Raw and post-processed data for the microscopic investigation of the effect of random envelope fluctuations on phoneme-in-noise perception

The current dataset consists of three main folders:

• 01-Stimuli/: Contains the three sets of noises (white noise, bump noise, MPS noise) for the 12 study participants (S01 to S12).
• 02-Raw-data/fastACI/: Contains the raw data as obtained for each participant, which are also available within the GitHub repository of the fastACI toolbox, using the same directory tree. The results for each (anonymised) participant (under: publ_osses2022b/data_SXX/1-experimental_results/) include their audiometric thresholds (folder: audiometry), the results for the Intellitest speech test (folder: intellitest), and for the phoneme-in-noise test /aba/-/ada/ for the three noises (savegame files in MAT format).
• 02-Raw-data/ACI_sim/: Contains the raw data as obtained for the artificial listener, i.e., the model osses2022a.m (available within the fastACI toolbox). Twelve sets of simulations (using the waveforms of participants S01 to S12) were run for the three types of test noises. The results of the simulations of the phoneme-in-noise test are stored in the savegame MAT files. The template derived from 100 repetitions of /aba/ and /aba/ at an SNR=-6 dB in white noise is also included (template-osses2022a-speechACI_Logatome-abda-S43M-trial-1-v1-white-2022-7-15-N-0100.mat). The same template was used in all simulations.
• 03-Post-proc-data/ACI_exp/: Auditory classification images (ACIs) derived from the participants' data (folder: ACI_exp) and from the simulations (folder: ACI_sim). For each participant (or artificial listener) there are three ACIs (MAT files) for each of the corresponding noises. Cross predictions are also included with performance predictions across 'participants' (Crosspred.mat, 12 cross predictions for each noise) or across 'noises' (Crosspred-noise.mat, 3 cross predictions for each participant). The cross predictions all have the same names but are stored in dedicated directories.

Use these data:

1. Download all these data, place them in a local directory of your computer. If you have MATLAB and you downloaded a local copy of the fastACI toolbox (open access at: GitHub) you can recreate the figures of our paper.
2. After initialising the toolbox (type 'startup_fastACI;', without quotation marks in MATLAB) and then type either of the following commands, to recreate the figure you want. To recreate the figures in the main text:

publ_osses2022b_JASA_figs('fig1','zenodo');    
publ_osses2022b_JASA_figs('fig2a','zenodo');
publ_osses2022b_JASA_figs('fig2b','zenodo');
publ_osses2022b_JASA_figs('fig3','zenodo');    
publ_osses2022b_JASA_figs('fig4','zenodo');
publ_osses2022b_JASA_figs('fig5','zenodo');
publ_osses2022b_JASA_figs('fig6','zenodo');    
publ_osses2022b_JASA_figs('fig7','zenodo');
publ_osses2022b_JASA_figs('fig8','zenodo');
publ_osses2022b_JASA_figs('fig8b','zenodo');    
publ_osses2022b_JASA_figs('fig9','zenodo');
publ_osses2022b_JASA_figs('fig9b','zenodo');
publ_osses2022b_JASA_figs('fig10','zenodo');

To generate the figures of the supplementary materials (Appendix in the BioRxiv preprint):

publ_osses2022b_JASA_figs('fig1_suppl','zenodo');
publ_osses2022b_JASA_figs('fig2_suppl','zenodo');    
publ_osses2022b_JASA_figs('fig3_suppl','zenodo');
publ_osses2022b_JASA_figs('fig3b_suppl','zenodo');
publ_osses2022b_JASA_figs('fig4_suppl','zenodo');    
publ_osses2022b_JASA_figs('fig4b_suppl','zenodo');
publ_osses2022b_JASA_figs('fig5_suppl','zenodo');
publ_osses2022b_JASA_figs('fig5b_suppl','zenodo');

References:

• Preprint: Alejandro Osses, Léo Varnet. "A microscopic investigation of the effect of random envelope fluctuations on phoneme-in-noise perception." BioRxiv.
• fastACI toolbox: Alejandro Osses, Léo Varnet. fastACI toolbox: the MATLAB toolbox for investigating auditory perception using reverse correlation (v1.2). Zenodo. doi:10.5281/zenodo.7314014. Supplement to: https://github.com/aosses-tue/fastACI/tree/v1.2