Is anyone aware of any studies of programmer performance under stress:

	* External pressure
	* Self-applied pressures
	* Long hours
	* etc ...

''Is it safe to presume that performance studies of workers in general would not be useful?  I would imagine that there is at least some applicability.''

Certainly studies of knowledge workers could be expected to correspond. I'd be willing to argue from studies of garbage men if the study supported my position!
----
I don't know if this will help nail something down or not:  In the novel TheDeadline, one of TomDeMarco's characters notices that productivity only increases by a fixed percentage regardless of how much pressure is applied.  He asks an oracle why this is.. the answer comes back: "because people don't think faster under stress."

It seems like it might be hard to do a controlled experiment of knowledge work like this.  You'd never know whether it was individual ability or something else which made the difference.
----
I believe there have been studies done on this kind of issue using 
programmers.  A good place to start would be the journals ''Work And Stress''
and ''Ergonomics'' (available at or though your local college/university).