Some relevant history to begin with…
The first ParEvo type exercise that I ever carried out was in a school classroom in Wales, circa 1996. Ten of the students, in a class of about twenty, were each given a filing card. They were then asked to write a paragraph describing what happened next, following the events described in a card that had already been placed on the blackboard in front of them. That card described how one of the students in the class had left school at the end of the year. Once the selected students had written their paragraph of text on a card they were asked to go to the blackboard and place their card in a column to the right of the paragraph that was already there (the seed). After doing that they were asked to read all the other paragraphs that there other co-participants had placed up there as well. They were then each given a new blank card and asked to write a second paragraph of text which would continue one of the ten storylines now present on the blackboard. Again, they were asked to place their card on the blackboard next to and to the right of the card whose story they wanted to continue. Again they were asked then to read all the new cards that had been placed there and appreciate how they had connected to the original seed paragraph. This process continued through a number of iterations until the available time for that class ran out. Here is a diagram showing how the storylines developed.

Some advice on running a face-to-face ParEvo exercise
One weakness of the Wales exercise was that participants could get a fair idea of who had contributed what paragraph of text and whose existing text they had added onto, just by watching who placed a new card of text next to an existing one. However one of the strengths of the ParEvo web application is that it enables participants to make contributions quite anonymously. When viewing the accumulating storylines online it is not possible for an observer or a participant to identify who made which contribution and whose existing contribution they added onto. However, it is possible for the facilitator to download this kind of data after the completion of an exercise and use it to analyse the social structure of people’s participation.
It is possible to make some small changes to the process used in Wales to ensure anonymity. Firstly, the participants should not directly place their new contributions to the developing tree structure. Instead, they should hand their new contributions to the facilitator, with two kinds of instructions on the back of the note/file card: (a) saying which existing storyline this card should be added to, (b) giving their name as the writer of this contribution. Once all participants have done this, the facilitator should then place the cards at the end of the various emerging storylines, as instructed by their authors.
Face-to-face ParEvo exercises may have an advantage that is not present in the online version. After each iteration has been completed i.e. after all the participants have added their current contribution to the existing storylines, it may be feasible to have an open discussion about the pros and cons of the existing storylines e.g. in terms of their probability, desirability, or other attributes. This option is available in the online version of ParEvo through the comment facility, but that does not involve the kind of back-and-forth discussion possible in a face-to-face situation.
Postscript: The 1996 exercise has been separately written up in more detail here: