Reader Comments

Post a new comment on this article

About equation (2)

Posted by syano on 31 Jan 2018 at 01:53 GMT

To achieve bayes' update (2), p(x, t | C=2) would be required.
It seems that the paper only gives p(x,t | C=1).
How did you defined that?

No competing interests declared.

RE: About equation (2)

msamad replied to syano on 05 Feb 2018 at 00:11 GMT

Thank you for your comment. We followed the formalism of the original model, which you can see derived more fully in citation (14) [Causal Inference in Multisensory Perception]. Specifically, we evaluate the integral separately for the visual and somatosensory signals, (Xv, Tv) and (Xp, Tt) respectively. I will try to write the equation in poorly formatted text below, but apologies for its lack of clarity, if you want more information please don't hesitate to contact me at the corresponding author's email address.

p(x,t | C=2) = int( int( p(chi_v, tau_v | Xv, Tv) p(Xv, Tv) dXv dTv )) * int( int( p(chi_p, tau_t | Xp, Tt) p(Xp, Tt) dXp dTt ))

No competing interests declared.