Dear Justanotherflyer,
You are referring to the concept of "Primacy". This is a very well accepted and well documented aspect of how people learn and how it affects subsequent behaviour. I am not a believer in citing "Wikipedia" but its a good source of references and for the purposes of this forum gives quite a nice explanation of "Primacy".
“...Primacy, the state of being first, often creates a strong, almost unshakable, impression. Things learned first create a strong impression in the mind that is difficult to erase. For the instructor, this means that what is taught must be right the first time. For the student, it means that learning must be right. “Unteaching” wrong first impressions is harder than teaching them right the first time. If, for example, a student learns a faulty technique, the instructor will have a difficult task correcting bad habits and “reteaching” correct ones.
The student's first experience should be positive, functional, and lay the foundation for all that is to follow. What the student learns must be procedurally correct and applied the very first time. The instructor must present subject matter in a logical order, step by step, making sure the students have already learned the preceding step. If the task is learned in isolation, is not initially applied to the overall performance, or if it must be relearned, the process can be confusing and time consuming. Preparing and following a lesson plan facilitates delivery of the subject matter correctly the first time...”
Retrieved 3 Jan 2011 from Wikipedia: Principles of learning - Wikipedia, the free encyclopedia
Most of the early work was done on animals (as it always is) - you may remember Pavlov's dogs that demonstrated the "conditioned response". A Google search on Thorndyke will reveal a lot. There have been a bucket load of "human" experiments and follow up work with pilots (as well as other professions) and the concept of primacy or "reverting to type" is often cited in accident reports as causal.
Your queries regarding whether primary learning can be "undone" is less clear in the literature but the basic answer is probably "yes" with some caveats.
New neural pathways can be created in response to a stimulus that now requires a different response, however the old ones can't be destroyed (or at least we don't think so). A good example is to look at the training of an experienced light aircraft pilot onto a swept wing jet. For several lessons, they find themselves unable to stop using the rudder in an attempt to balance turns. Every time they roll with aileron, on comes the rudder. Eventually, usually after a few simulator rides, they have learned to stop using the rudder. However, come the check ride / 1179 / type rating test, there goes the rudder once again! I have had the pleasure of conducting many such conversions and have observed this phenomena many, many times. I have also converted many experienced airline pilots back to light aircraft and it rarely takes more than 10 minutes for them to get back into finessing the rudder as they used to so competently many years before -so those neural pathways were not destroyed, they have just been sitting dormant.
However, this ability to "switch" behaviours based on situation does require some spare cognitive capacity, which is something that we tend to loose under stress. Under extremely stressful situations we are more likely to revert to the most basic of behaviours - i.e. those that were learned first or revert to even more basic behaviours that were not learned but exist by instinct e.g. dissociative and inappropriate responses such as "freezing".
Try the International Journal of Aviation Psychology as a good start point for further information.
With respect to VinRouge's suggestion regarding MBTI, this is an interesting hypothesis but, to my knowledge, has no scientific backing. The MBTI is a useful tool to explore personality types but is not a scientific one. There are many interesting theories regarding MBTI linkages but, to my knowledge, there is little or no lierature on the area under discussion.
Hope this helps. Have a good 2011 everyone.