I think I would swap these. The logic I manage to pull off is always top-down. It seems true to a top level concenptual model, and I dig for supporting data rather than building upwards. (could be Ne though)
I think it could be Ne. For my reasoning of the labels: Top-down processes need not always have external data, it's the big picture, the grand scheme which gets widdled down to little pieces to support the big picture, while bottom-up relies on a given data, an external, to build on it, to create the greater picture. Since, we cannot parse out exactly one single function, and leave behind its influences of others (like Ne), it may be hard to guage how exactly Ti or Te is worked, in isolation, in someone. Hence, me saying, 'likened to'...as we cannot know exactly how 'thinking' works in isolation in each of us, without any nod to the other fuctions we utilize to build, and/or, break, T.
Disagrees
Judgmental processes aren't really top down nor bottom up. You're looking more at the difference between S and N.
INTJ uses a top-down approach for problem solving.
I agree with the bolded, as I think a whole cognitive processing system must be called forth in order for a true labelling of top-down versus bottom-up. I.e., how it works in a person in conjunction with all their functions.
This is the very same reason why I reject your underlined assumption. I don't really understand the logic behind perceiving versus judging processes with regards to why the top-down/bottom-up would only apply to perceiving, as judging processes also has a part to play in 'processing information'? Would love to hear your thoughts.
Cognitive processes are just that, processes of thought (which often times leads to a conclusion...a judgement). Unless you reject that there's a processing that goes on with T.
As I said to happy puppy, I don't think a single function can be described in isolation (which is why I would more agree with your INTJ point)...hence the 'likened to'.