First thought: “No way! Abstaining from internet for the whole day and then losing the chocolate because of the wrong coin flip, that would make me really angry!”
Second thought: “Oh, maybe that’s what makes the random reinforcement stronger...”
Third thought: A part of the first objection is still valid, because I am the person who makes and protects the rules. So making myself angry or frustrated could engage my emotions and make the reinforcement stronger… but also on meta level, it could make me change or quit the game. (On the other hand, what is the worst possible outcome? If this fails, I can stil return to the original rules.)
First thought: “No way! Abstaining from internet for the whole day and then losing the chocolate because of the wrong coin flip, that would make me really angry!”
Second thought: “Oh, maybe that’s what makes the random reinforcement stronger...”
Third thought: A part of the first objection is still valid, because I am the person who makes and protects the rules. So making myself angry or frustrated could engage my emotions and make the reinforcement stronger… but also on meta level, it could make me change or quit the game. (On the other hand, what is the worst possible outcome? If this fails, I can stil return to the original rules.)
I was just throwing it out there, it seems likely that the longer time-frame of what you’re doing would make constant reinforcement optimal.