Group model building and mental model change

Abstract

This thesis explores the individual and group processes that occur in qualitative group model building, mostly using applied research in New Zealand government departments.System dynamics is an approach to understanding the behaviour of systems over time. Originally, system dynamics practitioners would advise corporate managers on business decisions. To improve the adoption of these proposed interventions, some practitioners experimented with involving client groups in the modelling process, now commonly referred to as “group model building”. Researchers not only discovered that group model building supports client commitment to implementing the conclusions of the modelling process, but also noted a range of intrapersonal and interpersonal changes in participants. However, group model building literature remains unclear on what these changes are, and how they occur.In Chapter 2, a literature review explores what intrapersonal and interpersonal factors predict effective strategy implementation. An observational study was conducted, and suggests that group model building workshops can have a positive impact on the intention to implement strategy. Several cognitive biases were proposed to explain this effect.Group model building literature claims mental model change and mental model alignment as outcomes, but it is not clear that the evidence supports these claims. Mental models are relatively enduring structures, and previous studies only evaluate changes observed immediately after the group model building workshop. Chapter 3 follows participants in four group model building workshops over a 12-month period. The results demonstrate that participants change their decision preferences through the workshop, and that without reinforcement these changes persist for 12 months, suggesting mental model change. Participants’ preferences also become and remain more alike, suggesting enduring mental model alignment. Individuals whose mental models are more aligned are more likely to agree, and this can be an important outcome where decision groups are expected to build shared understandings over time. The evaluations also reveal that participants are both persuaded by their peers and learn new insights from the modelling process. The new insights were much more easily retained than decision preferences from persuasion.This study did not rely on participants self-reporting their learning processes. A comparison of the results of this study and each participant’s questionnaire results (from Chapter 2) suggested that, in this case, questionnaire results do correlate with individual level outcomes. This finding increases the validity of many previous studies that use the same questionnaire tool. Having demonstrated that mental model change (and mental model alignment) occurred, this thesis turns its attention to how these changes happen. Several mechanisms have been proposed in the literature over the past 20 years, but none of these have been compared against each other. This deficiency is addressed in Chapter 4. Participants in group model building sessions were interviewed, to explore what they felt the workshop achieved and how these effects occurred. Participants’ experience of group model building workshops was most consistent with the Boundary Object mechanism, which suggests that tangible, modifiable objects that represent dependencies help to support participant trust and agreement. The Boundary Object mechanism provides important theory for refining group model building practice and informing where group model building may be most applicable.Finally, the thesis explores which (if any) of these findings matter to potential clients. Public servants conduct group-decision processes with increasing frequency. Chapter 5 describes interviews and survey questionnaires with public servants from several New Zealand government departments, who frequently commission or conduct group decision processes. Questionnaire results show that public servants value commitment to conclusions, consensus, communication quality and enduring mental model change. Interview results show that in some circumstances, shared understanding (mental model alignment) can be critical.Group model building workshops are believed to affect cognitive processes and behaviour of participants. There is currently a lack of understanding of the nature of these changes, their persistence, their importance to clients, and their causativ mechanisms. This thesis contributes to group model building knowledge by improving our understanding of each of these factors. However, this thesis also reveals new research gaps, and describes additional research that would advance our understanding further.