Dear Brad Jesness

This article might be useful, have a look:

1. https://usabilitygeek.com/what-is-eye-tracking-when-to-use-it/

2. Article Gaze tracking accuracy in humans: One eye is sometimes better than two

3. https://eyeware.tech/blog/what-is-eye-tracking/

4. Article A review of eye tracking for understanding and improving dia...

5. https://www.frontiersin.org/articles/10.3389/fnins.2021.664490/full

6. https://usabilitygeek.com/what-is-eye-tracking-when-to-use-it/

Kind Regards

Qamar Ul Islam

Dear Brad Jesness

I begin my answer with preliminary or trigger information for the subsequent discussion that I can stimulate, both for and against:

• Reality is built by the individual by interacting with the object of his attention as a researcher.
• The interaction during the subject / object observation; individual / context; It has different dimensions possible to measure, both at the macro and micro level.
• Depending on the measurement of the world to observe; nano or micro or nano, it will be the measurement instrument to be used, which, if not validated, will give erroneous results.
• If the researcher at the time of measuring the phenomenon does not have the appropriate instrument or it has not been validated; then he has to carry out an intermediate scientific investigation of the method and instrument, in order to design and validate both, before the laboratory or field test.
• It is always mandatory to use the statistical test or method appropriate to the type of variables and the number of observations; even to the root design of the research itself; failing that, the results will not be useful or valid.
• When it is handled with quantity of information and more if they come from different disciplinary fields and the phenomena are of different magnitude and quality or quality, it is necessary to do previously investigation of methods, instruments and procedures.
• Following the content of the previous paragraph; Databases already structured in great detail and computer programs based on deep learning artificial intelligence (AI dp) have to be designed; both for the filter of the input information, as well as for the processing and output of data, so that the information obtained is of quality (due to the quality of the input information) and already integrated in the form of indices and indicators . That is, entering the field of information analysis of large databases (BIg-Datas) with AI dp.

Regards

José Luis

Dear Jose Luis

I disagree with your perspective in many ways:

About "different dimensions possible to measure, both at the macro and micro level" :

If one can find a most VALID "level" to have all concepts ultimately in terms of (often, for example, the contents of the present working memory), THEN it is not in any way or to any extent CHOOSING A LEVEL (eschew anything from your pre-conclusions or from the pre-fashioned view(s) of the setting (or of behaviors); get any such out of your mind -- THOSE bad things being essentially arbitrary from the perspective of the behaving organism -- and definitely the case even if you have some major "ax or axes" to grind ).

One CAN NOT have valid logical-and-sound arguments with/between critical individuals without those discussions being consistent on some clear basic same level. Anything else is invalid. Intuition choosing a "level" OR anything (e.g. Re: about the setting) is non-science (as is "putting together" different "levels" or different fields in one's mind).

On the matter of un-validated instruments (and methods): the kind I speak of can self-validate BECAUSE the computer-analysis software is looking for ANY PATTERNING of eye movements within the setting. Any found repeated patterning or reliably found again and again is true (the analysis machine is NOT looking for any particular pattern -- that's why). Researchers, of course, can foul-up the data after the have collected it.

One can look for PATTERNING without in any way biasing or pre-concluding about ANY THING. We, the Researchers, should, in a real and true sense, DEFINE NOTHING because the Subject defines all. Anything not like this is non-science. Looking (via the analysis software) simply for reoccurring PATTERNS has nothing to do with the eye-tracking instrument (except the software's ability to have the analysis capabilities able to find a ANY pattern(s) and then finding it/them again and again). NOR should any research have any assumptions (presumptions), but perhaps a few basic ones : AND, with respect to such assumptions, there is essentially no disagreement regarding the what, where and when with regard to them. Any other "assumptions" (e.g. from a hypothetico-deductive systems or model(s)) would yield Researchers having preconceptions and defile the actual seeing of any true simply-existing PATTERNINGS (we want no misconceptions of what's there in behavior OR in the environment).

About statistics: Notice that many of the Laws and Principles of real science are so well-understood and reliable (that is: reliable and seen the same) unless other identifiable factors can be seen at work. NO STATISTICS ARE NEEDED in good foundational science. Today in Psychology: All statistics are based on groups (just trends of groups vs groups) or GROUPED "separate" data on the same individuals, within the self-same individuals The most important lesson is: it is the "pipe dream" that a scholar and/or Researcher can define things. ALL the statistic "stuff" is actuarial science, exactly like the data for insurance companies, AND not true science.

Truly, that is all it is.

About "quantity of information and more if they come from different disciplinary "fields": the ultimate object of study (and the core foundational unit-of-analysis) is the individual organism (here, human) -- this is all that is truly meaningful. Research that can be integrated meaningfully, meaning that ALL has to be grounded there in the well-researched, discovered, true, singular LEVEL of research --- otherwise you are conflating things AND "defining" them, and many others will not understand.

I do not rely on any existing databases.

All studied should be most-pure observational research -- the observations, I just described about as pure as it can be.

Regarding my Answer (above): I edited to improve understandability up to 14 hours after the first Posting. My apologies. (Unfortunately, this is not atypical for me.)

Dear Brad Jesness

Nor do I agree with your arguments, I see a lot of confusion in them.

You need to become more familiar with what artificial intelligence (AI) and human intelligence (HI) offer. HI never has to be blindly bound by the results of a computer program, even if it is from AI.

The human must always validate the information obtained from statistical programs and those derived from AI; not the other way around. The AI does not validate the HI.

He also needs to know more about objective disciplinary science at both the micro and macro levels; and also to look for evidences in the paradigm of the sciences of the complexity.

I offer you two references to expand your point of view on science.

Respectful greetings

José Luis

http://cursoenlineasincostoedgarmorin.org/images/descargables/Morin_Introduccion_al_pensamiento_complejo.pdf

https://www.researchgate.net/post/Quantum_Physics_Science_method_and_instruments_that_lead_us_to_the_truth_or_only_an_approximation_of_it

https://www.researchgate.net/post/Does_the_observers_gaze_influence_the_measurement_of_the_observed_object_or_subject

Dear José Luis

You say, "The human must always validate the information obtained from statistical programs and those derived from AI; not the other way around. The AI does not validate the HI" True, true, true. AND, I said NOTHING of the KIND otherwise.

My retort now must include saying that using a sort-of AI software DOES NOTHING FOR VALIDATION (or not) -- near nothing one way or another OR ... !!!!! ; but in ANY case, the machine may well be found reliable and may be found to be related to important aspects of environments or situations -- AND, NOTHING EXCLUDES THAT POSSIBILITY.

One could surely worry about persons that think by using a capable machine that can find ALL patterns, that the patterns thereby-found (eye-tracking, of course, also being in response to things in the environment) are somehow insufficient.

Nothing could be more clearly unbiased than just finding patterns (what you say implies the machine is biased). Unfortunately, modern psychologists typically are NOT scientists because they want to, THEMSELVES, do the "defining" (not really the biological organism (biological in the broadest but appropriate sense)); THEY define what the stimuli are; and, they want to set up the "measure" of what the response(s) are (that is, the "official, important response(s)") is/are) In real science THE SUBJECT defines everything and you should NEVER preclude that by doing the defining yourself. Just look at All OTHER ESTABLISHED SCIENCES (physical sciences). In my writings I provide a general definition of science that is true of ALL strictly empirical science (aka simply: "science")(you should read a LOT of me (my many. many papers and essays (all available on RG); I am near my end and such as me may not come around again (not because I am so smart, but because I am not system-"slotted"))).

P.S. I know of everything you discussed and have used all such myself. Problem is ALL that is of limited value (as well as it is NOT doing real science -- it is rather actuarial work (like that work done for insurance companies) (< -- good for some things, but not good for a continuing, growing science -- THUS, IT IS NOT GOOD SCIENCE, and really not science at all). You know you do not have a true science when progress depends on formulating major parts [ THINK: models ] of which are just from human minds (that, for sure, is something NOT TO TRUST). In science, THE SUBJECT DEFINES __ALL__ (I do mean ALL; well-defined behavior PATTERNS are defined by the settings they emerge in AND ALSO always: the surrounding behavior PATTERNS. If you do not have such, you do not have science.)

Western man is so darned messed up by quick (or often even a priori OR largely a priori) hypothetical-deductive systems (e.g. models) to plug reality into; THIS IS NOT SCIENCE -- it could be, but only with the stipulations I described.

Respectfully,

Brad

Again, I did editing for emphases 15 hours after initial answer,

Dear Brad Jesness

I know that he is a dedicated academic and that he seeks the truth about a great deal of information.

Nobody has the whole truth or knows everything there is to know about the external world and our inner universe.

• My reply in synthesis translates into:
• Learn the method in order to learn for life: Being, knowing/ to knowing and knowing how to do,
• The path of learning is endless: The scholar only knows a part of the whole and often knows a great deal about nothing. The sage knows that he does not know enough, but he aspires to know it in the course of his life.
• AI codes and algorithms are very useful, but so far, there is no evidence that they are smarter than the human who programs them.
• On the other hand, diseased and disembodied AI has no chance of reaching the emergent singularity of consciousness.
• The computer programs that man designs and elaborates, always have to be validated and find their reliability before using them in the information processing, which will finally be digested and assimilated by the human
• What is feasible with AI within the cyberspace of the Internet, is to function as an auxiliary or prosthesis of the subjects, expanding the possibilities and capabilities of human intelligence.

Greetings and success in your search

José Luis García Vigil

Dear José Luis García Vigil

You have missed my (and THE) major point; you do not have a way to see rightly (apparently), to see phenomenon, unencumbered -- and that is essential to science. You , frankly, seem incapable of "Learn[ing] the method in order to learn for life" -- , to have true self-correcting guidance through one's procedures:

Dear Michelle Morelli

My papers and essays quite clearly PROVE that most Psychologists ARE NOT SCIENTISTS (and I am even talking about those in academic institutions, doing research). Basically, the most of them just want to claim to be scientists BUT can only cite insufficient signs; with respect to what they do, THEY ARE NOT SCIENTISTS (in short, most are just "in church").

Apparently Psychologists cannot accept that "wishing does not make things so".

About "thorough and rigorous"; most psychology is those ways in ONLY the same sense as actuarial 'science'. Read my stuff to learn somethings different from what you 'know'. (Those supposedly "in the know" are not ; AND they are NOT "in the know" in very, very similar (if not exact-same) ways.)

There seem to be ONLY a handful of persons on Researchgate that think for themselves. (I could poke any undergraduate and get the same (or possibly better) responses, as ones I get here.) Provided (HERE, from others) are essentially only claims and assertions, not in any strict empirical ways testable.

Again, Michelle, YOUR "proof", that "proof", is as it is in actuarial 'science': not true science, because it can not be continuously and progressively becoming better (rather, it all dead ends).

Brad, I'm not sure if I understand your question correctly. However, as for the inference of personal information from eye tracking, you might be interested in our recent review paper: Chapter What Does Your Gaze Reveal About You? On the Privacy Implica...

For a summary of our findings, see also this twitter thread: https://twitter.com/JL_Kroger/status/1392789775569018881

To appreciate some major progress in the direction towards being able to "see" perceptual shifts in the Subject's attention, see "Looking ahead in working memory to guide sequential behaviour" Here's the link: Article Looking ahead in working memory to guide sequential behaviour

It is a VERY BRIEF summary of this recent work using advances in eye tracking.

Also see, "Multiple spatial frames for immersive working memory" for more relevant recent research (again, with new technology): "Multiple spatial frames for immersive working memory" Link:

Article Multiple spatial frames for immersive working memory

If one wanted to do research on the KEY developmental perceptual (perceptual/attentional) shifts I hypothesize, these pieces of research (pointed to the last 2 Answers I just posted) would certainly (at least partly) lead the way. (This research also proves the point I tried to make in the Question that begins THIS thread.)

Also, quite possibly relevant: "Gotcha: Working memory prioritization from automatic attentional biases" June 2021 Psychonomic Bulletin & Review , Abstract link:

Article Gotcha: Working memory prioritization from automatic attenti...

( NOTE: Do NOT think, or get the idea that, I can do all the work for you. I can't )

Dear José Luis García Vigil

I do value your participation and perspective, and realize that it is most agreeable to present behavior 'scientists' (you ALWAYS get more "likes" than me). BUT, as rather commonly recognized: there is NO WAY to define ANY INSTANCE of present behavior 'science' as strictly empirical in its definitions even "from the 'get-go'" -- i.e. even from the very beginning of the/any instance of 'science' work done. Some of the most respected psychologists "see" something of this kind of major problem; they somewhat-recognize/somewhat-and-variously-describe such an over-arching problem, but do not fully appreciate the basic problem and describe NO reasonable or decent way out of "the situation". Psychologists turn (VERY commonly) to "interdisciplinary studies" for the "answers" (which is not consistent with the individual organism ALWAYS BEING THE UNIT OF ANALYSIS -- a fatal problem). AND VERY, and even more commonly: the other main proposals to get out of this 'situation' is to have neuroscience (brain science) seemingly support us or correct us (I have essays where I address this, also a VERY unreasonable (irrational) way of seeing how "things will come together"; it is actually not only irrational BUT will not work (which would always be the case with the irrational); this present position of A LOT (maybe even the majority) of psychologists, in reality, has the status of a myth -- and it is impossible to show otherwise and to even "understand" WITHOUT RATHER EXTREME, irrational presumptions in from-the-start views) .

José Luis García Vigil, our major different positions (one vs one) very much relate to this ALL this, just described.

Our "arguments" or what I see as the basis of (and for) our general (over-arching) argument is basically (and put simply -- as it must be presently) : The extent to which we have to see data in a our-defined or a predefined way. My view is that ALL behavioral patterning can potentially be extracted (actually : abstracted) from raw data (including recognition of irrelevant behavior patterns OR of behavior patterns not-of-present-interest-or-concern -- all of which may well require an actual knowledge and understanding of development (ontogeny)(now, basically non-existent)). My position makes behavior science an inductive science almost exclusively, which I believe it must be to just " 'see' the 'landscape' " of any consistency in behavior patterning (I think it is VERY TELLING that the word "patterns" (as in 'behavior patterns') is almost never used -- this , I see, is an indication of how far from reality behavior 'sciences' are). My position is that hypothetico-deductive reasoning only should enter into it IF YOU CAN SEE NO OTHER WAY TO GO (and this is damned rare (or could be AND should be -- but maybe not in a "laboratory situation",

It may be helpful for those who read my writings to know that the VAST MAJORITY of the essays and papers are ALL part of ONE clear major perspective and approach: To wit: These writings are fully and _completely_ internally consistent AND, as far as anyone can know, consistent with physical reality, with grounded, demonstrable relationships/relations, and correspondingly with strictly empirically-grounded relationships between concepts, and such. (i.e. altogether valid -- at least testable (provable/disprovable) as such); THUS ONLY good hypotheses come out of all the words. I am saying these (my) many 100s of pages of writings are a magnum opus (my magnum opus), with the main goal of establishing an outlook, and pointing clearly to the evidence, for a STRICTLY EMPIRICAL perspective and approach to Psychology, to establish a real hard science FOUNDATION for/of Psychology. -- Algernon