I thought Google's Project Aristotle [1] nailed the problem.

TL;DR: They propose two traits of an effective team. First, long term everyone speaks about the same amount even if the meeting to meeting amounts differ. Second, there is a general awareness among team members that if they take interpersonal risk (i.e. say something that may turn out to be wrong) they wont be embarrassed, rejected or punished.

And I think this paper's results fall in line with that train of thought. If "risky"/creative ideas are shot down on sight your sense of psychological safety and probably your average amount of talking is going to take a big hit over time.

[1] https://www.nytimes.com/2016/02/28/magazine/what-google-lear... (If you decide to read it the second half is the meat/results, the first is background info)