Reply to thread

Message

<blockquote data-quote="Charlaquin" data-source="post: 9095253" data-attributes="member: 6779196">Sorry I’m late to the party, but I think the most interesting insight from Skinner’s experiments is being brushed over here. What was found was that if the rat is consistently rewarded for the desired behavior, it quickly learns that it only needs to perform the behavior when it wants the reward. Once the rat figures out that push lever = get treat, it leaves the lever alone until it’s hungry. However, if the reward is associated with the desired behavior but not predictably so - that is to say, if the rat has a random chance of getting a treat when it pushes the lever, the behavior becomes compulsive. The rat knows that it sometimes gets a treat when it pushes the lever, but it never knows if this time will be the time it works. So it keeps pushing the lever over and over again, long after it has built up an enormous stockpile of treats, and in fact will keep pushing the lever even if you stop refilling it with treats. This is so effective, you can even set it up so that sometimes the lever randomly delivers an electric shock instead of a treat, and the rat will keep pushing it, enduring the shocks for the promise of a chance of a treat.D&amp;D is already well on top of this, or at least its designers are. That’s (part of) why 5e has hoard tables for doling out treasure instead of specific wealth-by-level guidelines. Because randomized treasure actually makes for a stronger motivator than predictable payouts.</blockquote>

[QUOTE="Charlaquin, post: 9095253, member: 6779196"] Sorry I’m late to the party, but I think the most interesting insight from Skinner’s experiments is being brushed over here. What was found was that if the rat is [I]consistently[/I] rewarded for the desired behavior, it quickly learns that it only needs to perform the behavior when it wants the reward. Once the rat figures out that push lever = get treat, it leaves the lever alone until it’s hungry. However, if the reward is [I]associated[/I] with the desired behavior but [I]not predictably so[/I] - that is to say, if the rat has a random chance of getting a treat when it pushes the lever, the behavior becomes compulsive. The rat knows that it [I]sometimes[/I] gets a treat when it pushes the lever, but it never knows if [I]this time[/I] will be the time it works. So it keeps pushing the lever over and over again, long after it has built up an enormous stockpile of treats, and in fact will keep pushing the lever even if you stop refilling it with treats. This is so effective, you can even set it up so that sometimes the lever randomly delivers an electric shock instead of a treat, and the rat will keep pushing it, enduring the shocks for the promise of [I]a chance[/I] of a treat. D&D is already well on top of this, or at least its designers are. That’s (part of) why 5e has hoard tables for doling out treasure instead of specific wealth-by-level guidelines. Because randomized treasure actually makes for a stronger motivator than predictable payouts. [/QUOTE]

Verification