Masking staircase experiment

Instructions to load the staircase experiment results in Matlab

The results of each experiment run are stored in a sub-folder within the directory:


The subfolder names contain the subject, date, experimenter name and an additional tag (placed between brackets). This last bit is important because it is used to separate the data belonging to different versions/iterations of the experiment – for example if the experiment parameters were changed then the tag will be changed. The tags I’ve used so far for the masking experiment are DEBUG1, DEBUG2 or FREQS_8HZ. A folder name containing the DEBUG_2 tag for example would be named like:

NS64 02-04-2015 13.45 (DEBUG_2) (JIMMY)<a href="">View Page</a>

The results from folders with the same tag can be collated easily by:

1. Setting up the M3 toolbox by running (needs to be done only once after restarting Matlab):


2. Using the commands:

dir1 = 'V:\readlab\Ghaith\m3\data\mantisMaskingStaircase';
[~, resultSet] = loadDirData(dir1, {'DEBUG_2'}, {});

Which will load up the results from sub-folders whose names contain the tag DEBUG_2 (change appropriately).
The results are stored in the Matrix resultSet whose individual rows contain the results from each experiment (sub-folder). In the iteration DEBUG_2 there are 3 channels so each row of resultSet contains 3 thresholds obtained by one run of the experiment (3 interleaved staircases).

For example, to plot a histogram of channel 1’s thresholds after loading DEBUG_2 as above, we can do:

hist(power(10, resultSet(:, 1)))

(note thresholds are stored in log10 units)

To find out which parameters are simulated in each experiment iteration and their associated tags, check the function getConditions (line 48) in the script:


The experiment script also contains channel frequencies, bandwidths and noise settings if applicable.

Additional info

If needed, staircase stepping data (rendered thresholds and experimenter responses) are stored in a matrix called `history` within the file `results.mat` inside each sub-folder.

Each experiment sub-folder also contains:

• An entire backup of the toolbox source code sufficient to reproduce the experiment
• Video files for the individual trials
• A file (hardware_info.mat) which contains a lot of hardware info (CPU, GPU, monitors, OS etc)

Leave a Reply

Your email address will not be published. Required fields are marked *