This is what I've learned from reading a few books and researching online, but still have to experience using this theory for myself. I think using HR properly should allow you to improve your running and adapt to different running conditions - like the great running antoganists heat and high humidity.
One note on max HR. The value changes with fitness and actually lowers the better your cardiovascular system is. I suspected this for a few reasons of my own and finally have some confirmation here which points to a paper by Zavorsky, G.S. 2000, Endurance and possible mechanisms of altered maximum heart rate with endurance training and tapering. Sports Med 29(1):13-26
During sprint workouts according to Greg McMillan (2) you should not reach your max HR. One of these reasons appears to be the short nature of the workout. I think the HR curve over time during each sprint should be the same given the same effort and other physical conditions. This could probably be used to find your actual maxHR for running and can be used for targeting HRs for all other workouts.
This also makes sense for using a 12 minute test instead of a shorter time since an 8 min or shorter race should be at or above VO2max pace which will require being at an effort that should eventually reach maxHR. Reaching maxHR is probably a bad thing to do when testing physical fitness, and Im not sure if this means you will faint once reaching it or after staying at it for some amount of time.
For speedwork on the track the average HR should not be used due to the fact that maxHR cant be reached during sprints. The HR curve should level off given a minute or so and that level should be between 90-100% of your actual running HRmax. The actual HRmax is not the "220 - age" HRmax and I am doubtful that it is the HR that causes you to pass out regardless of the exercise used to come up with a number. Also it seems the HR curve levels off only approximately - not sure.
(2): YOU (Only Faster) Ch 19 Sprints, somewhere in the first few paragraphs