Leaving aside the career and intellectual aspects, consider your brain's reward system.
With big problems, you work a long time in ambiguity, and may never get a clear positive result. People may never care what you did. Lots of difficulty, little reward. The only way is to really believe in the value of what you're doing, typically as a contribution to humanity or at least the field.
In debugging, there's the rush of a new problem, the chase, and the solution. Couple that with the excitement/necessity/importance of being on call, and there's quite a thrill, plus a social persona as someone heroic and necessary.
So yes, debugging can be addictive.
The brain's reward system is very difficult to reprogram. After you get used to the reliable quick hits, it's very, very hard to go back to long-deferred gratification founded on a values-based identity.
You can tell you're addicted when nothing else seems interesting, you're impatient, and you can't really see yourself doing anything else. Often, the alternative opens up a whole can of worms about who you are, the value of your life, etc. So you don't.
Remember, Dr. House was an addict in pain, using work investigations to distract himself from his social isolation.
The reward system locks in valuations mostly by changing receptor populations. These take weeks or more to revert to the norm. The system also creates neuronal connections, which may never really atrophy to nothing. So stopping means a few days of angst and frustration, a few weeks of feeling out of it, and life-long defenses against falling back into it.
Here's a different recipe for impedance matching:
- When you're bored, try something larger, faster, and more challenging
- When you're overwhelmed, try something smaller, slower, and more exacting
Few have the self-understanding to realize their perspective is being shaped by their activities. The wisest choose their activities to produce the kind of perspective they aspire to. Only young adults have this opportunity. Kids have no perspective, and older adults are bound up in their psychological, social, and economic context to escape (unless some precipitating trauma severs their attachments).