[R] asymmetric mutual information

[JOHN VOIKLIS]

Hello,

In a human-subjects experiment collaborators (A & B) communicated
about how to categorize a number of multidimensional stimuli.
Following their conversation, each collaborator sorted the stimuli
into bins. This sort data was converted into square co-occurrence
matrices, where 1 indicated that the row stimulus and column stimulus
were placed in the same bin and 0 indicated that they did not
co-occur. I would like to figure out how much information A
transmitted to B and how much B transmitted to A; these transmissions
are likely asymmetric. Is there any way, using R, to estimate
asymmetric mutual information between the matrices? Should I be
looking at some other metric entirely?

Thank you,

John