In my opinion, one behavior with perhaps the greatest potential to kill to innovation is the aversion to using the phrase "I don't know."
Let's say you are leading a new project to develop a completely new offering for your company, the outcome is highly uncertain, and someone asks you a question like what you think the selling price of the new offering should be. Of course you want to give an answer that sounds like you are in control of the process; that you are capable, and confident, and you know what you are doing. But let's face it. If you are doing something truly new, you shouldn't have any idea of what the final selling price should be. At least not at the beginning stage of the process designed to figure out what type of offering you should be developing in the first place.
Would you be empowered at your company to say "I don't know what the selling price should be?", and would you be empowered to follow up that answer with "and at this point in this type of project we shouldn't know the answer to that question?" If you gave that type of answer, would you be removed as the project leader because you "don't know what you are doing?" And if you gave an answer that would assure everyone of a certain outcome, would you be sabotaging your ability to truly innovate?
As discussed in an earlier post, companies reward certainty. Unfortunately, this reward is often employed at every stage of every process. Would I ever advocate launching a new offering without thoroughly evaluating what the correct price will be? Absolutely not. And of course, "I don't know", may really mean that the person doesn't know what they are doing. What I am advocating is that we consider the context of the rewards we employ. Rewards should be consistent with the type and scope of the work we are doing. We should be clear about the expected outcomes of a project, and what types of decisions should be able to be made at each step.
There is a time and place for every question. If we're not careful, the "right" answer will be given at the wrong time. At that point, it may as well be the wrong answer.