That makes little sense to me. A necessary criterion, as mentioned above, is that an action is not determined by the environment, it has tom come from the organism itself.
Another criterion is that it the action cennot be predicted by the history of the organism.
In experiments, we can see that both criteria are met. Moreover, a third criterion is that the organism must be in control, i.e., pure chance is not a decision. Also that criterion is met.