THE AUDITORY MODELING TOOLBOX

This documentation applies to the most recent AMT version (1.6.0).

View the code

Go to function

exp_barumerli2024
Examples of FrAMBI reproducing figures from Barumerli and Majdak (2024)

Usage:

exp_barumerli2024(flag);

Description:

exp_barumerli2024(..) reproduces figures from Barumerli and Majdak (2024). It serves as showing example on the usage of FrAMBI. The flag can be one of the following:

'basic' Demonstrates how to use frambi_sample by applying a simple ITD-based sound-lateralization model barumerli2024_itdlateral as the agent and simulating response distributions for tasks. It uses sig_barumerli2024 as the enviroment.
'dynamic' Demonstrates how to create a dynamic scenario and apply frambi_plot and frambi_disp to visualize and display the agent-environment interactions. In this example, the agent tracks a moving sound source and interacts with the environment over time. This example also uses barumerli2024_itdlateral as the agent and sig_barumerli2024 as the enviroment.
'variants' Demonstrates how to handle multiple model variants. It compares the results for two model variants: with and without considering the lateral gain. This example also uses barumerli2024_itdlateral as the agent and sig_barumerli2024 as the enviroment.
'fitting' Demonstrates how to handle actual data from a behavioral experiment, fit model parameters to that data, and perform model comparisons. The model comparison is based on the Bayesian Information Criterion (BIC). It uses the mock-up model barumerli2024_mockup as the agent and sig_barumerli2024 as the enviroment.
'all' Runs all the examples.

Examples:

To run the basic example use :

exp_barumerli2024('basic');
exp_barumerli2024_1_1.png

To run the dynamic example use :

exp_barumerli2024('dynamic');
exp_barumerli2024_2_1.png

This code produces the following output::

time: 0.005
       angle: 0.45
       time: 0.01
       angle: 0.9
       time: 0.015
       angle: 1.35
       time: 0.02
       angle: 1.8
       time: 0.025
       angle: 2.25
       time: 0.03
       angle: 2.7
       time: 0.035
       angle: 3.15
       time: 0.04
       angle: 3.6
       time: 0.045
       angle: 4.05
       time: 0.05
       angle: 4.5
       time: 0.055
       angle: 4.95
       time: 0.06
       angle: 5.4
       time: 0.065
       angle: 5.85
       time: 0.07
       angle: 6.3
       time: 0.075
       angle: 6.75
       time: 0.08
       angle: 7.2
       time: 0.085
       angle: 7.65
       time: 0.09
       angle: 8.1
       time: 0.095
       angle: 8.55
       time: 0.1
       angle: 9
       time: 0.105
       angle: 9.45
       time: 0.11
       angle: 9.9
       time: 0.115
       angle: 10.35
       time: 0.12
       angle: 10.8
       time: 0.125
       angle: 11.25
       time: 0.13
       angle: 11.7
       time: 0.135
       angle: 12.15
       time: 0.14
       angle: 12.6
       time: 0.145
       angle: 13.05

To run the example comparing variants use :

exp_barumerli2024('variants');
exp_barumerli2024_3_1.png

To run the fitting example use :

exp_barumerli2024('fitting');
exp_barumerli2024_4_1.png

This code produces the following output::

Optimized Parameters
  angular_uncertainty: 33.034
  confusion_rate: 0.407
Statistical metrics:
  Log-likelihood: -753.133
  BIC: 1516.077
Optimized Parameters
  angular_uncertainty: 96.311
Statistical metrics:
  Log-likelihood: -868.238
  BIC: 1741.382
The model variant "considering front-back confusions" wins
  with a BIC difference of 225.30 to the next best alternative.

References:

R. Barumerli and P. Majdak. FrAMBI: A Software Framework for Auditory Modeling Based on Bayesian Inference. under review at Neuroinformatics, 2024.

Build with Bootstrap