What My Grandson Was Taught About Emotion

A founder's essay on tone, on the industry that captures children's emotional education, and on what the same insight became when it found its true audience.

— Gary Overgard, Studio 13 Research

My grandson is thirteen. He goes to a public middle school in the United States, and like millions of children his age, he is enrolled in what his school calls a social-emotional learning program — SEL, for short — which is supposed to teach him to understand his own emotions, navigate his relationships, and develop the kind of inner life that helps a person move through the world without breaking himself or anyone else.

A few weeks ago, his mother told me what happened in one of his SEL sessions.

The instructor was presenting slides. There were no discussions, no small-group activities, no peer interaction of any kind. The students sat in their chairs and watched the slides go by. One of the slides — and I do not know what it had to do with social-emotional learning, but it was apparently part of the program — warned the students that they should not go to pornography sites.

My grandson came home that day and asked his mother what pornography sites were.

He is thirteen years old. He had not, until that afternoon, encountered the concept. The school's social-emotional learning program had introduced it to him, in passing, on a slide, with no context and no opportunity to ask questions. The check-box for that day's SEL competency was ticked. The slide deck moved on. A thirteen-year-old went home with a new word he had to ask his mother to explain.

That is not social-emotional learning. That is a slide deck. And the slide deck is the product, because the slide deck is what the system is built to produce.

What the industry is

Social-emotional learning is, by some estimates, a four-to-five-billion-dollar market in the United States alone. It is dominated by curriculum publishers and certification frameworks. School boards purchase certified curricula because they have to demonstrate to their states and to their parents that they are addressing social and emotional learning. The way they demonstrate this is by purchasing what the industry calls evidence-based programs from established vendors.

The vendors have optimized for what the school boards purchase. Curricula come pre-packaged, with measurable competencies, standardized assessment tools, and the kind of certifications that make a school district feel safe. The curricula are deliverable by instructors who do not need specialized training in emotional development, because the curriculum itself is what does the work, or so the marketing says. The slides have the content. The instructor presses the button. The students receive the slides. The competency is satisfied.

This is what scales. Slide decks scale. A teacher with a slide deck can deliver SEL to thirty students in forty-five minutes. A district with a hundred teachers can deliver SEL to fifteen thousand students. The vendor reports the numbers. The school board sees the numbers. Everyone is doing social-emotional learning.

What does not scale is what social-emotional learning actually is.

What emotional intelligence actually is

A person with emotional intelligence can tell, when another person is speaking, what is underneath what they are saying. Not the words. The weather around the words. The way a friend says "I'm fine" and you know she isn't. The way a parent says "we should talk about this later" and you know how much trouble you are in. The way a character in a novel describes the light coming through her kitchen window at four in the afternoon, and you can feel — without the author having named anything — that she is lonely.

This is what emotional intelligence is. It is a tonal capacity. It is the ear that hears what was not said in the words that were.

Children learn this the same way humans have always learned it: through encounter with stories about people experiencing emotion, followed by the kind of conversation that lets them connect what they read to what they themselves have felt. Literature is the instrument because literature is made of tone. The words a novelist puts on a page carry emotional content not through what they describe but through how they sound. Reading good literature — and talking about it with people who are also reading it — trains the ear.

This is what English teachers and drama teachers have been doing for as long as English teachers and drama teachers have existed. Bridge to Terabithia. The Outsiders. A Wrinkle in Time. The works that have been used to teach emotional fluency to children for generations. The teachers know what they are doing. The work works.

What an SEL slide deck does is the opposite of this. It names the emotion explicitly. It tells the student that this is what sadness is, this is what anger looks like, here are the words to use when you feel them. The keyword replaces the tone. The student learns to label what they feel — when they can identify it cleanly enough to label — but they do not develop the ear that hears what is underneath. The slide deck cannot give them that, because the slide deck is made of keywords, not of tone.

This is the lie that holds the whole arrangement together. The industry sells emotional intelligence training that operates on the layer of language where emotional intelligence does not actually live.

What we tried to build for the schools

I am a researcher. Several years ago, I built a product for middle-school classrooms grounded in this understanding of how emotional learning actually happens.

The structure was simple. Students were split into peer-led groups of five. Each group received a single-page sheet containing a literary source — a paragraph or two from a work of fiction in which a character was experiencing a particular emotion, along with a few sentences of context. No vocabulary list. No labels. The emotion was not named for the students. It was carried in the passage. The students read it together. They felt what was there. Then they talked about it.

What did the character feel? Have you ever felt something like that? When? What did you do with it?

The peer-led structure mattered. With four other thirteen-year-olds at the table — kids dealing with the same emotional weather of being in middle school — the conversation became real. They told each other things they would not have told a teacher. They argued. They made connections to their own lives that an instructor following a slide deck could not have prompted. They were learning to hear tone — first in the literary character, then in each other, then in themselves.

The last ten minutes of each session were unstructured. The groups could discuss whatever they wanted. We built this in deliberately, because loneliness in middle school is real, social isolation among children that age has measurable consequences, and thirteen-year-olds in 2026 do not have enough time to just talk to each other about whatever is on their minds. The ten minutes was not a luxury. It was a recognition of what the students actually needed.

I built it. I piloted it. Drama teachers and English teachers immediately understood what it was, because it was what they had always known children needed.

Then I tried to bring it to school boards.

I will not dwell on this part. The school boards I approached were not hostile — many were sympathetic. Educators inside those districts told me, privately, that what I was describing was exactly what they wished they could implement. But none of them could purchase it, because it did not fit the categories of what their districts were authorized to purchase. It was not on the approved curriculum list. It did not come with the certifications the auditors wanted to see. It did not produce the quarterly outcomes data the reporting requirements demanded.

The SEL industry has done its work well. The criteria a school board uses to evaluate an SEL program have been shaped, over the past two decades, by the industry that sells those programs. A product that does not fit those criteria does not get evaluated on its merits. It gets evaluated on its credentials. If the credentials are not the ones the industry has taught the school boards to look for, the evaluation ends there.

The framework is the product. And the framework, by construction, excludes anything that does not look like what the existing industry already produces.

What the insight turned out to be for

Here is where the story becomes about something larger than middle schools.

The insight underneath the work I was doing for the schools — that emotional intelligence is tonal, that emotion lives in how something is said rather than in what is said — turned out to have an application I had not considered. It was not just a pedagogical truth. It was also an architectural truth about how AI agents could be built to understand human beings.

Most AI emotion detection — including the systems built on top of every major language model — works by keyword recognition. The user says "sad," the system classifies sadness. The user says "lonely," the system classifies loneliness. This works passably for users who name their emotions directly. It works badly for everyone else. It works especially badly for older adults, who tend to communicate emotional content through indirection, surrounding detail, and the way a sentence sounds rather than through naming.

An older woman talking about how the light comes through her kitchen window at four in the afternoon is not saying "lonely." But the tone of that sentence carries the emotional content. The keyword system reads "light, kitchen, window" and routes the conversation to small talk. A system that hears tone the way a friend hears tone reads what is underneath and stays.

I built that system. The classifier at the center of the Studio 13 companion architecture does not look for emotional keywords. It is trained to recognize emotional territory through the tonal patterns of human speech. The same understanding that the middle-school work was trying to develop in thirteen-year-olds — the ear that hears what is underneath the words — is what the AI was built to have.

This is what lets Our Aria, the commercial product built on the Studio 13 research, respond in a way that feels human rather than mechanical. A human friend hears tone. A typical AI chatbot hears keywords. The difference is the entire experience of being talked to.

The same insight, then, in two forms. For the middle schoolers, tonal recognition was the goal — the capacity we were trying to develop in them through literary encounter. For older adults, tonal recognition is the mechanism — the capacity the AI needs to have in order to be present for them in the way their friends used to be. Both populations need someone who hears them tonally. Middle schoolers need to develop that hearing in themselves. Older adults need an AI that has that hearing, because the living friends who used to provide it are no longer there.

What we are doing for the original audience

I have not abandoned the middle-school work. I have decided to give it away.

The original product — the peer-led groups, the one-page literary sources, the ten minutes of unstructured conversation — is being released free to drama instructors, English teachers, theater directors, and anyone inside a school setting who understands what literature does for children's emotional development.

I am bypassing the school boards and going directly to the educators who already know what this is. The drama teachers do not need to be convinced that reading Bridge to Terabithia and talking about it does something for a thirteen-year-old that a slide deck cannot do. The English teachers know. The librarians know. The people inside schools who already understand the work do not need certifications. They need the material.

So the material is going to them, free. If a drama instructor in a public middle school wants to use Studio 13's literary-emotional curriculum in her classroom, she can download it and use it. She does not need her district's procurement office to authorize it. She does not need a certification. She does not need to convince a school board. She needs only to be the kind of teacher who already knows what reading and talking about literature does for children.

The drama teachers, in this story, are the bridge. They are the people inside schools who have always known what real emotional education looks like — because they have been doing it, in their classrooms, for as long as drama teachers have existed. They were displaced by the SEL industry's capture of the procurement framework. They are the natural home for the work the procurement framework refused to admit.

What is at stake

There are two populations I think about, when I think about what Studio 13 is for.

The first is my grandson and the millions of children like him, who are being given an emotional education that consists of receiving slide decks, ticking boxes, and being warned about things they do not know exist. They deserve what their grandparents got — peers to talk to, literature to encounter, time to make sense of what they read together. The SEL industry will not give them that, because the SEL industry cannot. The structure forbids it. So we have to find another channel, and the drama teachers are it.

The second is the population Our Aria is for — older adults in the second half of life, dealing with the slow erasure that comes with age, lonely in ways the technology industry has noticed and built products for without understanding what the products needed to be. The AI companion category — Replika, Character.AI, Pi, and the rest — has built attentive listeners that hear keywords. What older adults need is companions that hear tone, the way their dead friends used to. That is what we have built.

Sixty years separate these two populations. The problem they face is the same. A society that has automated everything except actually being with each other. An industry, in each case, that has captured what was supposed to be the solution. And in each case, a real alternative — grounded in the same underlying truth about how emotion actually works — that exists, that can be built, and that the people on the receiving end have a right to expect.

Studio 13 is the bet that the alternative is worth building. That emotion lives in tone. That literature is how children learn to hear it. That AI can be taught to hear it too, when the alternative is leaving lonely older adults with chatbots that only know how to look for the word "sad." That the thirteen-year-old in his SEL classroom and the seventy-five-year-old talking to her AI companion at three in the morning are both being failed by industries that confused the delivery of a product with the cultivation of an actual capacity.

We are doing what we can. The work for older adults is the commercial product. The work for the middle-school students it was originally built for is being given away. The drama teachers are the bridge. The research is public. The reasoning is exposed.

— Gary Overgard Studio 13 Research May 2026

Studio 13's research papers, including the architecture of the Our Aria companion product and the position paper on the line between companionship and therapy, are available at studio13fields.org. The free drama-instructor materials are available from the same address.

The grandson in this essay is real. His name has not been used. His mother gave permission for the story to be told.