What is emotional intelligence: the real model with four branches

Sleeping eight hours does not guarantee you will rest. Being able to read does not guarantee you will understand a poem. With emotional intelligence, something similar happens: the phrase “I have emotional intelligence” floats around in conversation, self-help books, and course offerings, and by now it barely means anything. Reading a book about it is one thing; having the right model is something else. The confusion is dangerous because it mixes two things psychology separated 35 years ago —and trying to improve yours with the wrong strategy is like training for a marathon by working only your arms.

This piece does four things. First, it fixes the original technical meaning —the one Salovey and Mayer published in 1990 when they coined the term10.2190/dugg-p24e-52wk-6cdg— and separates it from the popular versions that drifted later. Second, it shows the four branches of the model, because that is where the difference between “being empathetic” and “knowing how to use emotional information to make better decisions” lives. Third, it explains why EI gets confused with personality, kindness, and empathy —and why that confusion costs you real results. Fourth, it leaves you with a workable distinction so you can tell whether what you have is EI, character, or something else the culture called EI but isn’t.

1. The uncomfortable fact: we confuse EI with three different things

Before going into the model, look at the background data. The popularity of emotional intelligence grew faster than serious research on it. By the late nineties, the concept was already being used to sell books, corporate trainings, and home tests that mixed cognitive abilities with personality traits and social competencies. In 1999, Mayer, Caruso, and Salovey published an explicit argument for why that mix was a mistake: emotional intelligence, they said, does not meet the criteria if it gets confused with kindness or personality —it only meets them if treated as an intelligence in the strict sense10.1016/s0160-2896(99)00016-1.

The distinction matters because it changes the strategy. If you believe emotional intelligence is “being a good person,” the training becomes moral: be kinder, more helpful, more smiley. If you believe it is “not feeling,” the training becomes suppressive: hold it in, swallow the anger, show up neutral. Both are wrong ideas, and both fail when put to the test.

What psychology made clear: emotional intelligence is a cognitive capacity. It is the ability to reason about emotional information and use that information to guide thought and behavior. It is not good or bad in itself: like mathematical or verbal intelligence, it is a tool that can be applied in any direction.

2. The technical definition: the four branches of Salovey and Mayer

The original model, called the ability model, organizes emotional intelligence into four hierarchical branches. Each branch is a distinct skill, and together they form a scale: the higher ones depend on the lower ones.

Branch	What it does	Everyday example	Question it answers
Perceiving emotions	Detect emotional information in yourself, in others, in images, in the body	Notice you are clenching your jaw before knowing you are angry	What is happening here?
Using emotions	Use emotional state as information for thinking and deciding	Use a state of high activation for a creative task	How does what I feel help me?
Understanding emotions	Understand how emotions combine, evolve, and transform	Know that guilt sometimes hides anger, or that sadness can hide relief	What exactly does what I feel mean?
Regulating emotions	Modulate intensity and duration of emotions in yourself and others	Lower anger without suppressing it, amplify curiosity when useful	What do I do with this?

The hierarchy matters: you cannot regulate what you do not perceive. People with poor emotional vocabulary try regulation techniques that first require identifying the emotion —and that is why they fail. It is not lack of will: the exercise asks for a prior capacity that was never developed.

Each branch is a measurable skill. Standardized tests exist (MSCEIT, among others) that assess each one separately. That measurement is what allows researchers to claim emotional intelligence exists as a construct rather than a slogan.

3. Why it gets confused with empathy (and the useful difference)

Empathy is the capacity to perceive and understand what another feels. That is: empathy lives almost entirely in the first two branches —perceiving and part of understanding—. But emotional intelligence does not reduce to that.

The difference shows up when decisions need to be made. Someone with high empathy can feel another’s pain perfectly and still not know what to do with that information: they can freeze, absorb the other’s mood, or want to help and worsen the situation. Someone with high EI uses the other’s emotional information to choose the most useful behavior —which is sometimes to intervene, and sometimes not to—.

A metaphor that works: empathy is musical ear. Emotional intelligence includes the ear, but also rhythm, harmony, and knowing when to play and when to be silent. You can have perfect ear and not know how to make music; you can have high EI without outstanding empathy —you can regulate your own emotions precisely without feeling others’ especially strongly—.

The confusion with empathy has a concrete cost. In organizations and relationships, highly empathetic people are usually the first to burn out: they feel a lot, understand a lot, and nobody taught them the fourth branch —regulation— so they would not absorb others’ pain. What gets called “toxic empathy” is almost never excess of empathy: it is a regulation deficit, which is branch four.

4. Why it gets confused with personality (and why that stalls progress)

Personality describes stable patterns of behavior, emotion, and thought. That is what traditional tests like the Big Five measure: openness, conscientiousness, extraversion, agreeableness, neuroticism. Emotional intelligence is conceptually separate: it measures a cognitive capacity, not a behavior pattern.

The confusion is popular because several commercial “emotional intelligence” tests are, in fact, personality tests in disguise. They measure agreeableness, extraversion, ability to read people. They predict some outcomes —because agreeable and extraverted people usually have better workplace relationships— but they do not measure EI in the strict sense. They treat the surface symptom (social behavior) without touching the mechanism (reasoning about emotions).

The data that shows why that confusion stalls progress: interventions designed to improve personality —therapy, deep self-knowledge— have slow and partially stable effects. Interventions designed to improve EI as a capacity —training in emotional perception, emotional vocabulary, regulation— show measurable gains within weeks. Both are legitimate, but they are not interchangeable. If what you want is to improve branch three (understanding how emotions transform), you need practice aimed at that branch, not a generic “soft skills” course.

The 2008 review by Mayer, Roberts, and Barsade on “human abilities” summarizes decades of evidence and separates EI from personality with empirical arguments: ability tests (MSCEIT) correlate with different outcomes than personality tests, predict different aspects of performance, and respond to different interventions10.1146/annurev.psych.59.103006.093646. They are not two versions of the same thing.

5. The most common error: treating it as a fixed trait

“I am not very emotionally intelligent” gets said as if describing eye color. The research says otherwise: as a cognitive capacity, EI is modifiable through practice. Meta-analyses make this clear. Joseph and Newman’s 2010 meta-analysis proposed a cascade model where EI precedes and affects personality, not the other way around10.1037/a0017286. That is: the stable personality traits you observe in yourself are partially held up by more basic cognitive capacities —including EI— that are indeed trainable.

The metaphor that helps: you did not know how to swim before learning to swim, and once you learn, swimming does not become “part of your personality.” You can improve your stroke with practice. The same happens with each branch of EI: each one gets trained separately, with concrete exercises, and changes show up in weeks, not years.

This has a direct implication about what to do if you want to improve yours. First identify which branch is weakest —do not confuse it with your personality, because that is the wrong shortcut—. Then train that branch with targeted exercises. The tool this hub offers does exactly that mapping: eight questions that point to where your difficulty concentrates —perception, use, understanding, or regulation—.

6. The piece popular culture lost: the fourth branch

The fourth branch —emotion regulation— is the one most distorted in public conversation. It got turned into “controlling emotions” or “not feeling” or “being strong.” All three are costly errors.

Regulating is not suppressing. Suppressing an emotion does not eliminate it: it pushes it below the surface where it keeps influencing decisions, physical fatigue, and relationships. Experimental evidence shows this: people instructed to “not feel” an emotion show afterward higher physiological stress responses, worse memory, and worse problem-solving. Healthy regulation is the opposite: recognize the emotion, modulate its intensity, and decide what to do with it —including, sometimes, feeling it fully on purpose—.

The useful metaphor: emotional regulation is not putting out the fire, it is deciding what to do with it. Sometimes the fire should burn (just anger, necessary sadness). Sometimes it should be stoked (curiosity, courage). Sometimes it should be contained so it does not burn others (legitimate anger in a work meeting). Each of those uses requires having identified the emotion first —branch one— and having understood what it means —branch three—. Jumping to regulation without the prior branches is trying to use a tool that has not been built yet.

7. A minimal timeline: how the concept got built

The concept of emotional intelligence has a precise history, and that history explains why so many versions got distorted.

• 1990. Salovey and Mayer publish the first formal article defining EI as the capacity to monitor one’s own and others’ emotions, discriminate between them, and use the information to guide thought and action10.2190/dugg-p24e-52wk-6cdg. A technical article, in a psychology journal, with a preliminary three-branch model.
• 1995. Salovey and colleagues refine the model and publish the Trait Meta-Mood Scale, a self-report measure of attention to emotions, emotional clarity, and emotional repair10.1037/10182-006. The scale becomes a reference and is still used.
• 1999. Mayer, Caruso, and Salovey explicitly defend that EI meets the standards of an intelligence —the criteria used to classify verbal or spatial intelligence as intelligence10.1016/s0160-2896(99)00016-1. It is the answer to critics who wanted to reclassify it as a trait.
• 2008. Mayer, Roberts, and Barsade publish the comprehensive review on “human abilities,” placing EI among studied intelligences10.1146/annurev.psych.59.103006.093646.
• 2010. Joseph and Newman propose the cascade model linking EI to personality and motivation: EI is not independent of personality, but it is not the same thing either10.1037/a0017286.
• 2014. Côté reviews the role of EI in organizations and documents its effects on performance, leadership, and burnout10.1146/annurev-orgpsych-031413-091233.

*Four decades of systematic research on the ability model. The popular version, mixed with personality, came later.*

This timeline matters because it tells you where to look when someone offers you a test or course. If the material is based on evidence from 1999 forward and explicitly separates EI from personality, it is working with the serious model. If not, it is mixing apples and oranges.

8. What emotional intelligence is NOT (clear the ground)

Four confusions worth dismantling before going further:

It is not kindness. You can have high EI and use it to manipulate. Kindness is a value, not a capacity. EI prescribes no moral direction.

It is not extraversion. Introverts can have high EI —they regulate their emotions and understand others precisely, just in intimate settings instead of large groups.

It is not “not feeling.” It is the opposite: it requires feeling and then deciding. Sustained emotional suppression is one of the things that gives the worst results over time.

It is not the skill of “reading minds.” It is emotional perception based on real information —facial expression, tone of voice, context— combined with understanding. People who say “I know what you feel without you telling me” are usually projecting, not perceiving.

Is there a risk of "psychologizing" everything?

Yes, documented. People who get trained in emotional vocabulary can fall into the temptation of labeling everything —in themselves and others— and turning each interaction into an analysis. It is the emotional equivalent of "looking at the finger instead of the moon." Mature EI does not label in every moment: it decides when the label helps and when it gets in the way. Practice also includes learning not to analyze, to let things flow without commenting.

Can EI be measured with a free internet test?

Not seriously. Valid EI tests as ability (MSCEIT) require professional application and normed scoring. What circulates free on the internet is usually disguised self-report: questions like "am I good at reading emotions?" that measure self-concept, not capacity. A person with high EI who does not consider themselves good at reading emotions scores low on those tests —and that does not mean they have low EI, it means the test is measuring something else—.

Does EI work the same across different cultures?

With nuances. The hierarchical structure of the four branches replicates cross-culturally, but the contents vary: what counts as appropriate emotional expression, the words available to name emotions, the norms about when to regulate —all of that changes between cultures—. Cross-cultural research shows that people from cultures with richer emotional vocabularies tend to score higher on branch three (understanding), simply because they have more material to distinguish. The model is not invalidated; it is modulated.

What happens if I discover I have high EI but use it poorly?

It is more common than it seems. People with high EI in high-power professions (business, politics, law enforcement) report using the capacity for personal advantage —negotiation, persuasion, reading others' weaknesses— more than for collective well-being. The literature on "dark EI" documents this pattern and distinguishes it from prosocial use. It is not a problem of the model: it is a reminder that capacity and ethics are separate axes. That is why serious EI interventions include explicit work on values.

Does a person's EI show in how they write?

Partially. Studies on written language find modest correlations between use of emotional words (not just positive or negative, but specific and varied) and EI self-reports. The correlation is low —not enough to diagnose— but suggests that internal emotional vocabulary reflects in how text gets built. Practical curiosity: if you re-read your own emails from last year, you will probably notice whether emotional vocabulary expanded or shrank.

9. Caja de crisis

If you are emotionally overwhelmed right now, to a point you cannot manage alone, please do not face this by yourself. EI is a trainable capacity, but when emotions overflow, the next step is talking with someone trained —not reading more articles—.

Colombia: Lines 123 / 106 (psychological and emotional care, free, 24h). United States: Suicide & Crisis Lifeline 988 (free, 24h). United Kingdom: Samaritans 116 123 (24h). Spain: Teléfono de la Esperanza 717 003 717. Argentina: Centro de Asistencia al Suicida (135 / (011) 5275-1135 from mobile). Mexico: SAPTEL 55 5259-8121. International directory: findahelpline.com

Get professional help through rdkterapia — no spam, you can unsubscribe whenever you want.

10. Frequently asked questions

Is emotional intelligence inherited?

Partially. Like most cognitive capacities, it has a genetic component and an environmental one. Interventions show the environmental part is modifiable: with targeted practice, levels can rise within months.

Is EI useful for work or only personal life?

Côté’s review on EI in organizations documents effects on job performance, leadership, and reduced burnout10.1146/annurev-orgpsych-031413-091233. The distinction matters: it is not magic that improves everything; it improves processes where emotional information intervenes —negotiation, team management, recovery after errors—. Two practical examples documented in workplace studies: in customer service, high-EI workers recover faster after difficult calls; in healthcare teams, high-EI leaders show staff with lower turnover and better adherence to protocols. Both effects are modest in size but consistent across samples —and they are not explained by extraversion or agreeableness, which the studies control for—.

How long until I see results from training?

Depends on the branch. Branch one (perceiving) shows measurable gains within 2-4 weeks of distributed practice —the vocabulary expansion alone changes what you notice in days—. Branch three (understanding) takes longer, typically 6-12 weeks, because it requires building mental models of how emotions combine. Branch four (regulating) builds on the previous ones, so realistic timelines are 8-16 weeks for visible change. These are not promises —they are averages from intervention studies, with substantial individual variation—.

Does high EI make you a better parent?

Evidence is suggestive but not definitive. Studies on parental emotional socialization show that parents with higher EI are more likely to discuss emotions openly with their children, validate emotional expression, and model healthy regulation. The effects on children’s emotional development are positive on average. The caveat: most studies are correlational, and what looks like “EI makes a better parent” could partly reflect underlying factors like education or mental health status that the studies do not always fully control.

Which of the four branches is most important?

Depends on the task. For helping professionals (psychologists, doctors, social workers), branches one and three (perceiving and understanding) are critical. For leaders and people in pressure positions, branches two and four (using and regulating) make more difference. What the evidence repeats: branch one is a prerequisite for the others —you cannot regulate what you do not identify—.

Is there a reliable test to measure it?

Yes: the MSCEIT (Mayer-Salovey-Caruso Emotional Intelligence Test) measures the four branches with ability tasks, not self-report. There are other valid tests too. What is not valid: “EI” tests that in reality measure personality and are sold as ability EI —those are measuring something else—.

What if I discover I have low EI in some branch?

It is modifiable. Each branch has specific exercises: for perception, expand emotional vocabulary and practice body self-exploration; for use, learn to read emotional state as information for tasks; for understanding, study how emotions combine; for regulation, practice techniques that are not suppression. The hub’s tool helps you locate where your biggest difficulty lives.

Can children’s EI be trained?

Yes, and there are school programs with published results. What they show: measurable gains in prosocial behavior, reduction in conflicts, and better academic performance in classes that integrate emotional education. The evidence is consistent, though the size of the effect varies by implementation.

11. To close: what you already know

If you got here, you have something most people confuse: the right model. Emotional intelligence is not a slogan, not personality, not kindness, not empathy alone. It is a cognitive capacity with four hierarchical branches —perceiving, using, understanding, regulating— that get trained separately and measured with specific tests.

The question this piece hands back to you is the one we opened with, but inverted: you already know “having emotional intelligence” means nothing without knowing which of the four branches you have strongest and which one holds up your difficulty. The hub’s tools —the eight-question map and the separator— are designed so you can see exactly that today, not next year.

The following pieces of this hub go deeper: the four branches in detail (what each does, how it gets trained, how the branches get confused with each other) and myth or reality (how solid the benefits attributed to EI really are, and where evidence supports the promises and where it doesn’t). Start there if you want to move from model to practice.