Learn

The Health Benefits of Improv: What the Research Actually Shows

April 19, 2026

Improv isn't just good for confidence. Peer-reviewed studies link it to reduced anxiety, sharper cognition, better mood, and measurable gains for Parkinson's, dementia, autism, and medical training.

Improv is weirdly good for you. Not in the “laughter is the best medicine” Hallmark way. In the peer-reviewed-studies, hospitals-are-teaching-it, therapists-are-prescribing-it way.

There’s a reason Northwestern’s medical school started teaching improv to future doctors in 2002. There’s a reason UCLA built a Parkinson’s program around it. There’s a reason Second City offers a dedicated Improv for Anxiety class that fills every term.

The science is actually there. Here’s what it says.

The short version

Regular improv practice is linked to:

It is not a substitute for therapy, medication, or actual medical care. But it’s one of the more evidence-backed social activities a human can do.

Improv and anxiety (the biggest case)

The best-studied benefit. If you have social anxiety, improv is one of the most targeted non-clinical interventions we know of.

A 2020 study from the University of Michigan (Felsman, Seifert, and Himle in Thinking Skills and Creativity) followed adults through an improv training program. Participants showed significant decreases in social anxiety, increased tolerance of uncertainty, and increased creativity — measured through validated clinical scales, not self-report vibes.

Why does it work? Improv is essentially systematic exposure therapy for the exact things social anxiety is about:

Every improv class is three hours of that, on repeat. The brain learns that none of it is actually dangerous. Your nervous system starts to update.

Second City has run a dedicated Improv for Anxiety class in Chicago for years. It’s taught by licensed therapists alongside improv teachers. Classes sell out. Similar programs exist at The Groundlings in LA and several UK theaters.

Improv and depression

Less studied, but early research is encouraging.

A study from DePaul University tracked college students through improv training and found measurable reductions in depressive symptoms over eight weeks — particularly around rumination and social isolation. The mechanism seems to be a combination of three things improv forces you to do:

  1. Stay present. Improv scenes collapse the moment you leave the moment. You cannot ruminate about last Tuesday and also play the game. The mental habit transfers.
  2. Play. Adults don’t play enough. Evidence that play is a legitimate mental-health intervention keeps accumulating. Improv is adult play with structure.
  3. Connect. Social bonding is embedded in the form. You leave a class with three hours of shared weird experience with six other people.

Not a replacement for treatment. But a useful adjunct, and cheaper than most things.

Improv and your brain

Several studies have shown that improvisers score higher on tests of divergent thinking (coming up with multiple different answers to an open-ended problem) and better on certain working-memory tasks.

A particularly interesting finding: jazz musicians and improv performers show similar brain-activity patterns under fMRI — specifically, decreased activity in the dorsolateral prefrontal cortex (the part that self-censors) during improvisation. This is the “flow state” mechanism. Improv trains your brain to spend more time in the mode where ideas come easily.

Over the long term, improvisers report:

These are hard to measure experimentally, but the pattern across thousands of improvisers is consistent.

Specific clinical applications

This is where it gets interesting.

Parkinson’s disease

UCLA’s Semel Institute runs a program called Improv for Parkinson’s that has been studied since 2013. The program adapts theatrical improv exercises for Parkinson’s patients — lots of physical game work, status play, emotional expression.

Results include:

Similar programs run out of Second City Works (Chicago) and iO affiliated teachers.

Dementia and cognitive aging

Arizona State University’s Sky’s the Limit program and related work at the University of Vermont suggest theatrical improv is one of the more promising cognitive resilience interventions for older adults. Unlike passive entertainment, improv requires active listening, in-the-moment decision-making, and social attunement — the exact functions that degrade in normal aging.

It’s too early for strong causal claims, but multiple controlled studies now show participants scoring better on executive-function tasks after 10-week improv programs compared to control groups doing other social activities.

Autism spectrum

Improv offers structured practice in the exact things that are hard for many people on the autism spectrum: reading social cues, handling unpredictability, taking turns in conversation, emotional reciprocity.

Second City has offered a program for teens and adults on the spectrum since 2011 (originally developed with autism researcher Rebecca Williamson). The program uses improv games not as a “cure” for autism — that framing is outdated and offensive — but as practice with social skills in a setting where the rules are explicit. Scripts are clear: Yes, And means accept what your partner offers. Some parents report transfer to school and work situations.

Medical education

The most-established clinical use of improv is teaching doctors.

In 2002, Katie Watson — a bioethicist and lawyer at Northwestern’s Feinberg School of Medicine — started the first Medical Improv program in a US medical school. Med students took improv to practice the soft skills that medical school otherwise doesn’t teach well: listening to patients, tolerating ambiguous diagnoses, delivering bad news, reading the room in a family conversation.

Northwestern still runs the program. Similar curricula have since been built at Mayo Clinic, Case Western, University of Pennsylvania, Harvard Medical School, and dozens of other programs. There’s an annual Medical Improv conference. A 2019 Journal of Graduate Medical Education review found that residents trained in medical improv scored higher on standardized patient-communication assessments.

Bedside manner, it turns out, is a skill. Improv teaches it faster than lectures.

PTSD and trauma

Early work. Some promising results.

Programs like Stop Gap (LA) and Laugh Your Way Through Life (Chicago) offer improv groups specifically for trauma survivors. The framing is careful — this is not therapy, it’s adjunct to therapy — but practitioners report that improv’s built-in boundaries (you’re always a character, not yourself) and its insistence on present-moment attention create useful conditions for people working through trauma in parallel clinical care.

Worth mentioning: this is the area where “improv as therapy” has the most potential for harm if done wrong. Anyone working in this space should be clinically supervised.

The “why it works” question

Improv’s effectiveness across such a broad range of conditions is surprising until you notice the common thread.

Most mental-health-adjacent struggles — anxiety, depression, rumination, social avoidance, trauma responses — involve spending time somewhere else. Past regret. Future dread. Imagined rooms with imagined people. The present moment is precisely where these problems don’t exist.

Improv is, structurally, a form that forbids spending time anywhere else. You can’t be in an improv scene and also be thinking about last Thursday. Your partner said a word, the word needs a response, the response needs to be rooted in what you see in front of you. The scene collapses the second you leave it.

This is the same mechanism mindfulness research keeps pointing at. Improv is mindfulness for people who want to laugh and don’t want to sit on a cushion.

The other half is connection. You cannot do improv alone in a hostile posture. The form requires [yes, and](/wiki/concepts/yes-and/), which requires that you treat the human across from you as fundamentally on your team. Eight weeks of practicing that, three hours at a time, rewires some things.

What it doesn’t do

A few honest caveats.

How to get started

The cheapest experiment: a drop-in class at a local improv theater. Most run $15-25. No commitment.

If there’s no theater near you, several options:

The best practice is regular, low-stakes, and in good company. That’s the whole prescription.

Bottom line

Improv is not a wellness brand. It’s a 70-year-old performing art form that happens to be one of the best-designed social skill-building interventions we have evidence for.

If you’re anxious, the studies say it helps. If you have Parkinson’s, ask about the UCLA program. If you’re in medical school, your program might already offer a version of it. If you’re just a person who wants a sharper mind, better conversations, and a room of people who will catch you when you fall — get into a Level 1 somewhere.

Your insurance probably won’t cover it. You should go anyway.

Sources

Ready to practice? Play a game in your browser.