Hacker News Reader

Comment by Dagwoodie | Hacker News Reader

This article is dangerous. On the surface it may sound good, but there is behavior so vile and harmful to the child, family or neighbors that never showing negative reinforcement is extremely dangerous with some children.

It does have some merits though, showing plenty of positive reinforcement will likely reduce the amount of negative reinforcement needed.

View on HN

You don't want never negative, or never stern to put a lighter touch on it—but many people, probably the audience of this article at large, simply don't know enough behavior theory to even know of the alternative.

Save the negative reinforcement and punishment for situations where it's absolutely clear and necessary—that's the best way to keep it an effective tool anyway. Maybe 60% positive, and 10% negative, with the other part just not being reinforced (an incredibly powerful behavioral tool all on its own).

It's the same with dogs (not as a derogatory comparison—just the truth of the science). I negatively reinforce for biting only, and only after they're at an age where they can comprehend the association between the bite and the punishment. It's remarkably effective. For nearly everything else, positive reinforcement is orders of magnitude more effective.

Overall, the most important part is to be a majority positive player, which makes you an overwhelmingly positive symbol in your child's (or pup's) life. This has primary effects that far outweigh the alternatives and negative side-effects of negative punishment and negative reinforcement methods.

I don't agree that the article is dangerous—people are unlikely to give up negative reinforcement altogether, but being introduced to a wider diversity of well-proven methods will only help, and we need more of it.

Note that negative reinforcement is reinforcement. It reinforces (i.e. encourages) a behaviour by stopping something unpleasant. e.g. your kid asks you nicely to turn off that dad music, you flip the stereo off, your kid feels better and learns to ask nicely. e.g.2. your kid screams that he hates carrots, you take the carrots off the plate, your kid feels better and learns to scream. When you say negative reinforcement you mean punishment.

Correct—I meant Positive Punishment (which is hard to remember, since it's decidedly not positive in the obvious sense).

I wonder how it equips children for the real world where there is both positive and negative reinforcement. If they only deal with positive reinforcement, will they have the emotional conditioning to deal with negative reinforcement. For example, dealing with a negative review or getting skipped for a promotion. Something we probably all experienced, getting yelled at by a manager.

Anecdotally, being raised by an asian mom. I was subjected to negative reinforcement of her disappointment if i didn't do well. I attribute that to both my and my sister success because we both didn't want to deal with the shame of failure.

Perhaps you are conflating failure with discipline.

We need to allow kids to fail, because you will fail in real life. Sports with no scores is dumb.

Fairly tangential to if we yell at our kids or not.

Sports with no scores is fun

From what I've seen, if the adults don't keep score, the kids will do it themselves.

Please quote sources to back this up. You are proposing the exact opposite of what the article says, because of what?

The concept of positive reinforcement to shape the behavior of children and animals comes from the work of B. F. Skinner [1], the world's most widely known and influential behaviorist psychologist.

Key Works

* 1938 The Behavior of Organisms: An Experimental Analysis

* 1948 Walden Two

* 1953 Science and Human Behavior

* 1957 Verbal Behavior

* 1971 Beyond Freedom and Dignity

[1] https://en.wikipedia.org/wiki/B._F._Skinner

As an aside, this is the reason that it's important to have some understanding of the liberal arts along with formal training in math and computer science.

Of course there is extremely bad and dangerous behavior. But then there is extremely bad and dangerous behavior we don't have true control over - typical examples are compulsive, drug induced or depressed behavior. And yes, all of them need professional attention but that does not preclude us spending significant time with such persons.

With negative feedback it gets tricky very quickly as there is such a thing called intermittent re-enforcement - one provides inconsistent strong negative feedback and behavior gets worse and more persistent.

Enforcing hard limits consistently requires careful thinking and acting. It does have value in some cases. There is however a risk getting stuck thinking about the problem not solution. The author argues a solution focused approach is more effective in most cases. I don't think he covered clinical or criminal situations.

There's a mild difference between punishment and negative reinforcement, and this article is just arguing against <punishment>.

The difference between punishment and negative reinforcement is in how one models the relationship between cause and effect, and how you are increasing or decreasing behavioral frequency.

The important aspect behind punishment or negative reinforcement isn't negativity or pain, but a model of learning that has since fallen out of favor versus bayesian causal learning models.

Fully understand what you're describing, though it's worthwhile to consider the articles position which essentially states that negative reinforcement will likely prolong and intensify the behavior which prompted the reaction in the first place.

Reply on news.ycombinator.com