Skip to content

Latest commit

 

History

History
50 lines (41 loc) · 2.59 KB

README.md

File metadata and controls

50 lines (41 loc) · 2.59 KB

Intersection information and SubPID (Subatomic Partial Information Decomposition)

DOI

This package is composed by four parts:

  • A MATLAB implementation of the partial information decomposition (PID) for distributions over three random variables using the definitions from Bertschinger, N., Rauh, J., Olbrich, E., Jost, J., and Ay, N. Quantifying unique information.. Entropy 2014, 16(4):2161–2183.
  • Two MATLAB implementations of the intersection information I_{II}(S;R;C) that, in perceptual discrimination experiments, quantifies the sensory information in the recorded neural response R that is relevant to behavior. This measure is defined and described in Pica, G., Piasini, E., Safaai, H., Runyan, C.A., Diamond, M.E., Fellin, T., Kayser, C., Harvey, C.D., Panzeri, S., Quantifying how much sensory information in a neural code is relevant for behavior, Advances in neural information processing 2017, 3687-3697. The first implementation, "src/matlab/intersection information.m", evaluates I_{II}(S;R;C) starting from the empirical joint probability distribution p(s,r,c). The second implementation, "src/matlab/intersection information_from_binned_response.m", evaluates I_{II}(S;R;C) starting from vectors containing the stimulus s, the response r, and the choice c, corresponding to each trial. Here, the response is discretized into equipopulated bins for a conservative estimate of I_{II}(S;R;C).
  • A MATLAB numerical implementation of the subatomic partial information decomposition proposed in Pica, G., Piasini, E., Chicharro, D., and Panzeri, S. Invariant Components of Synergy, Redundancy, and Unique Information Among Three Variables.. Entropy 2017, 19, 451, doi:10.3390/e19090451.

Requirements

Intersection information and SubPID requires glpkmex to be installed on your system.

Usage

The intersection information can be estimated from raw response data with "src/matlab/intersection information_from_binned_response.m", while the more general "src/matlab/intersection information.m" should be used if the user has already estimated p(s,r,c) from the experimental dataset. The PID and the subatomic PID are implemented by src/matlab/partial_info_dec.m and src/matlab/subatomic_pid.m, respectively. See the function descriptions for details.

License

This program is licensed under the GNU General Public License, version 3, or any later version. See LICENSE for details.