My reading results:

I'm in!

I had to modify the code a bit to accommodate the type of data the reading app generates, but it seems to be working. Tabulating the results for five people is taking pretty long for each run, and to be honest, we don't even need five people. Two would work perfectly fine and five takes awhile to run code for.
I was thinking about just using the code I already had that calculates based off of single dichotomies at a time, but the way the app works doesn't really work well with that strategy. The differences in percentage between people's most likely type guess and their second most likely type guess can vary by a few percentage points so I just reworked a lot of stuff from the ground up. @scientiam let me know if you're ok with me starting to post the results based on the feedback that's been generated up till now.

Apparently, Ni and Si are very hard to distinguish (at least without finished signal descriptions).

@auburn: "It’d be really neat to see which Si signals were selected though, since almost everyone had strong Si in their tally — so I can know where the confusion is."
-dulled eyes, as they generally looked bored (Ti influence?)
-confused stare, when he was confused by the drawing on the board
-concerned concentration, as he is very serious (general Pi signal?)
-dancing brow, very animated brows (Fe influence?)
-anecdotal rambles, like the car ride with his brother
-indented sockets, since his eyes are quite closed (general Pi signal?)
-concerned eye-drifting, as his eye-drifts are short and near
-silly-serious, as playful moments are very short (general Pi signal?)

I wasn't really able to decide between FeSi or FeNi (with Ti-developed). I forced myself to use only this video and the CT Reader to not mess up with the data. But using only one video, is not always optimal, imo... And the CT Reader doesn't take into account the development levels/functions' energetics (I got 61% for FeSi, followed by both TiNe and FeNi at 53% - and I wasn't thinking Ti-lead at all for him) - not complaining: it's a good tool to keep in mind all the signals when typing somebody.
The eye area seemed taut, indeed... But some of his ocular and eyebrow movements were a bit confusing… In this particular video, at least (in a real situation - without being constrained by the test's standardisation - i would probably search for more data before sending an answer).

@sander - Roger that!  The signal pages definitely need completion and that's my shortcoming. For the time being I've made 8 videos covering the 80 function signals in this thread here -- which can serve as primers. Here are videos of both all the Si and Ni signals to compare:

Si-1 through Si-10

Ni-1 through Ni-10

In CT the phrases like "confused stare" and "dulled eyes" are technical terms referring to very specific physical features -- so that their definitions differ from the typical nomenclature. The names are in some sense just labels used as shorthand for the visual aspects.

-dulled eyes, as they generally looked bored (Ti influence?)

So dulled eyes doesn't refer to boredom (boredom is actually more of an Ni signal, as a variant of "unimpressed" eyes). Dulled refers to a doe-eyed blank stare similar to this:

^ Si Dulled/Doe Eyes
^ I may just re-title this signal Doe-eyes, as that's been on my mind for a while now and it makes for a less confusing terminology.

-confused stare, when he was confused by the drawing on the board

I can't find the exact timestamp of this, do you think you could point me to it? But In general, anyone can be confused. Being confused is not the same thing as the CT signal.

-concerned concentration, as he is very serious (general Pi signal?)

Pi-5: Searching Scowling is a general Pi signal yes, and it differentiates into Si and Ni based on the shape of the brow. So we gotta look at that. Here's a still shot compilation from the two videos above:

-dancing brow, very animated brows (Fe influence?)

No no, I think you totally missed it. Going off the signal names (and what they sound like) alone is a bad idea. You gotta see the signal itself in some examples. Again sorry for not having more data on this but is what it looks like:

^ Si Dancing Brow

-anecdotal rambles, like the car ride with his brother

This could be a check for Si, yeah. Ni users are not incapable of recalling memories but the structure of the code would require this being a mark for Si which is fine since the confirmation of Si would come from a wider spectrum of signals being visible. 🙂

-indented sockets, since his eyes are quite closed (general Pi signal?)

I'm not sure what you mean by this, but if by eyes "closed" you mean narrowed/squinted, that would be Ni. We actually do have this one in the code pages as:

-concerned eye-drifting, as his eye-drifts are short and near

I'm not sure what's meant by short and near here, but I hope the video above answers this question better. Lemmy know if there's still confusion.

-silly-serious, as playful moments are very short (general Pi signal?)

Again I hope the videos above answer this one. 🙂
 

This is a very interesting project!  I didnt discover it before now.
I would guess that the experiment must be inspired by the experiment that Francis Galton did -  Wisdom of the Crowd?
Take a look at this wiki:  https://en.wikipedia.org/wiki/The_Wisdom_of_Crowds
My friend has done a lot of experiments like this with hundreds of people guessing the amount of  corn flakes in a big glass jar.  People give their wild guesses, some say 200, some say 5000.  But the average of the guesses are very, very close to the actual amount, and what is strange is that the average is always more correct than the closest guess.
I hope you will make yet another experiment like this, Scientiam.  The amount of people participating this time was far too few for the 'Galton effect' to show up.  I dont know how many is needed for the wisdom of the crowd is beginning to be seen, I can ask my friend if you want me to.
 

Hi Scientiam,
I just realized there is a big difference between Galtons/Brain games experiment and this one.
Galton asked people to guess the weight of a cow.If we say that the app will at some time contain all the gifs with signals as well defined as Kilograms/pounds are well defined measures,  it is still different from the Galton experiment as in his experiment people were free to make guesses as far out as they wanted to both sides . Whereas in this experiment there is a limit.  Say a person has 8 signals for his dominant function,  then you can only err 2 to the one side (by marking 10 signals), but its possible to make a very big error to the other side (by marking zero signals), like it happened in this trial.   This will of course influence the average score, so that even if 1000 people were participating, the crowd as a whole would necessary land on an average score much less than the correct one.
Since we will probably never be a 'crowd' of people in the experiment, it might not be of so much importance.  But still I think it might be worth to consider letting a deviation of +2 count as much more than a deviation of -2 in the above example, if you get what I mean.

Hi again,  you have formulated it perfectly clear, so I understand you now.   It sounds like a very good idea to let it be about a 'yes' or 'no' to each signal.
But this was not the idea I got looking at the 'spoilers' in the thread where people had just reported how many signals per each function they had found, but not which signals.  But if each marked signal will be reported from the participants what I wrote is irrelevant.
There is an interesting site that is inspired by Galton and uses yes/no responses instead of scaled answers.  Instead of guessing the weight of a cow, 'the crowd'  makes guesses about the future (but people are not unbiased about the future, so I guess it should be taken more as entertainment)
https://futuur.com/
 
 
 
 
 

I'm not great at the tedium of checking every signal, but he looks FeNi.