This repository contains the experiment files, data, and analyses used to produce the paper, "Modal spaces as a foundation for high-level cognition" by Eli Hecht and Jonathan Phillips.
This repository is organized as follows:
Modal spaces:
The data folder contains unprocessed data downloaded from Qualtrics that was subsequently anonymized. All analyses except algorithmic clustering of responses are completed within writeUp.rmd based on these data as well as the materials and manual/nlp coding files, and the results are compiled in writeUp.html. Figures are saved to figs. The code for alogirthmic clustering of responses is in nlpCoding.ipynb and the resulting tables and plots identifying repsonses by cluster are saved within folders that reference their DBSCAN parameters. For instance /nlpCoding/pg_decision_clusters/eps0.4_samp6/plots/S13.html is a plot showing the clustering of context 13 possibility generation and decision responses at epsilon = 0.4 and MinPts = 6. To save processing time when running other analyses, the outputs from a few of the computation-intensive analyses are stored in computationOutputs, allowing us to simply read in the results instead of recalculating them each time.