Emotions may reflect predictions, not just reactions to events

Change happens when feared outcomes don’t occur.

Avoidance can reinforce fear; staying present allows learning.

Most of us assume emotions work like this: Something happens → we feel something → we react.

An email arrives.Your chest tightens.You feel anxious.

A comment is made.You feel hurt.You withdraw.

It is exactly how we experience emotions in our daily life. Emotions are reactions to events. And if those reactions are too intense or disruptive, the task is to regulate them—calm down, reframe, cope better.

Much of therapy is built around helping people manage emotional reactions more effectively.

But what if emotion isn’t simply a reaction? Emotions may reflect predictions, not just reactions to events.

The Brain Predicts Before It Reacts

Research in neuroscience suggests that the brain is constantly making predictions. Rather than waiting for events and then responding, the brain is continuously asking:

What is happening?What does this mean?What should I prepare for?

To answer those questions, it integrates:

Signals from inside the body (heart rate, breathing, muscle tension)

Information about movement and posture

Sensory input from the environment

It then interprets those signals based on past experiences, learning, and cultural meaning. The result is a best guess about what’s going on—and that “best guess” may be what we experience as emotion.

From this perspective, emotion isn’t just triggered by events; it is constructed by the brain as part of its attempt to prepare and protect us.

Emotion as the Brain’s Working Hypothesis

Consider anxiety. In the traditional model, a situation causes anxiety. But in a predictive model, anxiety may reflect the brain’s hypothesis: Something uncertain is happening. Prepare for threat.

Sadness might reflect a prediction of loss.

Anger might reflect a prediction of violation.

Shame might reflect a prediction of social rejection.

Emotion, in this sense, is the brain’s best attempt to regulate and survive.

That changes how we think about emotional distress.

Instead of asking, “How do I get rid of this feeling?” we might ask, “What is my brain predicting right now?”

How Change May Happen

In neuroscience, learning often occurs through prediction error—the gap between what we expect and what actually happens.

Imagine someone believes: “If I let myself feel this sadness, I will fall apart.”

In therapy, they slowly allow themselves to feel it. They stay present. Their body tightens, tears come—and yet, they don’t collapse. The feared outcome does not happen.

That mismatch—between expectation and reality—creates an opportunity for the brain to update.

Over time, the prediction may shift: “Sadness is painful, but survivable.”

The same can happen with anger, anxiety, or shame. When we remain with an emotion long enough to discover that the predicted catastrophe doesn’t occur, something in the nervous system recalibrates.

Regulation vs. Avoidance

Of course, emotional regulation is important. No one benefits from being overwhelmed but sometimes, we regulate so quickly that we never learn anything new.

If every surge of anxiety is immediately numbed, avoided, or escaped, the brain never gets disconfirming evidence. The prediction—“This is dangerous”—remains intact.

In behavioral terms, avoidance can become negatively reinforced: Relief confirms the belief that escape was necessary.

By contrast, carefully staying with sensation—without being flooded—can allow the nervous system to register: “I thought this would destroy me. It didn’t.”

That is a powerful corrective experience.

This is why many therapists increasingly pay attention not just to thoughts, but to bodily signals:

Subtle shifts in breath.A tightening in the throat.Heat in the chest.

These are not just symptoms. They are part of the predictive system itself.

Learning often happens in those moments when a person:

Notices the sensation

Discovers it can move, shift, and pass.

Change is less about eliminating emotion and more about increasing flexibility.

Two major challenges arise in this process.

First, many predictions operate outside conscious awareness. A person may not know they expect rejection, failure, or catastrophe. The therapist and client often need to uncover these implicit expectations together.

Second, language has limits.

We rely on words to describe experience. But emotional experience is deeply embodied and often nonverbal. Language can clarify what we feel—but it can also simplify or narrow it.

Language is a powerful vehicle for communication, yet it cannot fully contain the nonverbal dimensions of lived experience. Interoceptive shifts, subtle affective tones, and implicit expectations often emerge before they can be named. When we translate experience into words, something is inevitably shaped, filtered, or reduced.

Sometimes change happens in ways that are sensed before they are spoken. This reminds us that therapy is not only a conversation. It is also an experience.

A Different Way to Think About Emotion

If emotion reflects the brain’s predictions, then it is not simply something happening to us; it is something our nervous system is generating in an attempt to help us adapt.

This doesn’t mean painful emotions are pleasant or desirable but it does mean they may carry information.

When therapy helps someone remain present with their experience long enough for new learning to occur, emotional patterns can shift because the underlying predictions were revised.

In that sense, emotion may not be the problem to eliminate. but a primary way the brain learns.

To find a therapist, visit the Psychology Today Therapy Directory.

For clinicians interested in exploring these ideas further, I write regularly on Substack.

Barrett, L. F. (2017). How emotions are made: The secret life of the brain. Pan Macmillan.

Barrett, L. F., Mesquita, B., Ochsner, K. N., & Gross, J. J. (2007). The experience of emotion. Annu. Rev. Psychol., 58(1), 373-403.

Friston, K. (2010). The free-energy principle: a unified brain theory?. Nature reviews neuroscience, 11(2), 127-138.

Seth, A. K. (2013). Interoceptive inference, emotion, and the embodied self. Trends in cognitive sciences, 17(11), 565-573.

© Psychology Today