A collection of generative models of response time distributions.
Focus is on providing efficient methods for sampling single responses given a collection of model parameter values so that probabilistic inference for the model can be conducted using simulation via pyEPABC. To achieve this the core functionality of a working model is implemented with JIT-compilation using numba.
Currently only some variants of the model reported in
- Bitzer, S.; Park, H.; Blankenburg, F. & Kiebel, S. J. Perceptual decision making: Drift-diffusion model is equivalent to a Bayesian model Frontiers in Human Neuroscience, 2014, 8, https://doi.org/10.3389/fnhum.2014.00102
- Park, H.; Lueckmann, J.-M.; von Kriegstein, K.; Bitzer, S. & Kiebel, S. J. Spatiotemporal dynamics of random stimuli account for trial-to-trial variability in perceptual decision making. Sci Rep, 2016, 6, 18832, https://doi.org/10.1038/srep18832
are implemented.
For a usage example see examples/fit_with_EPABC.py.