LESSWRONG

Feb 19, 2013

25 comments

A friend of mine has a rather precocious daughter with poor impulse control, and asked if I knew any behavior games that encourage children to think out the consequences of actions before they do them.

I'm familiar with the Good Behavior Game and the like, but standard conditioning hasn't been very effective with this child in the past. She's quite clever about subverting rules when possible, and shutting down entirely when subversion fails.

Please, one suggestion per thread so that the karma thing can do its thing.

6 points

Comments

sorted by

Highlighting new comments since Today at 8:51 PM

All comments loaded

TimS

7 points

In my experience, problems with conditioning / behaviorist interventions often arise because the function of the behavior is not well understood. Before you work too hard on designing interventions, I suggest thinking a bit more about the function (aka purpose) of the behavior.

For example, a toddler who tantrums because he doesn't want to get dressed and a toddler who tantrums because he doesn't get the sweet might perform similar behaviors. But the functions are totally different - my first example has a function of escape from task demand, while the second function is receiving a tangible object.

Until you have a good grip on the function of the behavior (and this can be very difficult), any intervention to change the frequency of a behavior is hit-or-miss at best. Some more discussion of function of behavior (in the context of autism) here.

Also, if the child is intellectually precocious, this doesn't mean that her other functional skills (mood management, etc) are also advanced. There is a limit about how much a parent should reasonably stress out about skill limitations that are developmentally normal.

7 points

Baughn

5 points

Well, since we're talking videogames...

Any Roguelike, for instance. Games where if you die, that's it.

Since she's going to die a lot at first, additionally rate-limit restarts.

5 points

PhilipL

1 points

Roguelikes are on average, way too hard for 6-8 year old kids.

1 points

Baughn

1 points

Minecraft, perhaps?

1 points

0 points

Terraria in one of the harder difficulty modes (lose items on death) might be more appropriate.

0 points

CronoDAS

0 points

The Fire Emblem series has a lot of the properties you'd want, but it's probably too frustrating. Considering her age, I'd go with one of the Pokemon games, actually...

0 points

Will_Newsome

4 points

For taking into account others' reasoning processes and how they interact with one's own, chess. Specifically the 'if I move here then he moves there then I move there' calculation algorithms. (Chess would probably also be good for general keeping track of consequences but other games might be able to teach that less discouragingly.)

4 points

army1987

1 points

I think that (for adults) poker is even better, because you have to think probabilistically (life isn't a perfect information game); but I'm not sure it is feasible for a young child to learn it.

1 points

Will_Newsome

2 points

Fights are messy but you practice kata precisely. (This applies to more than just chess vs. poker.)

2 points

0 points

It might be absolutely fantastic, except for the fact that you get it -wrong- so much of the time. You have to be adequately skilled at anticipating consequences in order for chess to teach you that you should. If you're pretty poor at guessing what somebody will do (or if your opponents are sufficiently unpredictable), it can easily teach you the opposite; don't bother wasting time planning things that won't happen.

0 points

cata

2 points

I've heard from chess teachers that they think chess serves this exact purpose very well. Seems plausible.

2 points

TrE

8 points

[Warning: Anecdotal Evidence. Don't take too seriously.]

As a chess teacher, I have to partly agree, partly disagree. For those children that stick with chess, I have indeed noticed improved impulse control and concentration. At the same time, of those children who quit chess, a good proportion had these problems.

So it might be the case that chess improves impulse control, but at the same time selects for it. In any case, the age of 6-8 is perfect for starting chess.

8 points

2 points

Is the goal to get her to anticipate consequences before acting, or to prevent certain kinds of action? And how old is she?

2 points

1 points

Both, and somewhere between 6 and 8.

1 points

2 points

I would recommend picking a particular goal, working on that, and moving to the next only when the first is complete.

At 6-8 years of age, anticipating consequences is a pretty big step. I'd recommend working on that first, as it may make the latter more easily resolved.

She's probably too young for chess or roguelikes, as other commenters have suggested. (If she's advanced enough to play roguelikes, Dwarf Fortress would probably be better.) Checkers should be within her age range. However, I recommend eliminating the rule that you -have- to capture a piece if the move is available.

If checkers is a bit advanced, Tic-Tac-Toe is also an option.

2 points

CronoDAS

1 points

(I don't know why this comment is sitting at -2. It seems perfectly ordinary to me.)

1 points

-2 points

(Perhaps because it's fully general advice followed by a non-helpful suggestion.)

-2 points

Kaj_Sotala

1 points

(How is "what the others have suggested is probably a little too advanced for her age, here are games that serve the same function but might be more age-appropriate" non-helpful?)

1 points

0 points

It's fully general advice for a problem you've seemed averse to specificity about. Only a fool would provide specific advice for a general issue.

0 points

CronoDAS

1 points

In terms of video games, the Fire Emblem series has a lot of the properties you'd want, but it's probably too frustrating. Taking her age into account, I'd suggest one of the Pokemon games. There's plenty of opportunity to use planning and thinking about consequences when playing them.

1 points

James_Miller

0 points

If the daughter loves video games then give video game time as a reward for good behavior, and take away video game time for bad behavior.

0 points

gwern

3 points

I'm reminded of http://lesswrong.com/lw/2ay/rationality_quotes_june_2010/2385

3 points

Qiaochu_Yuan

0 points

Does this actually work? My understanding of conditioning is that it works best if the reward or punishment immediately follows the action, which the taking away of privileges doesn't really: the consequences aren't felt until later (which I imagine is why conditioning doesn't stop people from drinking even if they consistently get terrible hangovers). I seem to recall that when my parents tried this kind of thing on me it just made me like them less.

If the goal is to get the child to think about the consequences of good and bad behavior, then you're the one tacking on artificial consequences to that behavior, and I think children understand this. The real goal should be to get the child to think about the natural consequences of good and bad behavior if you're going to use real-world behavior at all (as opposed to game behavior).

0 points

James_Miller

0 points

It works with my 8-year-old son. We have a learning/screen system set up where he has to do a learning exercise (such as math with daddy, Dual-N back or BrainPop ) to get video game time. Several times a day he will say "let's do learning/screen" to start the process. He has internalized the need to do work first before getting what he wants.

0 points

Luke_A_Somers

0 points

I'd try a puzzle game where you only get one try on each puzzle before moving on to the next, and there's no time limit.

0 points