The models aren’t wrong. But knowing them hasn’t changed most people — not in the body, not in the room, not in the moment when it matters. Here’s why, and what actually does.
The major models of EI are not wrong. They are genuinely useful. They have given millions of people a vocabulary for something that previously had no name, a framework for noticing what they couldn’t previously see, a language for conversations that couldn’t previously be had. That is real, and it deserves to be said plainly before anything else.
But here is what is also true: knowing the framework has not changed most people. Not in the ways that count. Not in the body, in the room, in the moment when it matters. And if you have spent time with the EI literature — if you can explain self-awareness, self-regulation, empathy, and social skill as fluently as any leadership seminar — and something still isn’t working the way you expected it to, the problem is not that you haven’t found the right model yet.
The problem is that understanding a model and being changed by it are two completely different things. And the field, by and large, has been far more interested in the first than in the second.
A brief and honest account of the models
The academic history of emotional intelligence begins in 1990, with psychologists Peter Salovey and John Mayer, who defined it as the ability to perceive, use, understand, and manage emotions — in oneself and in others. This was a rigorous, carefully bounded definition, and it remains the most scientifically defensible one.
What most people know, however, is not Salovey and Mayer. It is Daniel Goleman, whose 1995 book brought EI to a mass audience and expanded the construct considerably — adding motivation, empathy, and social skill alongside self-awareness and self-regulation, and making claims about its importance to life outcomes that the academic literature has since treated with considerable scepticism.
Reuven Bar-On offered another model, organized around emotional-social intelligence and measurable through his EQ-i assessment. The trait EI model, associated with Petrides, takes yet another approach — treating EI as a personality characteristic rather than an ability, measured through self-report rather than performance tasks.
These models disagree with each other in meaningful ways. Researchers have argued about them for decades. Whether EI is a genuine intelligence, a set of competencies, or a cluster of personality traits remains genuinely unsettled in the academic literature.
And here is the thing: for most purposes, it doesn’t matter.
What frameworks are for — and what they aren’t
A framework is a map. Maps are extraordinarily useful. They show you the terrain, name the landmarks, help you orient when you’re lost. A good map of emotional intelligence — and Goleman’s, whatever its academic limitations, is a genuinely good map — tells you what to look for, where the important territories are, what the major features of the landscape are called.
But a map does not move your legs. It does not train your body to walk the terrain. It does not prepare you for the altitude, or the weather, or the specific difficulty of the path you’re on. You can study a map of a mountain for years and still find yourself underprepared when you actually climb it.
This is not a failure of the map. It is a category error to expect a map to do what only the climbing can do.
The popular EI models are maps. The problem is not that they exist, or that they’re imprecise, or that researchers argue about them. The problem is what happens when understanding the framework becomes the point — rather than the beginning of something harder and more useful.
The competency frameworks, the assessment instruments, the five domains, the four branches — these are ways of organizing territory so it can be understood and discussed. They do their job. What they cannot do is close the gap between knowing and being. That gap is where most people live, and it is where the real work begins.
The gap the models don’t cross
Here is where most EI education stops, and where the actual work begins.
You can learn — genuinely learn, not just memorize — that emotions contain information. That anger points toward a value that has been violated. That anxiety points toward something perceived as threatened. That the body registers emotional experience before the mind has words for it. You can understand this clearly enough to explain it to someone else.
And then a colleague sends an email that lands as dismissive, and you are reactive for the rest of the afternoon — acting from irritation you haven’t named and wouldn’t quite admit to if asked. The knowledge did not make it to the body. The understanding did not change the response.
This is not a personal failing. It is what happens when insight travels only as far as the intellect. The emotional system — the nervous system, the body, the fast and largely automatic processes that generate feeling before thought has a chance to catch up — does not reorganize itself in response to conceptual understanding. It reorganizes in response to experience. Repeated, embodied, emotionally present experience.
This is the gap the models don’t cross, because crossing it was never their job. Mayer and Salovey set out to define and measure emotional intelligence, not to develop it in people. Goleman set out to communicate its importance to a broad audience, and largely succeeded. Bar-On set out to assess it. These are legitimate and valuable contributions to a field that matters.
But none of them are in the business of the slow, effortful, often uncomfortable process by which a person actually develops the capacity to work with what they feel — in the body, in real time, in the moments when it is most difficult.
One happens in the mind, through reading and reflection and conceptual grasp. The other happens in the body, through repeated practice, lived experience, and the slow accumulation of a different relationship with what you feel. Both matter. But they are not interchangeable — and mistaking the first for the second is how intelligent, well-read people end up knowing exactly what emotional intelligence is, and still finding themselves stuck in the same patterns.
What actually changes people
The research on what actually develops emotional capacity points consistently in a direction the popular frameworks tend to underemphasize.
The body comes first. Emotions are not primarily cognitive events — they originate as physical sensations, shifts in the nervous system, changes in heart rate and breath and muscle tension, before they become feelings the mind can name. Any approach to developing EI that doesn’t engage the body is working from the second floor up, with no foundation beneath it. Somatic practices, body-based regulation tools, and deliberate attention to physical sensation are not soft additions to rigorous emotional development. They are the ground floor.
Repeated practice in low-stakes conditions matters more than information. The nervous system learns through experience, not through concept. The person who takes a few minutes every morning to notice what they feel in their body and name it with some precision is building something real — slowly, incrementally, in a way that shows up differently when the stakes are high. The person who has read every EI book published in the last thirty years and never done anything like that has not built the same thing.
The quality of attention you bring to your inner experience is the mechanism. What changes the nervous system’s response to difficulty is not knowing about the difficulty but being present to it — repeatedly, in conditions where it can be tolerated. This is why self-compassion is not a soft add-on to emotional development but a genuine prerequisite. The inner life only becomes available as a resource when the relationship to it is not primarily one of judgment and management.
The work is relational. Emotional capacity doesn’t develop in isolation. It develops in contact — with a coach, a therapist, a trusted colleague, a partner willing to stay in the difficulty with you. The models tend to frame EI as an individual competency. The experience of developing it suggests it is better understood as something that grows in relationship, requires witness, and deepens through the specific challenge of staying present to another person while also staying present to yourself.
The question worth asking
So: which model of emotional intelligence is most accurate? Is Goleman right, or is the Mayer-Salovey-Caruso model more defensible? Is EI a genuine intelligence or a personality trait? Does the EQ-i measure what it claims to measure?
These are interesting questions. Some have reasonably clear answers. None of them are the question most worth asking.
The question most worth asking is simpler and harder: what would it take for this to make a real difference?
Not a conceptual difference. Not a difference in vocabulary or framework or the ability to give a more sophisticated answer when someone asks what emotional intelligence actually is. A difference in the room. In the relationship. In the moment when the body floods and the old response is right there — available, familiar — and something else, something slower and more deliberate and more true, is also possible.
That difference is not delivered by a better model. It is built, gradually and imperfectly, through the kind of practice that most EI content gestures toward without quite getting to.
The switch is where it always was. The question is whether you’re willing to use it.