Page 273 -

272 O. Barreteau et al.

role-playing game. A switch to the following category occurs when this explanation
leads to a debate that makes the model open to modifications. Otherwise, it remains
mere information.

12.4.2.2 Consultation and No Control

A R

P M

Participants are consulted about the model’s content and its simulation that is by
the researchers, who are the only designers. They provide information and solicit
comments on the model. Mere data collection through a survey does not fall in
this category because it assumes active involvement from participants in providing
information to the modellers. Some knowledge elicitation techniques, such as BBN
design, tend to fall mostly in this category. Translation of the inputs originating from
participants into pieces of a model is performed only by researchers. This translation
is not necessary transparent. No control over use or dissemination of the model is
deputed to participants as such. Compared to previous category, participants have
the ability to frame marginally more of what is performed by the model through
their inputs to the model’s content However, the extent of this ability depends on
the participants’ skills to identify potential uses of a model. As in any participatory
process, when there is an unbalanced power relation between parties, the process is
also a way for policymakers to gain information from stakeholders, information that
could be used for strategic purposes. This bias can be alleviated if the involvement
of A includes all members of A, including the subset P. The constructed model in
this case may be used by the members of P.

12.4.2.3 Dialogue with Modellers and No Control

A R

P M

In this category, iterative and genuinely interactive processes between stake-
holders and modellers start to appear. There is still a translation of inputs from