I think it's accurate in certain problem domains, and the causality runs both ways.
I think that for technical tasks, ones with a good deal of depth that need to be performed precisely, insecurity correlates highly with success. For two reasons. One is that insecurity is a very potent emotional driver: if you think that being good at X is going to give your life meaning, then you'll probably spend a lot of time getting good at X. Time spent mastering a subject is probably the best predictor of how good you get at it.
The other works in the other direction. The more skilled you get at something, the more you understand the complexities of the problem domain, in sort of a reverse Dunning-Kruger effect. And so as you become more successful, you start seeing that your knowledge, as a fraction of the total knowledge out there, is much less than you thought. The result is insecurity.
For people tasks, where a significant part of your success is convincing other people to go along with you, success is correlated with confidence. There's a game-theoretical explanation for that. When evaluating the credibility of a proposal, a random person has very little information to go on. They also know that the person proposing it has more information than they do. So a major information channel for them is how sure the proposer seems to be of their proposal, and they naturally follow subconscious signals that telegraph how the proposer really feels about himself.
Whether high confidence or low confidence is better for you depends on which stage of your career you're at, and also who you're talking to. In particular, you don't want to project high confidence when the listener has backchannels that can tell them unequivocally that you're wrong: you lose all credibility then, which makes you seem deathly insecure. When the listener has poor or no information, though, you want to project high confidence because that's the only information they have available to them. When evaluating yourself, you'd ideally want to have no confidence as that would give you the greatest drive to improve. Unfortunately, confidence doesn't work that way: it's very difficult to have no confidence and then suddenly "turn it on" for an important client. Instead, an optimum solution usually involves having supreme confidence in your ability to learn anything required for the job, no confidence that you actually know anything right now, and the ability to fudge the distinction between those when you talk to other people.
The scientific method actually enshrines this into a well-known process.
Comments
I think it's accurate in certain problem domains, and the causality runs both ways.
I think that for technical tasks, ones with a good deal of depth that need to be performed precisely, insecurity correlates highly with success. For two reasons. One is that insecurity is a very potent emotional driver: if you think that being good at X is going to give your life meaning, then you'll probably spend a lot of time getting good at X. Time spent mastering a subject is probably the best predictor of how good you get at it.
The other works in the other direction. The more skilled you get at something, the more you understand the complexities of the problem domain, in sort of a reverse Dunning-Kruger effect. And so as you become more successful, you start seeing that your knowledge, as a fraction of the total knowledge out there, is much less than you thought. The result is insecurity.
For people tasks, where a significant part of your success is convincing other people to go along with you, success is correlated with confidence. There's a game-theoretical explanation for that. When evaluating the credibility of a proposal, a random person has very little information to go on. They also know that the person proposing it has more information than they do. So a major information channel for them is how sure the proposer seems to be of their proposal, and they naturally follow subconscious signals that telegraph how the proposer really feels about himself.
Whether high confidence or low confidence is better for you depends on which stage of your career you're at, and also who you're talking to. In particular, you don't want to project high confidence when the listener has backchannels that can tell them unequivocally that you're wrong: you lose all credibility then, which makes you seem deathly insecure. When the listener has poor or no information, though, you want to project high confidence because that's the only information they have available to them. When evaluating yourself, you'd ideally want to have no confidence as that would give you the greatest drive to improve. Unfortunately, confidence doesn't work that way: it's very difficult to have no confidence and then suddenly "turn it on" for an important client. Instead, an optimum solution usually involves having supreme confidence in your ability to learn anything required for the job, no confidence that you actually know anything right now, and the ability to fudge the distinction between those when you talk to other people.
The scientific method actually enshrines this into a well-known process.