Well, the really obvious answer of "Studying how action potentials are generated is important for understanding behavior, and definitely warrants a serious consideration" is just too pat and saccharine for me to allow myself to write. I think there is a decent case for each side. Spending a week on action potentials may be fine for some, but I like synapses, and what happens there, more than I like neurons and axons and depolarizations and thresholds and permeabilities, etc. Although neurons have thousands of connections to and from each other, action potentials don't communicate at those connections, neurotransmitters do. A neuron can only transmit one action potential, but it can receive input from receptors sensitive to many different NT's, and possibly transmit different NT's (I can never remember whether they can or can't).

The flip side, though, is that to know which NT's to release with what frequency in reponse to which dendtritic receptors, an action potential needs to communicate down its length, and the frequency and firing pattern of these AP's adds another whole dimension to the complexity of nervous system activity. What we learn about potentials is a complement to synapses, an important base, well understood and much observed, that links one synaptic release to another, adding a new level of complexity to the process. While understanding synapses is what interests me, it's impossible to do without knowing what lies in between them.

It isn't actually so saccharine a question. Lots of people argue quite seriously that what is critical for behavior is some set of more abstract information processing rules, and that those can be derived and understood independently of the particular physical substrate in which they are instantiated. In which case, its not only action potentials but synaptic potentials which can safely be ignored. Assuming we don't buy that argument, the action is indeed at the synapses, but what goes on there depends (as I've argued) on exactly the same set of basic phenomena one sees in action potentials. Hence, its all part of the same understanding, no? PG