This OSF page contains all the data and model files for the manuscript entitled "True and false recognition in MINERVA2: A computational implementation of fuzzy-trace theory" by Minyu Chang, Brendan Johns, and Charles Brainerd.
Explanation for the raw data files containing word vectors:
1. beagle_sparse.txt: this file includes all the BEAGLE vectors used in the current simulations
2. drm_hologram.txt: this file includes all the holographic word-form vectors used in the current simulations
3. drm_fit_rep.txt: this file includes all the optimized BEAGLE vectors used to enhance list-level fits, which was discussed in the General Discussion.
Explanations for the CSV files that contain word lists
1. R&M1995.csv: this file includes DRM lists from Roediger and McDermott (1995)
2. Gallo2002_Exp1.csv: this file includes DRM lists from Gallo and Roediger (2002; Experiment 1)
3. Gallo2002Exp2.csv: this file includes DRM lists from Gallo and Roediger (2002; Experiment 2)
4. category.csv: this file includes categorical lists used in Simulation 1.4
5. Stadler1999.csv: this file includes DRM lists from Stadler et al. (1999)
6. Stadler1999_Gallo2002_combined.csv: this file includes DRM lists from a combined pool of Stadler et al. (1999) and Gallo and Roediger (2002)
7. SL1999.csv: this file includes DRM lists from Sommers and Lewis (1999)
8. Watson2003.csv: this file includes DRM lists from Watson et al. (1999)
Explanation for model codes:
1. holoword_python3.py: this is the code we used for constructing holographic word-form vectors (Cox et al., 2011). We thank Gregory Cox for sharing this code.
2. All the MINERVA2 model scripts are labelled based on the simulations they correspond to in the manuscript. Please note that the models in Simulations 1.1-1.6 are a simplified version of the full morel where only gist representations figure in the model (which is equivalent to the situation when gist-verbatim scaling parameter was set to 1). The models in Simulations 2.1 -3.4 are the full model that features both verbatim and gist processing.
3. Simulations 1.7 and 2.3 share the same script.
BEAGLE_validation.xlsx: this file contains the raw data of list-level MBAS, MGS, and semantic similarity values calculated using BEAGLE vectors.
If you have any questions about the files, please feel free to contact Minyu Chang at minyu.chang@mcgill.ca.
