I would like to share some intuition for function identification and testing. I'll just go through my own stack, so the later functions will perhaps be more prone to error and less detailed. I will try to make this a general, all depth overview of cues. Systematically, we can then start to generate detailed descriptions of functions based upon depth.
Ti
essence: past tracking of external opaqueness
testing: logical deduction and missing generalization
Logical deduction starts with some givens (e.g. if this then that, A or B implies C, etc.) and then demands a resolution to a question that is presumably answerable by what is given.
Givens and question: If some cats are blue and some blue animals eat cheesecake, then do some cats eat cheesecake?
Answer: Not necessarily, because it is conceivable that the set of cats that are blue does not overlap with the set of blue animals that eat cheesecake.
When testing for Ti, other functions can be better factored out by inventing an absurd scenario like above. No cats are blue and usually do not eat cheesecake, so it is less likely that someone would be right based upon real-life inference (Te is especially good at picking up on real-life logical chains). Longer chains will require higher depth Ti as well as higher IQ.
Missing generalizations are deductions in reverse. The givens are not given, instead you start with a statement that makes hidden assumptions.
Statement: Tommy is a cat therefore he eats olives.
Missing generalization: All cats eat olives. OR Animals named "Tommy" eat olives.. etc. (generally an infinite number of solutions can be given through recursion).
Ti being introverted means that it is not usually directly apparent and requires probing a bit below the surface. A strong Ti presence will tend to bring up prior conversations and claims made within a conversation. Inconsistencies may be pointed out, or if Fe is co-developed, Ti-users may ask questions that lead to someone rethinking their position. They may even be able to do this with the other person thinking they changed their own minds on a topic. Yet, from the Ti point of view, these deliberations can go on an entire lifetime as far as I am aware (hence decades long academic debates over subtleties).
Another sign being that Ti-users will tend to make strong claims seemingly out of the blue. Then, if asked why they said or believe that will continue on to give their reasoning, usually in a way that can be followed from prior shared assumptions. Though, sometimes Ne paired with Ti just goes off the rails.
Ne
essence: present creation of internal concreteness
testing: diverging stream of thought
As an extroverted function, Ne will be more readily apparent in loose conversations where creativity is appropriate. Often Ne will not really have a destination or objective and will just be throwing mud at the wall to see what comes of it. On the internet this will often come across as trolling when the respective feeling functions (Fi and Fe) are low depth. It is more challenging to invent a closed procedure for testing the extroverted functions since they diverge out of boxes (that is what they do!), yet for Ne this could perhaps be done by measuring (formally or informally) how unique and random a train of thought is without any clear objective. Simply ask them to tell a story that no one has ever heard before or to rap.
High depth Ne fantasy stories can get pretty weird, fast. I would give an example, but I figure we have all encountered such and I respect the patience of my audience, ha.
Polar or 7th slot Ne will come across as strict speaking of facts in as few words possible. Which is great in a situation with a clear, existential objective that requires a no non-sense solution ASAP. Think of an ISxP firefighter that doesn't hesitate to act in a crisis to save lives.
Si
essence: past tracking of external concreteness
testing: detailed recall of past
This function is by far the easiest to test in a strictly controlled way. The time period of the past is flexible. The further back in time you go inevitably the details will start to become chapters filled with stories that may jumble details. A basic test is to ask about what happened thus far today in as much detail as possible. High Si may be able to go on for an hour about just today if they really wanted to. Low Si or even polar Si (ENxJ's) may just say strait up that they would have to think about it. This isn't because they couldn't do it but because it would take effort and they have learned (or otherwise fell into the pattern) of not recalling the past.
If I had to give some numbers.. for say a mid-aged person (40-60 yo):
High depth Si will remember life in 5+ chapters each with 5+ stories each with 5+ details without much recall effort. Mid depth Si will remember life in less than around 5 chapters with either a dark or light shadow that distorts serial recall and stories/details that are sometimes misremembered or out of order. Low depth Si often seeing the past as one thing without chapters that is scrambled with light and dark shadows (hard to arrange).
Though, don't take my word for it, I have limited evidence (say around 20 high quality data points and 200 much lower quality data points with each point being a human life).
Fe
essence: present creation of internal opaqueness
testing: multiple interfacing styles
Fe may be one of the easiest functions to observe directly in a situation with lots of people interaction. The giveaway is the range and fluidity of facial expressiveness. High depth Fe is often required to sustain being "the life of the party" where contextual switching between both light and dark situations may be warranted. I have also noticed this pattern whereby people with relatively lower Fe (usually conscious but sometimes subconscious) can become allured by a charming, high depth user (especially if they are of opposing sex). Another way to spot or test such is through imitation ability. Fe can pick up on accents, body language, speech patterns, and so forth to convincingly play a role.
Though, this can often backfire in which the role being played is too deep (Fi) to be imitated on short notice and thus comes across as superficial. Hence, a high depth Fi user will see right through the act and write it off as inauthenticity (unless it is all for laughs, of course).
----------------------
The next four will be in my subconscious much of the time and thus should be taken with a grain of salt. I would love to be corrected or given counter evidence as this is also an exercise in honing my own intuition and getting thoughts written down.
----------------------
Te
essence: present control of external opaqueness
testing: problem solving with constraints
Of the extroverted functions, Te may be one of the easiest to formalize or enclose in a "procedure". The basic idea being that Te works well under constrains to solve a problem. This is being done constantly here on the forums. When testing, it would probably be best to invent an unlikely scenario that is still realistic yet never actually happens. The Eric guy on Talking with Famous People likes to give the "troll problem" where a small village is separated by a river from crop fields that are ready to harvest. The problem being that a troll is in the middle of the river and refuses to leave, instead demanding a troll tax to pass the river. What do you do? There are many solutions involving fighting, moving elsewhere, paying the tax, gaining leverage on the troll, and so forth. Eric will then proceed through storytelling to explain what each strategy leads to. The Te test being to basically stumble upon solutions he has not yet anticipated or to cycle quickly and efficiently through the entire solution tree.
Thus, perhaps a pattern with many of the testing techniques is that Ne is quite helpful in creating novel questions and situations that detach enough from reality to avoid familiarity. The opposing force of Ni perhaps helpful in insuring the fantasy isn't so far off as to lead to non-nonsensical typing outcomes.
Ni
essence: future matching of internal concreteness
testing: synchronous perception that fills gaps
Which leads us to Ni as being a rather difficult function to test for. Though, I suspect high-depth Ni users can identify other high-depth Ni users quite reliably given that the universe can be interconnected synergistically such that alternative synergy paradigms will tend to match even if all the details or words are different. People coming from a theoretical physics, an evolutionary biology, a theology, or a meditative background will all have likely pondered many of the same deep existential questions each with their own latticeworks of generally applicable understandings for what the universe is.
As far as testing is concerned, what is being sought is evidence that the person has a vastly applicable web of internal connections. So, a test would generally give some pieces of a puzzle and seek an intuition for what the puzzle is about or looks like. As distinct from Ti missing generalization problems, the puzzle pieces would not presume to know anything and instead be concrete particulars that fit together in only so many ways.
This is a great resource for RAT's or remote associate tests:
https://www.remote-associates-test.com/
When testing it is best to do at least 10+ of these as the success variance is quite large even for high-depth Ni users. Ni, along with Ti, also seems to be quite readily tested by IQ, so should be factored in. Interestingly, for longer deduction chains, Si becomes important and tends to counter high Ni (so it's a bit loopy and avoids linear correlations).
Se
essence: present control of external concreteness
testing: smooth engagement in unpredictable tasks
This function is most easily seen when someone is in an alpha(*) flow state. As when they are engaged on a familiar yet challenged task, the ability to improvise will be telling of Se depth. I suspect Se users will likely enter into many of these flow states throughout the day/week for a variety of tasks. In contrast with Si, these tasks will require non-linearity of movement and execution whereas Si will tend to be habitual and linear. Se crosses items off a list as quickly as can be potentiated, whereas Si would rather generate a list to be done sequentially.
(*) That is, a frequency of brain waves.
A sure tell being a generalist doer with little hesitation to try novel concrete tasks, especially in the presence of a challenging yet rapid feedback cycle for learning.
This being my polar function, I have had to slowly build it up in constrained contexts over long periods of practice. Though, I can readily use Se consciously when cooking, cleaning, dancing, some sports, computer navigation, and gaming.
Fi
essence: future matching of internal opaqueness
testing: peeling back onion of why's
I feel/think
that Fi is future oriented, though it may not seem like it when it is indirectly expressed as an introverted function. Surely enough though, it seems to be focused on the internal harmony of emotions into the future. Part of this involves not being blindsided by faulty assumptions and hence high-depth Fi users will tend to have answers (expressed or not) for why they are saying what they say or do the things they do. Fi, like Ti, is very deliberate and tends to speak about principles, but the primary distinction is that Fi principles are highly personal and the assumptions underlying them are not arbitrary. If Ti users derive logical trees in the sky, then Fi users peel back like an onion towards the heart.
Hence, Fi is probably the least testable function in that peeling the onion may lead to unexpected triggers. Fi users will tend to have a boundary at which questions become too personal. Though, in my experience, high-depth Fi users will tend to have explored their own depths enough to feel perfectly comfortable letting loose with the right people (those they trust).
-------------
Probably one of the best, if not the best (i.e. most consistent) typist in the world is Eric Strauss. Here is a playlist with nearly 100 of his typing sessions:
https://www.youtube.com/playlist?list=P ... BCQX6cF-IF