Code illustrating the "ventriloquist effect" using a generative model and an ideal Bayesian observer model, inspired on Kording et al. (2007) and also described in Ferreira & Dias (2014).
A generative model will be used to simulate experimental trial data, which models visual and auditory stimuli corresponding to either one source or two independent sources (currently includes stub code instead, generating fixed experimental data for only 10 trials: to be edited by students).
An ideal Bayesian observer is then used to implement a model to estimate whether stimuli arise from a common cause (the "ventriloquist effect") and to estimate individual stimulus source locations from the generated experimental data. Estimates and posterior distributions are plotted when number of trials is small (i.e. <=10).
Finally, the ideal observer is evaluated for the simulated experiment w/ multiple trials, and metrics summarising its performance over all trials are plotted.
The UML class diagram for the ideal Bayesian observer class in question is shown below:
This project requires that the following packages are installed:
numpyfor scientific maths.scipy/scipy.statsfor statistics and probability distributions.matplotlibfor plots (native).seabornfor advanced plotting wrapper.
Kording, K.P., Beierholm, U., Ma, W.J., Quartz, S., Tenenbaum, J.B., Shams, L.: Causal Inference in Multisensory Perception. PLoS ONE 2(9), e943 (2007), doi:10.1371/journal.pone.0000943
João Filipe Ferreira & Jorge Dias, Probabilistic Approaches for Robotic Perception, Springer International Publishing, 2014, Chapter 4.# portfolio2