Talking about the likelihood of risks: an agent-based simulation of discussion processes in risk workshops

Purpose This paper aims to explore drivers of the effectiveness risk assessments in workshops. Design/methodology/approach study uses an agent-based model simulate Combining notions transactive memory and ideal speech situation, this establishes a assessment benchmark then investigates real-world deviations from benchmark. Specifically, models limits information transfer, incomplete discussions potentially detrimental group characteristics, as well interaction patterns. Findings First, transfer among workshop participants can prevent correct consensus. Second, increasing required number stable discussion rounds before improves high but not low likelihood risks. Third, while theoretically advantageous characteristics are associated with highest correctness for all risks, Fourth, prioritizing who particularly concerned about leads level correctness. Originality/value shows that by duration simulated workshops, change – rule underestimating overestimating unraveling trade-off facilitators. Methodologically, approach overcomes limitations prior research, specifically lack process benchmark, inability disentangle multiple effects difficulty capturing individual cognitive processes.