Code for the Naive Generalization Model

Caplan S (2025) Word learning as category formation. PLOS ONE 20(7): e0327615. https://doi.org/10.1371/journal.pone.0327615

Please contact me with any questions: scaplan@gc.cuny.edu

---

Setup

• Java code is written in verion 8 (OpenJDK 1.8) with no external libraries.
• Python code is written in version 2.7 with no external libraries.
• R script have been tested on version 3.3.3.
• The following R packages are required to evaluate regressions and create plots: dscales, ggplot2, Hmisc, lme4, plyr, dplyr, cowplot, tidyr, forcats, lmerTest.

No other setup is required.

Running

The following script runs all simulations and analysis:

$ runall.sh

Resources

Human experimental data used for comparison from:

• Lewis, M. L., & Frank, M. C. (2018). Still suspicious: the suspicious-coincidence effect revisited. Psychological science, 29(12), 2039-2047.
• Spencer, J. P., Perone, S., Smith, L. B., & Samuelson, L. K. (2011). Learning words in space and time: Probing the mechanisms behind the suspicious-coincidence effect. Psychological science, 22(8), 1049-1057.