English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Computational models are powerful tools for understanding human cognition andbehavior. They let us express our theories clearly and precisely, and offerpredictions that can be subtle and often counter-intuitive. However, this samerichness and ability to surprise means our scientific intuitions andtraditional tools are ill-suited to designing experiments to test and comparethese models. To avoid these pitfalls and realize the full potential ofcomputational modeling, we require tools to design experiments that provideclear answers about what models explain human behavior and the auxiliaryassumptions those models must make. Bayesian optimal experimental design (BOED)formalizes the search for optimal experimental designs by identifyingexperiments that are expected to yield informative data. In this work, weprovide a tutorial on leveraging recent advances in BOED and machine learningto find optimal experiments for any kind of model that we can simulate datafrom, and show how by-products of this procedure allow for quick andstraightforward evaluation of models and their parameters against realexperimental data. As a case study, we consider theories of how people balanceexploration and exploitation in multi-armed bandit decision-making tasks. Wevalidate the presented approach using simulations and a real-world experiment.As compared to experimental designs commonly used in the literature, we showthat our optimal designs more efficiently determine which of a set of modelsbest account for individual human behavior, and more efficiently characterizebehavior given a preferred model. At the same time, formalizing a scientificquestion such that it can be adequately addressed with BOED can be challengingand we discuss several potential caveats and pitfalls that practitioners shouldbe aware of. We provide code and tutorial notebooks to replicate all analyses.

Designing Optimal Behavioral Experiments Using Machine Learning