One of the things that I've lamented over the last few years is the loss of what might be called innocence in observing and interpreting people's behavior, a certain connection to the world out there. When I first started studying cognition, I looked at how people behaved, and used that to inspire my search for research and theories that might explain that behavior. There were problems with this approach. I was not yet sufficiently familiar with theory to interpret people's behavior in deep and systematic ways, and was thus limited to surface features. This sometimes made it difficult to find the relevant research, but it also made me evaluate theories and experiments based on my real-world experience with the behaviors that they were trying to explain.
By now, that innocence is long gone. It is difficult for me not to interpret people's behavior using my knowledge of research and theory. In other words, my approach to people's behavior is the opposite of it once was. Instead of going from behavior to theory, I go from theory to behavior. This has its advantages. I am able to interpret behavior at a much deeper, more systematic level. However, it also has its disadvantages. One of the most common criticisms of cognitive psychological research is that it is not connected with the real world. We cognitive psychologists lock ourselves in our labs and conduct strange experiments, in strange contexts, on university undergrads. Few of us perform experiments out in the real world, because we're worried about control. Before, I frequently noticed the disconnect between cognitive scientific explanations and real world behaviors. Now it's more difficult for me to notice them.
Interestingly, this is exactly what you would expect based on the research on the development of expertise. The primary difference between novices and experts concerns depth of processing. Novices tend to lack the relevant theoretical knowledge that would allow them to notice complex relational structures, and are thus limited to surface features when processing information. This is how novices solve physics problems, play chess, and perceive the information in x-ray photos, for instance. Experts, on the other hand, reason about information in their area(s) of expertise at a deeper, more relational level. Their theoretical knowledge allows them to immediately recognize deep relations between information that are not apparent on the surface, and thus to reason about that information at a deeper level. The development of expertise follows a pattern that moves from using surface features to understand theory to using theory to understand observed surface features based on underlying relations and principles. This is exactly what has happened to me.
Sometimes I wish that weren't the case. I was able to come up with some very interesting and novel ideas when I first started grad school, because I was coming at cognitive scientific problems from a naive perspective, and using my real-world experience to develop ideas about them. It's virtually impossible for me to do that now. Notice how I can't even write a blog post about it without invoking the theoretical explanation. I'm afraid that my ideas have suffered in some ways as a result.