Shadow Work Prompts: 15 Questions That Go Somewhere (and What the Research Actually Shows)

The answer, before anything else

Shadow work prompts are journalling questions aimed at the parts of yourself you tend not to look at — and the ones that go somewhere share a shape: they ask about a specific moment rather than a general trait, and each is followed by one more question that turns a complaint into a pattern. The 15 below are written that way, and you can use any of them tonight with a sheet of paper.

Two honest notes first. "Shadow" is a term from Carl Jung's analytical psychology, not from clinical practice. And the research on reflective writing — the closest thing to shadow work that has actually been tested — shows a real but small average effect, with some trials finding nothing at all. Both are below, with sources.

Where the idea comes from

Jung used "the shadow" to mean, in his own compact phrase, "the thing a person has no wish to be" (*Collected Works* 16, para. 470). His fuller essay on it, titled simply The Shadow, appears in Collected Works 9ii and dates from 1951, where he describes it as "a moral problem that challenges the whole ego-personality".

What that gives you is a usable idea rather than a diagnosis: the traits you most firmly believe you do not have are worth a second look, because the firmness of the belief is doing work. That is a pattern to examine, not a condition anybody has.

What the term does not give you is a treatment. Shadow work is a self-reflection framework from a mid-twentieth-century school of psychology; it is not an evidence-based clinical intervention and not a substitute for therapy.

What the evidence actually says

The testable version of this is "expressive writing" — writing about difficult experiences and the feelings attached to them. It has been studied for decades.

The largest synthesis is Frattaroli's 2006 meta-analysis in Psychological Bulletin, pooling 146 randomised studies: "experimental disclosure is effective, with a positive and significant average r-effect size of .075", 95% confidence interval .052 to .098.

That number is the point. It is positive and statistically reliable — the interval does not cross zero — and it is also small, below the conventional r = .10 benchmark for a "small" effect.

And it does not hold everywhere. A 2023 meta-analysis in Cognitive Therapy and Research pooled 11 randomised trials with 1,188 participants among Asian populations and found "no significant group differences between expressive writing and control conditions at the first follow-up assessment, g = 0.05", nor at the second, g = 0.04.

So the realistic claim is: writing about difficult things tends to help a little, on average, for some people. That is a reason to try it and a reason not to expect it to be transformative — and if you have tried journalling and found it did nothing, that is a normal result rather than a failure of effort.

The shape of a prompt that goes somewhere

Three rules separate a prompt that produces a page of insight from one that produces a page of complaint.

1. Name a moment, not a trait. "Am I a jealous person?" has no answer. "When did I last feel a flash of something unpleasant at someone else's good news?" has a date and a scene attached, and scenes contain detail you had not noticed.
2. Follow it with "and what did I do next?" This converts feeling into pattern. The feeling is information; the behaviour it produced is what you can change.
3. Write for a fixed short time and then stop. Ten or fifteen minutes. The trials behind the evidence above used short, bounded sessions, and a stopping point is also what keeps this from turning into rumination.

A prompt that breaks rule 1 will almost always produce a vent — the most common reason people conclude this does not work for them.

15 shadow work prompts, each with its follow-up

Pick one. Not five.

1. What did I most recently pretend not to mind? → And what did I do instead of saying I minded?
2. Who irritates me out of proportion to what they actually did? → What do they get away with that I do not let myself do?
3. When did I last feel relieved by someone else's bad news? → What did that relief tell me I had been comparing?
4. What compliment makes me uncomfortable? → What would be true about me if I simply accepted it?
5. What am I certain I would never do? → Under what specific circumstances might I do it anyway?
6. Which of my parents' habits do I most dislike? → Where does a version of it appear in my own week?
7. What was the last thing I said that I immediately wanted to take back? → What was the real sentence underneath it?
8. Where am I generous in a way that I keep count of? → What am I expecting in return that I have never asked for?
9. What do I most want people to think about me? → What would I have to stop doing if I stopped managing that?
10. When did I last avoid someone rather than have a conversation? → What was the conversation going to cost me?
11. What is the compliment I give others that I most want to hear? → Who have I never asked for it from?
12. What part of my day do I dread, quietly? → What have I decided it would be unreasonable to change?
13. What did I apologise for most recently that was not my fault? → What was the apology actually buying?
14. Which of my own rules do I break in private? → What would it mean if the rule were simply wrong?
15. What would the people closest to me say I am like when I am tired? → Is that a different person, or an unguarded one?

How to use them without it becoming rumination

- One prompt, fifteen minutes, then stop. Set a timer. The bounded session is what the tested version looks like.
- Write by hand if you can. It is slower and harder to delete, and deleting is how honesty leaks out of this.
- Do not reread for at least a week. Immediate rereading turns into editing your own account.
- Stop if it is making things worse. Writing that circles the same injury without moving is rumination, and rumination is not what the research found a small benefit for. If it consistently leaves you worse off for hours, that is a signal to stop and talk to someone rather than write harder.
- Nothing here diagnoses anybody. These prompts describe common patterns. They do not identify a condition in you, and they are not a way to decide what someone else is.

Take the quiz

Not sure which of the 15 to start with? The Shadow Work Starting Point Quiz asks six questions and points you at the three prompts most likely to go somewhere for you. No email needed.

Frequently Asked Questions

Q: How do I do shadow work as a beginner?
A: Take one prompt, write for fifteen minutes by hand, answer the follow-up question, and stop. That is the entire practice. Beginners usually go wrong in one of two ways: choosing a prompt about a trait rather than a moment, which produces generalities, or trying to do several prompts in a sitting, which produces a summary instead of a scene. Search demand for the phrase "how to do shadow work" has risen over the last 30 days measured on YouTube, which suggests a lot of people are at exactly this point.

Q: What is shadow work, in plain terms?
A: It is structured self-reflection on the parts of yourself you would rather not claim — the irritations that are out of proportion, the things you are certain you would never do, the generosity you secretly keep score of. The term comes from Carl Jung, who defined the shadow as "the thing a person has no wish to be". Mind Decoded treats it as a framework for noticing patterns, not as a clinical method, because that is what it is.

Q: Does shadow work actually work?
A: Shadow work as such has not been tested as a discrete intervention. The closest tested thing is expressive writing, where Frattaroli's 2006 meta-analysis of 146 randomised studies found a small but statistically reliable average effect (r = .075), and a 2023 meta-analysis among Asian populations found no significant effect at all. The honest summary is: small average benefit, inconsistent across populations, worth trying cheaply rather than expecting a lot from.

Q: Is shadow work dangerous?
A: For most people it is a sheet of paper and fifteen minutes. The realistic risk is rumination — circling a painful memory without it moving, which tends to leave people worse rather than better. The bounded session, the follow-up question and the week before rereading all exist to reduce that. If the writing consistently leaves you distressed for hours, stop and speak to a professional.

Q: How often should I do it, and what if a prompt produces nothing?
A: Once or twice a week is plenty — the trials behind the evidence above generally used a handful of short sessions rather than a daily habit, and a daily practice tends to run out of specific moments and start recycling general complaints, which is rule 1 breaking down. If a prompt returns nothing, move on and note which one it was; it is sometimes worth revisiting in a month, but staring at a blank page is not a result worth forcing.

Q: Can I do shadow work instead of therapy?
A: No. It is a self-reflection exercise with a small measured average benefit, and therapy is a different thing with a different evidence base. The two are not substitutes, and anything describing journalling prompts as a replacement for treatment is overselling.

Subscribe to our YouTube channel: https://www.youtube.com/@minddecoded987

Connect

Questions or thoughts? Reach out anytime.

Email

subscribe to our newsletter

hello@our-minddecoded.com

© 2026. All rights reserved.

Terms of Service Privacy Policy Disclaimer