter-egg Variance will fluctuate

Red is real data, Green is Pseudorandom

Autocorrelation on Sept 11Structure where there should be none

Calculations: Peter Bancel

What is there to say?

Repairing, Restoring The Social Fabric

HealingMeditationsPrayer VigilsCeremonies

Holidays

Chicago, Moment of Silence September 14 2001

Three Minutes of Silence September 14 2001

Focused Siddhi Meditations September 23 to 27, 2001

Peak Day, Maximum Numbers

Synchronized Johrei Ceremonies Around the World, April 1 2001

Four Million People Pilgrimage For Ritual and Prayer

A Substantial Agreement on Getting Our Ecological Act Together

Natural DisastersAnd Accidents

Earthquakes, FloodsPlanes, Trains, Ships

The Great Earthquake in Turkey August 17 1999

Variance Analysis, Bierman

Standard Analysis

US vs Europe, Bierman

Volcano in Nairobi January 17 2002

Typhoon in Orissa, India October 29 1999

The Concorde Crash July 25 2000

The Kursk Submarine Disaster August 12 2000

Train Crash in India August 2 2001

All Eggs, 4 hrs

Nearby Eggs, 1.5 hrs

The Lighter Side

Celebrations: What better example than

New Years?

Global Attention: Sharing New Year’s Eve

All over the world, people celebrate the change to a New Year. Since 1998, we have recorded data from a network of physical random event generators (called “eggs”). Here we look at what happens at midnight around the world.

The scientific prediction is that there will be a pattern of increased correlation among the eggs. We test for trends away from the expected “random walk”. We have learned to expect reductions in the variation across the eggs.

The figures on this page almost speak for themselves. They are pictures of our engagement with each other.

Variance Drop, Midnight, 1999-2000

In the first year, 1998-1999, we looked for a change in the average deviation, and compared Maxi- and Mini-celebration time zones.

Then, for the infamous Y2K transition, we looked at a measure of the variability among the eggs and predicted it would decrease as we all focused on midnight.

We predicted the same pattern for the following year.

And again for last year. The pattern is replicated for the third time.

More at http://noosphere.princeton.edu

An obvious prediction: New Years celebrations Concatenation across all (24) time zonesCumulative excess deviation of means

Weak Replication

Model Prediction

Taking the Electrogaiagram metaphor to heartMaxi-celebrating time zones, Signal averaging

Figure by Richard Broughton

Evoked Response in the Electrogaiagram:Signal averaging: Mini-Celebrating time zones

Figure by Richard Broughton

Y2K New Year 1999-2000: Coherent engagement? Radin makes an independent prediction

Reduction of Variance across eggs

Odds, GMT

New Years 2000-2001: Variance Reduction Signal Average over 37 time zones Normalized, Squared, Smoothed

New Years 2001-2002: Variance Reduction Signal Average over 37 time zones Normalized as Z-scores, Smoothed

5-Min Smoothing Window

New Years 2002-2003: Variance ReductionShows ambiguous response,

But Meanshift is significant again

Questions outnumber answersGreat numbers contribute

Deep engagement is powerfulbut

Does distance from the focus matter?How about relevance to local people?Is human consciousness necessary?

Are “experimenter effects” the source? What kinds of events are “strongest”?Is the effect repeatable and reliable?

Columbia Explodes: Seven astronauts lost But the eggs apparently do not respond

Why do we sometimes see no effect?

Perhaps we do, But not in the form we expect

The Columbia Explosion:Viewing through a different instrument

Note, however, about 1/3 of permutationsHave such an extreme deviation sometime in 24 hours

Categories: What seems to touch The Global Consciousness?

Calculations, Dean Radin

Does the distance of the eggsFrom the event make a difference?

September 11 2001Effect Size as a function of Distance

0

0.5

1

1.5

2

2.5

3

NorthAmerica

SouthAmerica

Europe Australia Asia

Increasing Distance

Z-s

core

Calculations, Dean Radin

Does the distance of the eggsFrom the event make a difference?

Effect vs Dist, Stouffer Z, Full Day, Sept 11

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

1.2

1.4

USEastern

US Middle

USWestern

England,Scandin

WesternEurope

Israel,India,Brazil

Africa,Austral,

New ZealIncreasing Distance from NYC

Effe

ct S

ize

(r)

Current Result: Formal Database132 global events over 4.5 years

Odds: about 100,000 to 1

What do we have in hand?Where do we want to go with it?

Four years of data

55 eggs around the world

More than 130 formal studies

About 65% positive outcome

About 18% individually significant

Many analyses remain to be done

Bigger Picture: What is our aspiration?

Sharpen and focus our questions

Aim for theoretical understanding

Capture insight about creative mindConsider evidence that we are one

Contribute to better future for culture

We think the world apart. What would it be like to think the world together?

-- Parker Palmer, educator

http://noosphere.princeton.edu

Homepage

StatusDay Sum ResultsExtract

Magic Buttons

Primary Links Menu at Bottom

http://noosphere.princeton.edu

Berger: Web Design

Barry Fenn 1950 – 2000

A memorial assessmentof GCP data

Perhaps the eggs

Respond to their hosts

A respectful, exploratory look at the death of egg host Barry Fenn

All 28 eggs

Same time period, Barry’s egg only.After he passes, a 20 hour falling trend

Symmetry in the data, maybe even meaning: Barry’s egg and that of his friend, Sze Tan

A total solar eclipse was of Special interest to people in the path

For most of the formal predictionsWe specify a “Standard Analysis”

Normalized signed deviation of mean, zi = (mi-)/ Composite across eggs: Stouffer Zs = (zi)/N1/2

Composite Z is squared for 2 distributed statistic

Large cumulative sum of Zs2 – 1 or 2 – df

Reflects inter-egg correlation, or Consistent large deviations, or both

Global Emotion: Transfixed by Tragedy

On September 11 2001, early in the morning, a network of physical random event generators (called “eggs”) took on a striking trend. By 8:45 the non-random behavior was unmistakable. It peaked at about 10:30 with odds against chance of a thousand to one. See the red trace below.

Other measures also deviated from expectation on that day, creating an unmistakable pattern where there should be none. The eggs became linked across distance and time in some subtle way that we do not yet know how to explain.

This is not a physical or electromagnetic effect. It’s not due to extraordinary mobile phone use, or saturation TV.It appears to be related to our profound engagement.

On 9/11 deviations began that persisted for 2 days

On 9/11 the data showed extraordinary moments

On 9/11 the data contained unique sequential structure

The jagged red line shows three days of a measure (squared cumulative deviation of variance) that represents the composite randomness of 37 eggs.

On September 11, the data show clear structure where there should be none.

am

am

am

More at http://noosphere.princeton.edu

…………………………………………………………………………………………………………………..…….…………………. Normal range of variation

Squared Cumulative Variance, Sept 10 - 12The departure begins nearly 4 hours before the first attack

Bottom Line Comparison:Formal outcome sequence versus Random draws from distribution

Recommended