Identifying and modelling cognitive biases in mobility choices

Chloe Conrad,Carole Adam

2024-02-15

Abstract:This report presents results from an M1 internship dedicated to agent-based modelling and simulation of daily mobility choices. This simulation is intended to be realistic enough to serve as a basis for a serious game about the mobility transition. In order to ensure this level of realism, we conducted a survey to measure if real mobility choices are made rationally, or how biased they are. Results analysed here show that various biases could play a role in decisions. We then propose an implementation in a GAMA agent-based simulation.

Computers and Society,Artificial Intelligence,Multiagent Systems

What problem does this paper attempt to address?

The paper attempts to address the issue of cognitive biases in daily transportation choices. Specifically, the researchers aim to identify and understand how these cognitive biases influence people's transportation choice behavior through surveys and modeling.

### Background and Problem Description
1. **Sustainable Transportation Transition**: The paper points out that transitioning to more sustainable transportation modes is one of the keys to building an ecological society. However, merely relying on public policies to encourage the use of low-carbon transportation modes (such as cycling, walking, etc.) is not sufficient to achieve this goal.
2. **Importance of Educational Tools**: In addition to policy measures, developing educational tools (such as serious games) is also an effective way to promote transportation transition. Through game mechanisms, players can be prompted to reflect on their daily transportation choices.
3. **Limitations of Existing Models**: Existing transportation choice models are mainly based on rational decision-making processes, ignoring the influence of habits and cognitive biases. Therefore, these models are often not accurate enough in practical applications.

### Research Objectives
1. **Identify Cognitive Biases**: Identify the cognitive biases that play a role in daily transportation choices through a questionnaire survey.
2. **Model Cognitive Biases**: Incorporate the identified cognitive biases into an agent-based model to improve the model's realism and predictive ability.
3. **Validate Model Effectiveness**: Experimentally verify whether the model with cognitive biases can better simulate people's actual transportation choice behavior.

### Main Methods
1. **Questionnaire Survey**: Designed and implemented an online questionnaire survey, collecting transportation choice preferences and perception data from 625 participants.
2. **Data Analysis**: Analyzed the questionnaire data to identify major cognitive biases, including confirmation bias, reactance bias, and overestimation bias.
3. **Agent-Based Model**: Developed an agent-based model using the GAMA platform to simulate individual transportation choice behavior under different environmental conditions and introduced the aforementioned cognitive biases.

### Experimental Results
1. **Simulation Without Cognitive Biases**: Without considering cognitive biases, the model's predictions showed significant differences from the actual survey data, especially in the usage proportions of public transportation and bicycles.
2. **Simulation With Cognitive Biases**: After introducing cognitive biases, the model's predictions were closer to the actual survey data, particularly in correctly reflecting the order of transportation mode usage.

### Conclusion
1. **Impact of Cognitive Biases**: The study shows that cognitive biases play an important role in people's daily transportation choices, especially in explaining irrational choices.
2. **Directions for Model Improvement**: To further improve the model's accuracy, more in-depth research and precise modeling of cognitive biases are needed, while also considering the impact of social interactions and social biases.

In summary, by identifying and modeling cognitive biases, this paper enhances the realism and predictive ability of transportation choice models, providing a theoretical foundation for developing more effective transportation transition tools.