How You Feel About How You Feel About How You Feel About the Job (Candidate): a ‘Parity Check’ on the Mind

That's not a valid work email account. Please enter your work email (e.g. you@yourcompany.com)
Please enter your work email
(e.g. you@yourcompany.com)

 You’ve got a good feeling about the job (candidate). This fit seems right, good. But feelings, like bits of evidence, are not always reliable.

Sometimes, if not always, the smart thing is to hit the pause button and ask yourself how you feel about that good feeling, especially whether you feel you can or shouldtrust it.

(Old joke of mine: From the standpoint of the father of a teenage girl, there are two kinds of teenage boys: those she can’t trust and those she shouldn’t. It is to be hoped that, in general, and with respect to trustworthiness, our other feelings offer us a broader menu.)

Of course we can have feelings about our feelings, just as we can have information about our information, e.g., that the source of our information is or isn’t reliable, or awareness of our awareness, e.g., knowing that we just got a whiff of Chanel.  Such common meta-observations can be very illuminating and useful. But they pose some unique challenges and quirky puzzles.

 Two such challenges and puzzles are these:

1. How many levels (of reflection, observation, information, etc.) do we have to employ in order to assure the reliability of the lowest level feeling, information, observation, etc.? Is it enough to trust my initial feeling because I trust my feeling about that feeling? Or do I have to add at least one more level—my feeling about my feeling about my feeling (like a jittery, doubt-plagued groom on his wedding day)?

2. How many levels are our brains capable of handling? Presumably, your dog can manage one; you, at least two; your child, probably not more than, but maybe, two. Advanced aliens or God? Only they know. But the more levels that can be generated, recognized and applied, the greater the self-awareness of sentient beings and the stronger the claim to and form of “personhood”—claims that some forms of AI (artificial intelligence) software will probably be making for themselves very soon.)

(Note: If your dog “feels guilty” about chewing your socks, it would be a feeling about an action, and therefore would not count as “a feeling about a feeling”, despite possibly being meta-level in some sense, which it may not be, since it could just be “guilt-like” behavior in anticipation of punishment).

Can Your Dog Meta-Feel?

Start with the second question first: Suppose that a mix of meta-feelings and initial feeling is like a multilevel set of quality control or airport security checks—as confirmatory checks on a check (either as repeat checks or as qualitatively different ones added on). Further suppose that it can be proven that a 3-level response to anything is sufficient to assure reliability of judgment or information with a 97% probability (this, at this point in the argument, being a purely hypothetical, illustrative example—for both feelings and airport screenings. However, something like these numbers will be argued for below).

What if it were also proven that most people can’t manage more than two, i.e., they are capable of having and identifying their feelings about their feelings—but utterly incapable of experiencing, much less describing any feelings about their feeling about their feeling that the job (candidate) is a good fit or 4th, 5th, etc., levels of feelings or awareness.

To suggest that they must be able to have x-number of feelings about feelings just because we all understand what the words “feelings about feelings about a feeling” mean is mistaken. That’s because even if each and every word in an “about-based”  concept, phrase or sentence is understandable, and even if the concept as a whole is comprehensible, this is not enough to guarantee that the process can psychologically be performed by a given brain or that experience of it can occur and be recognized.

Example: Having to mentally or physically “rotate the center of a circle about the center of a circle aboutthe center of a circle” (shown in the accompanying image) is likely to be beyond the performance or imaginative capabilities of many, if not most, young children, even though the instruction is perfectly meaningful English.

Even if we allow that your feeling about your feeling about your feeling about the job (candidate) makes sense logically or linguistically, that is not sufficient to guarantee that the experience itself is possible for anyone or everyone, e.g., for your dog—or do you imagine your dog has feelings about his feelings about the kibble? 

Taking this skeptical line of argumentation to its logical conclusion, it may very well be the case that even if n-levels of observation or information (as a form of confirmation) may be required, they may not be psychologically possible for some or all people, and that, brains being what they are, there has to be an upper limit of meta-observing, e.g., four, five or fifty levels of reflective awareness, depending on the complexity of the brain or mind doing the reflecting.

As suggested above, presumably, the greater that number, the greater the (self-)awareness and reliability of judgment of the mind in question.

Parity Checks on Your Mind

I suggested three levels as hypothetically guaranteeing reliability of an initial feeling, by means of a feeling about a feeling about that first “object-level” feeling.  In fact, a case can be made for precisely that number—three—as a reliability check, at least in terms of basic information theory and one of its core concepts: a “parity check”—which in addition to shedding light on the meta-feelings question in hand, also explains why so many signals come in threes, e.g., “Ho, ho, ho—Merry Xmas!”, “three on a match is bad luck”, three strikes in baseball, or “Hear ye! Hear ye! Hear ye!..The court is in session.”

The concept of a parity check is as simple as it is powerful: Let’s say I want to send you a simple message—the simplest possible, namely, either a binary “0” or a “1”. So, I send it over a line. If there’s an error in transmission, and you receive a “0” instead of a “1”, you won’t be able to detect the error.

But suppose, for safety’s sake, I add a copy of my intended message to the original. If I want to send you the intended “1” and to confirm there’s no mistake, I’ll add a second “1”. I hit “send” and off goes “11” to you, so that if you receive “11” you’re sure there’s no mistake (on the assumption that 2 mistakes in one message just don’t happen) and that my primary message is “1”, confirmed by the second “1”.

However, if there is a mistake and you receive “10” or “01”, you know that you can’t trust the message, that there’s an error, because the two digits are not the same. The problem is, you don’t know whether the correct message is “11” or “00”, i.e., fundamentally a “1” or a “0”.

To ensure that you not only can detect an error, but also identify it, I add a third signal, e.g., a third “1”. Now, even if there is one mistake, you will be able to identify it—as the “odd man out”. For example, I send “111”, but you receive “101” or “011” or “110”. You’ll know for certain that my primary message was “1” and that the transmission can be trusted, because of the two additional (meta-) digits, the second confirming the first, and the third confirming both. (Note: This is but one kind of parity check. For common alternatives, which include checks on the number of bits, check out “even parity” and “odd parity” bit checks. )

The gist of this is that to ensure reliable transmission of one bit of information, three bits are required, as a parity check. In psychological terms that can be applied to the phenomenon and requirement of meta-levels of feeling or information, the three bits can be redescribed, in sequence as, “stimulation”, “identification” and “confirmation”.

The first “1” is the orienting stimulus—the attention-getting “wake-up call” (like your first impression of a job or job candidate). The second “1” represents the identification stage: Having detected something, the next thing we do is identify it; the final digit or signal serves as confirmation. To make this clear, the well-known combat maxim, “Three on a match is bad luck” will suffice.

More Magical Threes

In trench warfare, it is understood that worse than lighting a cigarette with a match is passing the match to another soldier. Worst of all is to pass that match to a third comrade in arms. Why? An enemy sniper detects the sudden flash of a match (detection step). When the match is passed in a smooth arc to the second soldier, the sniper tentatively identifies it as a match rather than a firefly. Still, he needs to be sure. When the match is passed to the third, unlucky infantryman, the sniper confirms it is a match being passed, shoots and completes the 3-level, 3-step process.

A child hears “Ho!” in a shopping mall (detection). Hearing a second “Ho!”, the child tentatively identifies it as being emitted by Santa Claus. Upon hearing the third “Ho!”, the child confirms it and races into the toy section of the department store to line up with the other kids waiting their turn to prep Santa. Such examples and this analysis can be multiplied endlessly. (Or do you need threedetailed commonplace illustrations to be convinced? If so, go back and look at “three strikes, yer out!” for starters.)

What are the implications for your feelings, observations, judgments and information about a job (candidate)? Should you, as a minimum that reflects the value of a parity check, check your feelings about your feelings about the job or job applicant? When you ask for or provide references, should you ask for or provide references for the references? (Why isn’t that ever explicitly done? Perhaps, in this regard, we trust websites, hearsay, phone books and media too much.)

Gut vs. “But…”

This parity check model of meta-evaluations clearly clashes with an “old school” “gut instinct” approach to assessment: “I’ve got the guts to go with my gut feeling about this one.” What would serve as extraordinarily valuable research is an investigation into the comparative reliability of this kind of “intuition” and gut instinct vs. 3-step parity check meta-evaluations.

My gut instinct tells me that any research like that would either have to be repeated at least twice or be the subject of at least two meta-studies (the highest-level meta-study, of course, being about all the other meta-studies).

….assuming that I can trust my gut instinct about my gut instinct.

By Michael Moffa