Emotional Intelligence

Miles: Nineteen ninety five. A science journalist publishes a book arguing that the thing we've spent a century measuring, the thing schools sort children by and employers screen for, accounts for a modest fraction of what determines how a life goes. And that another set of capacities, mostly unmeasured, matters more.

Miles: Within ten years the phrase is sitting in job postings, in leadership programmes, and in school curricula on several continents. Very few ideas in this field have ever been taken up that fast. And the research underneath the phrase is still being argued over now.

Miles: Today: Emotional Intelligence, by Daniel Goleman. In two sentences: recognizing, understanding and managing emotion, in yourself and in the people around you, predicts how a life turns out at least as well as cognitive ability does. And unlike IQ, the claim goes, it can be taught.

Miles: Sixty Books. Presented by Stoar Consulting and Coaching.

Miles: This is Sixty Books, episode twenty five, and I'm Miles. One management book each time out, sixty in total, pulled from the last hundred years, and each of them gets the same two questions put to it: what's in this for leaders and teams who think differently, and what's left of it now that AI is doing part of the job. The research is Nora's.

Miles: Nora, yesterday took us half an episode to keep two states apart. There's another pair like that waiting in here.

Nora: Flow and hyperfocus divided on how you get in and how you get out. This book's pair is caring what another person feels, against looking as though you do, and the version of it that operates in workplaces grades the looking. That's our second half.

Miles: And the first half is unusual for us, because the evidence here isn't thin. It's genuinely mixed, and the trouble starts at the point where a mixed finding turns into a hiring requirement.

Miles: Start with the model.

Nora: Goleman's framework has five domains. Knowing your own emotions, which is self awareness. Managing them, which is self regulation. Motivating yourself. Recognizing emotions in others, which he calls empathy. And handling relationships.

Miles: Where does it come from?

Nora: The academic construct predates the book. Peter Salovey and John Mayer published on emotional intelligence in nineteen ninety, defining it as a set of abilities for perceiving, using, understanding, and managing emotion. Goleman's contribution was popularization, and a substantial broadening.

Miles: Broadening how?

Nora: Salovey and Mayer proposed an ability, something you could in principle test with right and wrong answers. Goleman's version absorbed motivation, optimism, persistence, and a range of personality traits. That expansion is what made it useful to corporations and what made it hard to defend scientifically.

Miles: Mm-hm.

Miles: And the famous claim.

Nora: That emotional intelligence matters more than IQ for success, with various percentages attached in the popular retellings. The record has no clean study behind those numbers, and Goleman has publicly separated himself from the strongest versions of them.

Miles: What's the neuroscience piece?

Nora: The amygdala hijack, his term for the moment when an emotional response outruns deliberate thought. He draws on Joseph LeDoux's work on fear pathways. The underlying research is real. The picture of a hijack, with an emotional brain wresting the controls off a rational one, is the triune brain story once more, and we took that apart in episode eighteen.

Miles: Right.

Miles: Who wrote it?

Nora: Daniel Goleman, born nineteen forty six in California. He has a doctorate in psychology from Harvard, then spent twelve years writing about behavioural science for a major newspaper. So a trained psychologist doing the job of a journalist, which is an unusual pairing and mostly a useful one.

Miles: And afterward.

Nora: A long run of follow up books, and consulting work built around measuring emotional competence inside organizations. Which is the thing to keep in view all episode: there's a large commercial market sitting on top of this construct.

Miles: There's a business application chapter that became the whole industry.

Nora: He later argued that emotional competencies distinguish outstanding performers from average ones, especially in senior roles, and that the higher you go the more they matter. That claim, in various numerical forms, is the engine of the corporate training market around this idea.

Miles: And its basis.

Nora: Competency modelling studies run largely inside companies, a good deal of that work published by firms that also market emotional competence assessments, with the organizations' own performance ratings used as the outcome. Performance ratings are themselves substantially social judgments, which means the finding may partly be that people rated as socially skilled get rated highly.

Miles: A circle.

Nora: Yeah. A circle with a certificate at the end of it.

Miles: Closed loop, neatly bound.

Miles: And what came apart later?

Nora: Three problems, and they compound.

Miles: One.

Nora: The construct fragmented. There are now at least three distinct things called emotional intelligence. An ability model, tested with performance items. A mixed model, Goleman's, combining abilities and traits. And trait models measured entirely by self report. They correlate only moderately with each other, which means two studies using the phrase may be studying different things.

Miles: Two.

Nora: Incremental validity, which is the question of whether it predicts anything beyond what we could already measure. Once you control for cognitive ability and standard personality traits, particularly conscientiousness and emotional stability, the additional prediction from mixed and trait measures is modest. That's the same finding we hit with grit in episode nine, and for the same reason: a new construct that overlaps heavily with an old one.

Miles: Three.

Nora: Self report. A large fraction of workplace emotional intelligence assessment asks people to rate their own emotional skills. The people least accurate about their emotional impact are, by definition, unlikely to know it. And a great many of these instruments reach the buyer inside a training package sold by whoever developed them.

Miles: Is there a version that survives?

Nora: Yes, and it's worth saying. The narrower ability model, particularly emotion perception and emotion understanding as measured by performance tests rather than by self report, holds up better and shows real if modest relationships with outcomes. The version that survives is much smaller than the version that sells.

Miles: Huh.

Miles: So where does the book land?

Nora: A genuine research construct, popularized into something considerably larger than the evidence, with an assessment industry attached, and a defensible core hiding inside it. And this is a book written for practitioners and not a peer reviewed finding, which is worth holding on to, because the popularized version is the version that got quoted. Read it and decide for yourself which of the three versions of the idea you're being handed.

Miles: Here's the turn, and it's the episode where I expect the most disagreement from listeners.

Nora: Because emotional intelligence, as it gets deployed in workplaces, is one of the most effective mechanisms for excluding autistic professionals that exists, and almost nobody using it intends that.

Miles: Walk through the mechanism.

Nora: A job description says strong emotional intelligence required. What actually gets assessed in the interview is a set of observable behaviours. Reading the room. Matching the interviewer's affect. Fluent small talk. Eye contact. Picking up on unstated cues.

Miles: None of which is emotion.

Nora: Yeah. None of which is the underlying construct. They're the conventional expression of it in one neurotype. A person can care enormously, read the situation accurately, produce none of those signals, and be scored low on emotional intelligence.

Miles: And there's a research point that goes further.

Nora: Damian Milton's double empathy problem, from twenty twelve, and we've come back to it more than once in this run. It reframes the breakdown as mutual. Autistic and non autistic people both misread each other. And there's experimental work in that tradition finding that autistic people pass information to each other about as effectively as non autistic people do among themselves, with the loss appearing in the mixed pairs.

Miles: Which means the deficit is in the pairing.

Nora: And emotional intelligence testing measures accuracy at reading one group's signals, then reports the result as a property of the individual. Non autistic people aren't tested on their ability to read autistic expression, and if they were, the scores would look different.

Miles: Say more.

Nora: There's a specific clinical piece too. Alexithymia, which means difficulty identifying and describing one's own emotional states. It occurs at elevated rates in autistic populations and it's a distinct thing from autism itself, and that matters, because a good deal of what gets attributed to autistic empathy differences may track alexithymia instead.

Miles: Why does the distinction matter practically?

Nora: Because alexithymia sits in the internal labelling channel and not in the caring. A person may feel something strongly, act on it appropriately, and be unable to name it on demand. Goleman's first domain, knowing your own emotions, gets assessed by asking people to describe their emotions in words. That's a naming test.

Miles: So what survives for us?

Nora: A great deal, if you keep the goal and change the assessment. Emotion regulation is real, learnable, and useful for everyone. Working out what other people are likely to feel and want is real, and it can be reached analytically rather than intuitively, which is how a lot of autistic professionals do it, with results that are frequently better because the reasoning is out in the open.

Miles: And the leadership instruction.

Nora: Assess the outcome, not the channel. Does this person leave colleagues feeling respected. Do they handle conflict without damage. Do they notice when something's wrong and act on it. Those are answerable from behaviour over time. Whether somebody held eye contact through a forty minute interview isn't the same question, and it's the question that actually gets asked.

Miles: Now the AI end of it.

Nora: Two directions, and one of them is genuinely new. One piece of admin before either, though: the standing caution about tools like these is read out after the credits now, in the same words every episode, so I won't score it a second time in here.

Miles: Then take the useful direction.

Nora: These tools are unusually good at explicit social reasoning. What's this message likely trying to convey. What might this person be worried about. How will this land on somebody in that position. That's the analytical route to the same destination, available on demand, and for a person who reaches social understanding by reasoning rather than by intuition it's close to a prosthetic for a channel that was never open.

Miles: And it works in both directions.

Nora: It does. Translating an intended meaning into the register the recipient expects is the other half, and it's the half that reduces the daily cost of being misread.

Miles: There's a third use, going by segment four.

Nora: The naming channel, and it's the one I'd hand to anybody who recognized that alexithymia description. If you can feel a state and can't label it, then something you can reason out loud with, at any hour, with nobody watching, is work on exactly the domain Goleman put first. And it happens without a room full of people waiting for the word.

Miles: And self regulation.

Nora: Same logic from the other end. His hijack chapter is about the gap between the reaction and the reply, and everything in this category widens that gap, because drafting a response is slower than saying one. That's not a small thing for anybody whose difficulty is timing rather than intent.

Miles: Now the concerning direction.

Nora: Emotion recognition systems in hiring and performance management. Software that scores facial expression, vocal tone and word choice during an interview. That technology is being sold, its scientific basis is contested, and it automates precisely the bias we've spent this episode describing. If a human panel misreads autistic affect, a model trained on human judgments will misread it identically and at scale, with a number attached that reads as impartial.

Miles: And that's a policy question.

Nora: It is, and it's worth a leader asking whether anything in their hiring process scores how a candidate appears rather than what a candidate produces.

Miles: Fair enough.

Miles: Last word on it.

Nora: The book's own promise is the thing to hold all of it to. Goleman argued these capacities can be learned, and that's the claim that made the idea travel. Anything that leaves a person more able to do this themselves is inside that promise. Anything that scores a face and files the number has quietly abandoned it.

Miles: That's Emotional Intelligence.

Miles: Next time: Thinking, Fast and Slow, by Daniel Kahneman, and the public correction a Nobel laureate published about one chapter of his own book.

Nora: A prize winner telling readers not to trust a chapter he wrote is the rarest item on this shelf. I'll bring the replication record with me.

Miles: Nora, thank you for the care on that one.

Nora: Bring the next one, Miles.

Miles: That's us for today.

Miles: This show belongs to Stoar Consulting and Coaching. Logan Williamson is a Licensed Professional Counselor in Texas and Louisiana. Treat what you just heard as education about a single book, put together for working leaders and for the people on their teams. Nothing in it coached you, nothing in it counseled you, and it offered no opinion of you.

Nora: Under every claim there's a citation, and the page for this episode holds them all, transcript included. Book twenty six lands tomorrow.

Miles: Now the disclosures. These are two synthetic voices and nothing besides. There's no person here, there's no license here, and no part of what was said was aimed at you. The drafting uses AI assistance; the edit and the approval come from Logan Williamson, a Licensed Professional Counselor in Texas and Louisiana, at Stoar Consulting and Coaching.

Nora: And one note about the AI segment. The evidence that these tools help neurodivergent people at work is early. There are controlled trials now, but they measure things like conversation skill rather than how a job actually goes, and for work itself most of what exists is individual providers reporting success. So take it as promising and not proven. The tools also get rebuilt every few months, so whatever you set up today is built on something that moves.