Academic Bibliography

 Search 200 years of publications by Ghent University researchers.
Advanced search
2 files | 4.56 MB
Add to list
  1. Search results
  2. How much data do we need to estimate ...

How much data do we need to estimate computational models of decision-making? The COMPASS toolbox

Maud Beeckmans, Pieter Huycke (UGent) , Tom Verguts (UGent) and Pieter Verbeke (UGent)
(2024) BEHAVIOR RESEARCH METHODS.
Author
Organization
Project
Abstract
How much data are needed to obtain useful parameter estimations from a computational model? The standard approach to address this question is to carry out a goodness-of-recovery study. Here, the correlation between individual-participant true and estimated parameter values determines when a sample size is large enough. However, depending on one's research question, this approach may be suboptimal, potentially leading to sample sizes that are either too small (underpowered) or too large (overcostly or unfeasible). In this paper, we formulate a generalized concept of statistical power and use this to propose a novel approach toward determining how much data is needed to obtain useful parameter estimates from a computational model. We describe a Python-based toolbox (COMPASS) that allows one to determine how many participants are needed to fit one specific computational model, namely the Rescorla-Wagner model of learning and decision-making. Simulations revealed that a high number of trials per person (more than the number of persons) are a prerequisite for high-powered studies in this particular setting.
Keywords
Computational models, Statistical power, Toolbox

Downloads

  • Download
    (...).pdf
    • full text (Published version)
    • |
    • UGent only
    • |
    • PDF
    • |
    • 3.57 MB
  • Download
    Beeckmans COMPASS.pdf
    • full text (Accepted manuscript)
    • |
    • open access
    • |
    • PDF
    • |
    • 990.61 KB

Citation

Please use this url to cite or link to this publication:

MLA
Beeckmans, Maud, et al. “How Much Data Do We Need to Estimate Computational Models of Decision-Making? The COMPASS Toolbox.” BEHAVIOR RESEARCH METHODS, 2024, doi:10.3758/s13428-023-02165-7.
APA
Beeckmans, M., Huycke, P., Verguts, T., & Verbeke, P. (2024). How much data do we need to estimate computational models of decision-making? The COMPASS toolbox. BEHAVIOR RESEARCH METHODS. https://doi.org/10.3758/s13428-023-02165-7
Chicago author-date
Beeckmans, Maud, Pieter Huycke, Tom Verguts, and Pieter Verbeke. 2024. “How Much Data Do We Need to Estimate Computational Models of Decision-Making? The COMPASS Toolbox.” BEHAVIOR RESEARCH METHODS. https://doi.org/10.3758/s13428-023-02165-7.
Chicago author-date (all authors)
Beeckmans, Maud, Pieter Huycke, Tom Verguts, and Pieter Verbeke. 2024. “How Much Data Do We Need to Estimate Computational Models of Decision-Making? The COMPASS Toolbox.” BEHAVIOR RESEARCH METHODS. doi:10.3758/s13428-023-02165-7.
Vancouver
1.
Beeckmans M, Huycke P, Verguts T, Verbeke P. How much data do we need to estimate computational models of decision-making? The COMPASS toolbox. BEHAVIOR RESEARCH METHODS. 2024;
IEEE
[1]
M. Beeckmans, P. Huycke, T. Verguts, and P. Verbeke, “How much data do we need to estimate computational models of decision-making? The COMPASS toolbox,” BEHAVIOR RESEARCH METHODS, 2024.
@article{01H6B0N2MJYPGXV9D7GJ68PKV4,
  abstract     = {{How much data are needed to obtain useful parameter estimations from a computational model? The standard approach to address this question is to carry out a goodness-of-recovery study. Here, the correlation between individual-participant true and estimated parameter values determines when a sample size is large enough. However, depending on one's research question, this approach may be suboptimal, potentially leading to sample sizes that are either too small (underpowered) or too large (overcostly or unfeasible). In this paper, we formulate a generalized concept of statistical power and use this to propose a novel approach toward determining how much data is needed to obtain useful parameter estimates from a computational model. We describe a Python-based toolbox (COMPASS) that allows one to determine how many participants are needed to fit one specific computational model, namely the Rescorla-Wagner model of learning and decision-making. Simulations revealed that a high number of trials per person (more than the number of persons) are a prerequisite for high-powered studies in this particular setting.}},
  author       = {{Beeckmans, Maud and Huycke, Pieter and Verguts, Tom and Verbeke, Pieter}},
  issn         = {{1554-351X}},
  journal      = {{BEHAVIOR RESEARCH METHODS}},
  keywords     = {{Computational models,Statistical power,Toolbox}},
  language     = {{eng}},
  title        = {{How much data do we need to estimate computational models of decision-making? The COMPASS toolbox}},
  url          = {{http://doi.org/10.3758/s13428-023-02165-7}},
  year         = {{2024}},
}
Altmetric
View in Altmetric
Web of Science
Times cited: