Interaction Effects

[Brian]

Hi,

Suppose that I have an INLA model with main and interaction effects x1, x2, x1:x2, where the interaction effect x1:x2 is the interaction between x1 and x2 and was added as a variable. Is it possible to have an interpretation of the effect of the interaction effect on the dependent variable, apart from the mean (log odds)? 

An interpretation such as when x1 is high and x2 is high it has this effect..

Thanks.