MSNBC: "We learn more from success than failure"
Learning from failure is overrated and its redux [both from SvN] sparked a healthy debate. Well, now there’s a new study that backs up the idea “we” learn more from successes than from mistakes. Ok, it’s a study of monkeys. But still…
Josh Catone
on 06 Aug 09I still don’t think I understand this meme that Jason created about success vs. failure (except to create a conflict so he has material to blog about and talk about at conferences).
I still have honestly never heard anyone (smart) say that you can only learn from failure, and I have never heard anyone say that you can learn more from failure than from success. It’s fairly obvious that you can learn more from what you did right (at least assuming you’re trying to figure out what works).
But the way you’re framing the idea it seems like you guys are implying that you can’t learn from your screw ups. And that’s just as wrong as saying you can only learn from failure. You can learn a lot from the times you mess up.
The idea that you should “fail early, and fail often” isn’t supposed to imply that failure is required for success or failing offers some great insight that success can’t, but rather that you shouldn’t be afraid of failure. Don’t worry about things not working out, just get out there and try it. If you fail, no biggie—figure out what went wrong, learn from your mistakes, and move forward.
At least, that’s my interpretation. But the whole argument you guys keep presenting seems completely fabricated, imho.
Vesa
on 06 Aug 09We earn more under adequate amount of stress, no matter what is the outcome. Stressful situations occur probably more often when we fail – or at least when there is danger of failure? (Stress just for sake it can destroy us). We learn by doing, I think the failure part is included by someone who wanted to persuade someone else to do something unwillingly (destructive stress).
But its a fact that our brains remember bitter taste of loss longer than sweetness of victory.
Happy
on 06 Aug 09I agree with Josh. Jason set up a false premise in order to have something to shoot down.
On a related note: that study was more accurately about the affect of consequence on learning. The monkeys only got a consequence when they were right. No consequence or feedback at all when they were wrong. Some would say the acknowledgement of a guess and feedback on wether the guess was right is what caused the faster learning. With the wrong answer, they got no feedback. In the situations Jason’s talking about, you certainly get feedback and consequences from failed decisions.
Anyway… please continue highlighting interesting things from around the web. I love this blog for it’s spirited debate. And though my posts recently have seemed to be adversarial, it’s not my primary intention as a reader here. :)
gvb
on 06 Aug 09To dredge up a sports-ism, in skiing “they” say that, if you don’t fall, you aren’t learning.
IMHO, this is what is happening: if you aren’t pushing the boundaries, you won’t fail, but you also won’t excel.
The New Development Methodology (Open Source, agile, etc., and enabled by distributed source repositories), the philosophy is to sandbox developers and development so failures do not corrupt the mainline development. The result is that failure is cheap because it is isolated to a single developer. This encourages experimentation and results in breakthroughs.
The Old Development Methodology (CMMi, ISO900x, SA/SD, etc.) is to prevent failure at all costs. This is (IMHO) a result of strongly coupled development (characterized by centralized, shared, repositories) where one developer’s failure will take out the whole team for hours or days. This makes failure too expensive to tolerate.
The problem with this development model is that the benefit/cost ratio (where cost goes to infinity) drives the system such that any change results in risk of infinite cost and thus the benefit/cost going to zero. Thus, in this system, no change is the optimal solution. Of course, no change also results in stagnation and, ultimately, death. Oops.
Reference: How company policies are made.
JF
on 07 Aug 09except to create a conflict so he has material to blog about and talk about at conferences
I don’t make up arguments so I have material to blog about or speak about. I say things that I believe. You may not agree with me, but my arguments aren’t fabricated for attention.
I believe you learn more from your successes than your failures. And I believe studying your failures is mostly a waste of attention.
There’s something to learn from mistakes – I’ve never suggested there wasn’t – I’m just suggesting that if you’re going to pay attention to something, you’re better off paying attention to what works. There are better lessons there.
Everything’s a choice. We don’t have unlimited attention, interest, or motivation. So I choose to focus my limited resources on the things that work, not the things that don’t work. I think the lessons are more valuable.
This is what I believe.
Happy
on 07 Aug 09JF: “If you’re going to pay attention to something, you’re better off paying attention to what works” That, I can definitely agree with. Well stated.
I think what Josh is saying is that you offer the advice as the antidote to conventional wisdom, yet conventional wisdom does not favor study of failure over study of success. From academia case-studies to popular business press, the best sellers are those about success, not failure. Have you heard someone say different: that they’d rather pay attention to what does not work?
p.s. I took a fresh read of Getting Real the other day and now I’m really looking forward to the new book!
sensei
on 07 Aug 09I think your focus is fucked.
“Learn from your mistakes” means “if you’re going to focus on your mistakes, take something good from that focus”.
The whole of this is aimed at USING everything for something good – mistakes and successes alike.
That’s why we have this saying.
Don Schenck
on 07 Aug 09Do what I do: Don’t fail.
That is, I might not achieve my intended result, but I DO get a “result”. So, I always succeed … in getting results.
Semantics, to be sure. But it works for my (feeble) mind!
This discussion is closed.