For two years now, The Association for the Study of Dreams (ASD) has hosted a Psiber Dreaming Online Conference where dreamers participated in Mutual Dreaming, precognitive dreaming and telepathy in dreams. Each year, there was a Mutual Dreaming contest as well as other dreaming contests. This year, a number of participants were curious about how many similar elements they saw cropping up in the dreams of that night. Two questions arose.

I then decided to explore those questions to see what I could find. Here is a copy of the proposal I put to the dreamers at last years (2003) Psiber Dreaming Online Conference.

"The purpose of this research project (Intentionally As An Important Factor In Mutual Dreaming Similarities) is to attempt to show that having the intention of dreaming mutually will cause more frequent similarities in dream elements among the mutual dreamers.

The process will be as follows:

A computer program called DreamSpinner will be used to input all the dreams submitted to this project. This program uses the words in the narrative (dream report) to create frequency lists of all elements in the dream report. One data processor, Bjo Ashwill, will maintain reliable consistency in her method of linking each word in the narrative to the appropriate categories in the program. (An expanded version of the Hall/van de Castle content analysis categories.)

Three different groupings of dreams will be kept in separate databases in DreamSpinner to be able to compare these different groupings with each other.

First grouping will be dreams that occurred during a mutual dream contest with full intentionally in place (Dreams from 2 different contests 2002 and 2003 will be used, each in a separate database)

Second grouping will be dreams from those same mutual dreamers but from dates chosen randomly in the past so no mutuality intention was present. (Assuming no precog time warp!)

Third is a control group of dreamers who do not necessarily consider themselves Psi dreamers who will present dreams from a randomly chosen night or nights.

I would like three things from each of you willing to participate.

1. The Dreams:
The mutual dreams for 2003 you entered into the contest.

The mutual dreams you entered into last years contest (for those of you who did last years Mutual Dreaming contest.)

Your dreams for the night of July 15, 2003 and/or the night of April 4, 2003.

(Although I am not requesting you fill out Ed’s questionnaire, I am asking you to reexamine your dream report with his questionnaire in mind. Fill in all those missing details we take for granted. The fact that I don’t know you personally and I will be inputting the information into a computer program means that only the details you write in the report will be examined. Do not assume I will know the shape of your table (square, round, oval etc.) Label the characters with age (adult, child, baby or teen), sex (male or female or blended), numbers of them (group or individual), and identity (relation? Friend? Husband or wife or child? Occupational, Famous or prominent?) Colors, shapes, sounds, feelings need to have descriptive words in the dream report for it to be counted in the DreamSpinner software. More is better."

168 dreams total.

83 dreams were from the participants of the Mutual Dreaming Contest night 9/25/03.

85 dreams were from the two control groups. (53 from non Psiber dreamers) and (32 from Psiber dreamers on nights other than 9/25/03.

ZIGGLE is code for a female participant of the Psiber Dreaming Conference.

ZIGGLEM male Psiber dreaming conference participants.

ZAGGLE is the code word for female control group participants.

ZAGGLEM for male control group dreamers.

Unless otherwise noted, Ziggle stands for all Psi dreamers and Zaggle for all non Psi dreamers. When the groups are separated out according to male and female, it is specified.

These dreams were entered into DreamSpinner, a content analysis computer program. Each dream then was subjected to a similar linking process.

1. Words that were automatically NOT counted were: a, an, and, the, to, who’ve. These are the most frequently used words in narratives and do not add significant content. This first step chopped the length of the dream narratives. For example: one dreamer had 1808 words in 5 dreams. This was chopped down to 1072 Hits or 59.20% of the narrative was linked. Another dreamer had 1158 words in 5 dreams which was chopped down to 813 Hits. This is 70% of the narrative that was linked. This simply indicates different writing styles.

2. Pronouns, prepositions, helping verbs, negation words (no, not) were not counted. (The intent of the negations words were carried in the word it negated. "I was not happy" was linked to the category "Unhappy".

3. Some words had multiple category links. All character names were linked four times. One in the Character’s Age Category, one in the Character’s Numbers category, One in the Character’s Sex category and one in the Character’s Identity category. Other examples of multiple category linking are words like "box". It is linked to the Object category, and to the Object Relationship category as a Container.

4. Root words are recognized in their different forms. For example, the root word "slow" has the following ending possibilities: slows, slowed, slowing, slower, slowest, slowly, slowness. Any word with these different endings are linked to the root word "slow". In addition, there are different definitions:

Slow1=A motion that is not fast. It takes more time to do.

Slow2=Not slowing down. The velocity continues.

Slow3=Mentally slow, not catching on quickly.

Slow4=Slow down, reduce velocity in a vehicle.

Slow5=To walk less quickly. To slow down.

The person inputting the dream into DreamSpinner chooses which definition is correct within the context of the dream being linked.

5. DreamSpinner only reports one incident of a word per dream. So if a dreamer repeats a word, it will not effect the total count. For example: A dream report might say "I saw the box. The box was square and solid. The man picked up the box. Another man took that box from the first man." Several situations arise from this way of counting. It is necessary to use different definitions of root words to denote more than one when it is important. For example it is important to have an accurate count of characters. So the first mentioned man would be linked under man1 and the second man would be linked under man2. It is not important to note how many times the word is mentioned when it has the same definition. So the word box1 would be linked just once per dream. Now some dreamers had 5 or more dreams counted in the samples. If a character name was used in more than one dream, it would show up. For example someone might dream of a family member in 5 different dreams. So that family member count would be 5. We have no way of knowing (except of course to manually find those dreams and see whether one or more dreamers drempt about that character.) So some frequency counts can be influenced by one dreamer.

6. A word about dream narrative length. The more words in a dream, the more possible similarities will be found. This research project chose to use all the dreams regardless of length. A more rigorous protocol would be necessary to completely compensate for the dream narrative lengths. (Such as only examining dreams between 50 and 300 words in length. This unfortunately would remove the Designated Dreamer as well as some other meticulous and detailed reporters of their dreams. Keeping this in mind, I have made some attempts to equalize the dream length problem.

a. Stripping out the many extra words as described above left only meaningful words to be linked.

b. Words that were repeated were only counted once.

c. The HP method of comparing different dream sets helped equalize the size issue. The HP factor is found by comparing the number of hits for a word or sub category in one dream set with the number of total hits in a category. Then the percentages that created are compared with the same ratio of hits to total hits in a category in the second dream set. For example: