Data and analysis code for the Masters thesis titled "Dyadic Physical Activity Planning with a Virtual Coach: Using Reinforcement Learning to Select Persuasive Strategies"
doi: 10.4121/2796f502-0610-4a7d-a8ee-ebc36639e0b1
CC BY 4.0
This is the data and analysis code for the Masters thesis titled "Dyadic Physical Activity Planning with a Virtual Coach: Using Reinforcement Learning to Select Persuasive Strategies." In this thesis, we used data gathered through an observational study with a virtual coach to build a reinforcement learning model for persuading people to commit to plans for walking. Once the model was created, it was analysed using the code provided in this dataset, including simulations of how the model would perform in a real-life setting.
Study
The study was conducted on the online crowdsourcing platform Prolific in June and July 2023. The Human Research Ethics Committee of Delft University of Technology granted ethical approval for the research (Letter of Approval number: 3089). 114 participants participated in a conversation with the virtual coach named Jamie. During these conversations, the virtual coach measured people's confidence in following the plan, their perceived usefulness of planning, and their attitude towards planning. Then, the virtual coach used a persuasive strategy to try to change these aspects of a person's state, after which it measured the three aspects again. This repeated until at least two persuasive strategies were used, and until the three aspects were high enough to satisfy pre-set heuristics for when the person is likely to commit to the plan. When these conditions were met, the virtual coach measured the reward signal, composed of satisfaction with the conversation, commitment to the fist two weeks and to the whole plan, and confidence in reaching the goal.
Data
We provide data on:
- participant characteristics (e.g., age, gender, stage of change for becoming physically active),
- the state-action-next state-reward samples used to create the reinforcement learning model,
- free-text responses to the persuasive strategies used by the virtual coach,
- free-text responses about what people found motivating and demotivating in motivational messages.
The file OSF pre-registration explains in detail how each variable was measured.
Analysis
We provide code to reproduce our analysis with Docker. For more information on this, please refer to the README-file in this repository.
In the case of questions, please contact Nele Albers ([email protected]) or Willem-Paul Brinkman ([email protected]).
- 2023-11-16 first online, published, posted
- Part of Perfect Fit, funded by NWO, project number 628.011.211
DATA
- 5,119,635 bytesMD5:
ba77fd44a31de329e06e652ed081196fAnalysis_code_and_data_for_Dyadic_Physical_Activity_Planning_with_a_Virtual_Coach_Using_Reinforcement_Learning_to_Select_Persuasive_Strategies.zip -
download all files (zip)
5,119,635 bytes unzipped
