Emotional Intelligence: The Research Behind the Buzzword

Someone in your workplace is described as having high emotional intelligence. What do you picture? Maybe someone who stays calm under pressure, reads the room well, gives thoughtful feedback. Or maybe someone who’s just… likable. The kind of person people want to be around.

That last description is where the concept starts to blur. Because likability, social skill, and warmth are real and valuable things, but they aren’t the same as emotional intelligence, and treating them as if they are has muddied a genuinely useful area of psychological research.

Two Very Different Definitions

The popular understanding of emotional intelligence comes largely from Daniel Goleman’s 1995 book of the same name, which described EQ as encompassing self-awareness, self-regulation, motivation, empathy, and social skill. Goleman’s framework resonated with a broad audience, made its way into corporate training programs worldwide, and has largely shaped how most people think about the concept.

The scientific model came earlier and is considerably more specific. In 1990, psychologists Peter Salovey and John Mayer published a paper defining emotional intelligence as the ability to perceive emotions accurately, to use emotions to facilitate thought, to understand emotional information, and to manage emotions effectively. John Mayer, along with David Caruso, later developed the MSCEIT, an ability-based test designed to measure these capacities in a way more consistent with how intelligence is measured in other domains.

The key difference between these two frameworks is consequential. The Mayer-Salovey-Caruso model treats emotional intelligence as a genuine cognitive ability, something you either do more or less accurately, like spatial reasoning or verbal comprehension. The Goleman model expanded EQ to include personality traits, motivational dispositions, and social behaviors. Critics of Goleman’s approach, including Mayer and Salovey themselves, have pointed out that by the time you’ve included all of those components, you’re no longer describing a specific ability. You’re describing a cluster of socially desirable characteristics that already have their own names.

What the Ability Model Actually Measures

The four-branch model from Mayer, Salovey, and Caruso describes emotional intelligence as a hierarchy of abilities, each building on the one before.

The first branch is perceiving emotions accurately. This means reading emotional signals in faces, voices, art, and posture. It’s the most basic capacity and the foundation for everything else. If you frequently misread how someone is feeling, you’re working with faulty data, and the rest of your emotional processing will be compromised accordingly.

The second branch is using emotions to facilitate thinking. Emotions aren’t just noise that interferes with rational cognition. They direct attention, influence what seems important, and can actually improve certain kinds of judgment when used deliberately. A mild melancholy state, for example, has been associated with more careful analytical thinking. Recognizing that your current emotional state is affecting what you’re noticing and how you’re evaluating things is a form of intelligence.

The third branch is understanding emotions: knowing how emotions work, how they blend and transform, how they fit into sequences over time. If you’re angry at someone and you understand that anger often follows a feeling of being disrespected, you have better information about what’s actually happening than if anger just appears to you as a raw state with no internal logic.

The fourth branch, managing emotions, is the one most people focus on. But in this model, it doesn’t mean suppression. It means the ability to choose how to engage with emotional experience, deciding when to remain with a difficult feeling, when to shift it, and what approach is most appropriate given the situation.

What EQ Research Actually Predicts

When researchers use ability-based measures of emotional intelligence rather than self-report questionnaires (which are more susceptible to people simply rating themselves as they’d like to be), the findings are interesting and more modest than the popular version of EQ would suggest.

Ability-based emotional intelligence predicts performance in domains where emotional information is central. It’s associated with better social relationships, more effective leadership in interpersonal contexts, and somewhat better mental health outcomes, particularly around emotional regulation. Some research finds associations with life satisfaction, though the effect sizes are typically smaller than what Goleman’s popularization implied.

It does not appear to predict general job performance across all domains. A highly emotionally intelligent accountant isn’t necessarily better at accounting. What it does seem to predict is performance in jobs that require managing relationships, navigating complex social dynamics, and handling emotionally charged situations.

Emotional intelligence also doesn’t predict academic achievement as powerfully as standard cognitive intelligence does. This was a claim Goleman made that researchers pushed back on pretty firmly. Academic performance is still better predicted by general intelligence (g) and conscientiousness than by EQ measures.

What Emotional Intelligence Is Not

Emotional intelligence is not being nice. Niceness is a disposition. You can be pleasant, accommodating, and easy to get along with while being quite poor at accurately reading emotions, understanding their causes, or managing them effectively. Conversely, someone with high emotional intelligence might be direct, occasionally uncomfortable to be around, and still excellent at navigating emotionally complex situations.

It’s not empathy, exactly, though they’re related. Empathy is the capacity to understand or share another person’s emotional experience. Emotional intelligence is a broader set of abilities that includes perceiving and understanding emotions, your own and others’, but isn’t identical to empathic responding.

It’s not emotional expressiveness. People who display their feelings openly aren’t necessarily more emotionally intelligent than people who are more reserved. Emotional expression and emotional processing are different things.

And it’s not a fixed trait. The ability-based model treats EQ as something that can be developed, just as other cognitive abilities can be developed with practice and knowledge. The research on training emotional intelligence is mixed, but the weight of evidence suggests that emotional perception and understanding, in particular, respond to deliberate practice and feedback.

What This Means in Practice

One thing the research consistently underscores is that accurate self-assessment of emotional intelligence is itself difficult. Studies routinely show that people’s self-ratings of their EQ correlate poorly with their performance on ability-based measures. That’s worth sitting with: the people who most confidently describe themselves as emotionally intelligent are not reliably the most emotionally intelligent by objective measures.

This doesn’t mean self-awareness is useless. It means that emotional intelligence, like other forms of intelligence, benefits from honest external feedback. Therapy can serve this function. A skilled therapist creates conditions where you can observe your emotional patterns in real time, where blind spots tend to surface, and where you get feedback that isn’t filtered through politeness or social obligation.

Understanding the distinction between the ability model and the trait model also matters for how you interpret moments of difficulty. If you treat emotional intelligence as a fixed personality trait, then struggling to manage your emotions in a hard situation feels like evidence of who you are. If you treat it as a set of learnable abilities, the same struggle becomes information about what you haven’t yet practiced enough, which is a fundamentally different and more generative frame.

Emotions are data. They carry information about what matters to you, what you’re perceiving, what your nervous system is registering as significant. Getting better at reading that data accurately, using it thoughtfully, and managing your relationship to it isn’t a soft skill. It’s a capacity that shapes almost everything else.

This article is for educational purposes only and is not a substitute for professional mental health treatment. If you are experiencing a mental health crisis, please reach out to a qualified mental health provider or call 988.