Adaptive-GRT

Repository of code, data, and analysis scripts for running Adaptive General Recognition Theory experiments (Glavan et al.)

Import the AGRT library to use the following classes and functions:

• class AGRTHandler: PsychoPy handler object that initializes and manages the internal adaptive algorithm. Similar to PsychoPy's StairHandler.
• class GRTHandler: PsychoPy handler object that implements the complete identification paradigm trial structure. Similar to PsychoPy's TrialHandler
• function RunAdaptiveBlock: Creates an instance of AGRTHandler with the given parameters and runs the requested number of trials. Returns stimuli values corresponding to the requested level of accuracy.
• function RunGRTBlock: Creates an instance of GRTHandler with the given parameters and runs the requested number of trials. Returns None.
• function RunAdaptiveGRTExperiment: Uses the above functions to first run an adaptive block and then a main GRT block that uses the adapted stimuli.
• function GRTSubjectModel: Example trial function used in the simulation study from Glavan et al.

Additional Files in this repository:

• AGRT_Exp_x_x_x.py: Runs the human subjects experiments from Glavan et al.
• AGRT Human Analysis.R: Analysis of the human subjects experiments from Glavan et al.
• AGRT Quick Accuracy Check.R: Script that prints the accuracy of human subjects in the pilot condition. Used to inform the selection of stimuli.
• test_simulation.py: Runs the simulation study from Glavan et al.
• AGRT Python Simulation Study.R: Analysis of the simulation study from Glavan et al.