Mapping the landscape of behavioral reinforcement learning research
Year: 2025
Type: preprint
Abstract: As global research output increases, maintaining a comprehensive overview becomes challenging, particularly in high-volume research fields spanning multiple disciplines and research traditions. Behavioral reinforcement learning---an influential approach to understanding how people learn from interactions with the environment---is one such field. What characterizes this research landscape, and how is it interconnected? Here, we introduce a novel bibliometric approach---combining computational semantics, large language models, and clustering methods---to explore article clusters within behavioral reinforcement learning. Our analysis provides a comprehensive map of the field, highlighting broad lines of behavioral and neuroscientific research and documenting a wide array of interdisciplinary topics. We characterize research clusters by frequent research topics and methods, key journals, and publication timeline. Moreover, we examine the relationships between clusters and visualize the distribution of research topics and methods across the landscape. Finally, we discuss implications for facilitating scientific exchange and present an online tool for exploring the landscape.
Authors: Anna Isabel Thoma, Florian Bolenz, Kevin Tiede, Yujia Yang, Stefano Palminteri, Ralph Hertwig, Dirk U. Wulff
Language: English
Subfield: Statistics, Probability and Uncertainty
Field: Decision Sciences
Domain: Social Sciences
Open Access status: gold
OpenAlex ID: W4415252245