Managing a Chocolate Factory and Controlling Forest Fires: How to Learn to Make Smarter Decisions

Managing a Chocolate Factory and Controlling Forest Fires: How to Learn to Make Smarter Decisions poster

Audio Presentation:

Audio Transcript

Research Authorship:

Yoannis Hermida, C. Dominik Güss

Faculty Mentor:

Dr. C. Dominik Güss | College of Arts and Sciences | Department of Psychology


Within much of society, people have to make decisions under time pressure and in unpredictable, changing situations. The current study explores the effects that training has on improving DDM in two computer simulation tasks with different characteristics. A total of 100 undergraduate students participated in the current study. Participants either managed a computer-simulated chocolate factory—ChocoFine (Dörner, 2000) or were a fire rescue chief controlling a forest fire, WinFire (Schaub, 2017). ChocoFine lasted 1-hour involving high complexity and lower time pressure. WinFire lasted 15 minutes with lower complexity and higher time pressure. We employed a between-subjects design comprised of an experimental group with training and a control without training. Participants were later given a list of common errors and instructed to identify which errors they committed throughout the simulation. We hypothesized that training in DDM errors would lead to better performance in both simulations. Furthermore, we hypothesized that participants in the slower-paced, more complex simulation, ChocoFine, will exhibit different errors compared to participants in the fast-paced, less complex simulation, WinFire. Preliminary results showed marginally significant differences in performance between experimental and control groups, with experimental groups performing better while also reporting fewer errors than their respective control group. Participants working on ChocoFine reported more errors regarding information gathering; and planning, decision making, and action compared to participants in WinFire. The implications of this study are critical for organizations encountering dynamic and uncertain situations, as these organizations could utilize DDM training, focusing on human error and self-reflection for future personnel trainings.


My name is Yoannis Hermida.

I am a first-year graduate student in the psychology department at the University of North Florida.

I will be showcasing the work I have done thus far pertaining to my graduate thesis.

My poster is titled, “Managing a Chocolate Factory and Controlling Forest Fires: How to Learn to Make Smarter Decisions”.

My thesis project is about dynamic decision-making, abbreviated as DDM, and human error training across two computer simulated tasks, WinFire and ChocoFine.

DDM refers to novel and ever-changing problem situations, where no specific protocol is  known to reach a solution.

In this study we assess DDM performance via the two computer simulations.

WinFire is a simulation where participants’ objective is to take the role of a fire chief and oversee a current fire and salvage the town as well as the forest.

This simulation requires quick action and is less complex in nature, with little room for self-reflection.

On the other hand, ChocoFine, is another simulation where participants take the role of a chocolate factory manager in charge of production and marketing chocolate products.

This simulation is longer, more complex and therefore, participants have more time for self-reflection.

Participants were given an error sheet either before or after the completion of the simulation to identify any errors they committed throughout the simulation.

The error training sheet provided to participants included the 6 steps involved in the dynamic decision making process where participants circled specific errors they thought to have committed.

The first step of the dynamic decision-making process is problem identification which includes two possible errors: inaccurate perception and denial of reality and methodism.

The second step is goal definition which includes two errors: no goals or secondary goals determine actions.

The third step is information gathering which highlights three possible errors: entrenchment, misinterpretation of information, and conflicting information in teams.

The fourth step is elaboration and prediction which focuses on three errors: oversimplification and overgeneralization of conclusions, not considering side effects and future developments, or not considering time developments.

The fifth step in the dynamic decision-making process is planning, decision making, and action which include 5 human errors: ego-enhancing decisions, avoidance of making decisions, over planning and horizontal flight, under planning and actionism, and single focus strategy and lack of holistic strategy.

The final step in the dynamic decision-making process is evaluation of outcome and self-reflection which include two errors: no monitoring and self-reflection and ballistic action

After participants completed the simulation and circled their errors, they were then given a demographic questionnaire to complete.

The current study is a between-subjects design with participants randomly selected to complete either the WinFire or ChocoFine simulation and assigned either to the control group, with no error training conducted before the simulation, or in the experimental group, with error training conducted before the start of the simulation.

A total of 100 participants participated in the current study, 53 included males and 47 were females with an age range between 18 and 41 years old.

Preliminary results revealed a marginal significance in differences between the experimental and control groups.

Those who received training before the start of the simulation performed better than those participants in the control group, with no training beforehand.

Results also revealed that participants both in the WinFire and ChocoFine simulation reported less errors in the experimental group than their respective control groups.

Specifically, we found that participants in the ChocoFine simulation reported more errors in the DDM process of information gathering; and planning decision-making and action compared to participants in the WinFire simulation.

The results of the current study can translate into IO psychology specifically, for managerial personnel who often times experience dynamic and complex work environments.

Encouraging the use of self-regulatory decision-making training for business organizations can have positive effects on the efficiency and quality of their performance.