Original Post:
The design of a slot machine is tricky. If it paid out too much, the house would go broke. If it paid out to little, clients would learn to avoid it. Somewhere between those extremes is an optimal point (from the POV of the house). The trick is to find that point.
There is a related issue: For a given payback percentage, should the machine be configured to make lots of small payouts, or fewer but larger payouts?
I would suppose slot machine designers pay a lot of attention to all this. But maybe not.
Response:
As I psych major in college, I learned some of the principles of conditioned behavior -- which undoubtedly have application in human behavior at the slots... I'm pretty sure my explanation below is a correct recollection of the long-ago learned information from way back then.
If you reward a rat for a behavior (such as press a lever) with food, it will learn to do the behavior. If you gradually decrease the frequency of the reward, so that the behavior is not rewarded every time, the animal will learn to do the behavior repeatedly to get the reward - even if the reward does not come at a regular interval, but is instead random, but (for example) every third time. Gradual decrease is the key, as opposed to abrupt.
If you then stop the reward, the rat will learn NOT to do the behavior anymore -- BUT it takes a while. The number of times it will continue the behavior without a reward before it stops doing it is larger and larger the greater the interval was between rewards when conditioning was happening. So if you were rewarding every third behavior, the behavior would extinguish quickly, compared to if you had stretched the reward frequency out to every 20th behavior.
Since slots have both small rewards and large rewards (and in between), it gets a little more complicated. Certainly if a machine paid back 105% but didn't pay at all on most plays, say for example it paid only a jackpot every 10,000th play (on average), most players would not learn to play that machine. And certainly if it paid back 85% long-term, but did so only with small but frequent rewards, a player is more likely to learn to keep playing that machine over the one that pays only a rare jackpot.
Again, with a combination of large and small rewards and varying frequencies, "it depends". It is undoubtedly more complex. The individual player's actual experience during the first time playing would determine their initial learning of the behavior: "play this machine." So a pretty good frequency of small rewards is essential to hooking a player on a machine; if they put their $100 in and it gets swallowed up with apparently unrelenting losses, they're very unlikely to put in a second $100, while if the first $100 lasts a while (and especially if it goes up to $200 or more while they're playing), they're more likely to put in another $100.
If a player plays a machine with an infrequent jackpot, but hits it early in their playing experience, they are more likely to continue playing that machine, since their experience supports continued play. Such an early jackpot is a big incentive to keep playing. How many "sure to lose in the long run" players do you know that make repeated trips to a casino because they won on their first trip?!
Secondary conditioning is the one where the behavior is rewarded with a bell and food together. The bell sounds whenever the behavior is performed, whether the primary reward of food is awarded or not, and so the bell becomes the secondary reward, which can help prolong the behavior even after the food is discontinued, by sounding every time the behavior is performed, and will extend the time needed to "un-learn" the behavior when the food is discontinued. This is why slots make noises.
I can't imagine that machine designers and casinos are unaware of this science, or at least understand the underlying patterns even if they don't know the actual science. It is pretty much a fundamental principle that is being applied to slots to "condition" people to play them, and to do so with a long-term negative expectation.
I wouldn't be surprised to learn that a game that awards something on the screen that is very entertaining, even when no money is paid, might also be considered as a "good machine" to play. The rewards don't have to be only monetary.
Of course, knowledgeable players such as those on this list probably ignore all this and play rationally, but most players are not that knowledgeable.
--BG===================
[Non-text portions of this message have been removed]
Posted by: Barry Glazer <b.glazer@att.net>
| Reply via web post | • | Reply to sender | • | Reply to group | • | Start a New Topic | • | Messages in this topic (1) |
