I think the right terms for this are "variable reward" and "variable interval" here:
https://en.wikipedia.org/wiki/Reinforcement