I am performing a moderated mediation as shown below. In this particular example, this is depicted as a "second-stage moderated mediation". According to Muller et. al. 2005, moderated mediation occurs when moderation of one of the paths is significant (in this case b3); but according to Hayes 2015 one can compute the index of moderation for more direct evidence (a1b3).
Turns out, in my analysis, the conditional indirect effect was significant (a1b3 path). I can claim a moderated mediation. I further probed the mediation at difference levels of the moderator. This yielded the following results: significant estimate at low value (1SD below) but not significant at medium or high values. I am confused at what my conclusions can be? We know there is moderated mediation; but what are the takeaways from this further probing?
Lastly, and this may require a separate question, but what is the difference between simple slopes and conditional indirect effect (as I have just explained). I would like to depict simple slopes for the moderating relationship (in this case between loneliness on PTSD-Depression) but this is not the same as the conditional indirect effect, correct?