Most of my work is doing blockbusting like that, not doing easy things. But certainly, if someone on the team knows something about the tech I'm researching, I check in with him.
If you're working on rare things where you are the only expert on the team then yes, you can have the opportunity to go into full focus mode and hammer something out. And as the other poster said, sharing that knowledge after you've finished is part of being a team player. If you simply say 'I am an individual contributor, I turn in my work and then I'm done' that to me sounds like someone that's not interested in sharing knowledge.
You dont need to be ONLY expert on the team for the above. Other people knowing the same thing dont prevent your knowledge of the system. For the above, you need to have passing knowledge of the thing you work on. And you should have build that knowledge while doing previous taks.
The situation in which majority if tasks is assigned to developer who has no clue about that area of the system, again and again, is unreasonable.
You seem to be describing micromanagement.
Genuinely, if there are so many questions that few hours of solitude are red flag, then something is massively wrong with the process at hand. It really means that tasks are less then underspecified and likely means testing/demoing processes are completely lacking.
> In more complex systems you might be working on a very small part, and exiting into the sun with a system that doesn't fit the rest of the system means you've spent the past 7-8 hours engineering a solution to the wrong problem.
Sure, you work in small part. But if all this is routine issue due to 7-8 hours if engineering, then the system is likely mess too difficult to decipher. It also means the development is overly chaotic. The developers working on the project should build up knowledge of the rest of the system overtime, this is somehow failing.
Like let me make this clear, I am a born introvert. I prefer being by myself and doing my own thing. That makes me happy. But that doesn't mean I hate being a team player, nor do I think programming is a solo activity. It is very much so a group one.
And in my experience, the people that tend to go into a cave for a few hours and produce a solution will tend to produce the wrong solution. Because no spec or ask is perfect; you should be asking clarification and follow-up questions (even obvious ones!) to understand what is actually wanted. In more complex systems you might be working on a very small part, and exiting into the sun with a system that doesn't fit the rest of the system means you've spent the past 7-8 hours engineering a solution to the wrong problem.