Technique, examined
Affirmations
This is the rare case where the research is genuinely encouraging and the popular version still gets it wrong — because the effect reverses depending on whether you believe the sentence you are repeating. For a large number of people, the affirmations they were handed make things measurably worse.
Two different practices sharing one name
What gets called affirmation covers two things that behave differently.
The first is self-affirmation in the research sense: briefly reflecting on a value that genuinely matters to you — honesty, your family, craft — usually by writing a short paragraph about why it matters. This has a substantial literature behind it.
The second is the popular version: repeating a positive statement about yourself that you do not currently believe. I am wildly successful. I am worthy of abundance. This is the one on the phone lock screens, and it is the one with the problem.
The self-affirmation evidence is real
Values-affirmation exercises have been shown to reduce defensiveness under threat and to improve performance in stressful settings. Cohen and colleagues found that a brief values-writing exercise, repeated a few times over a school year, narrowed an achievement gap among students under stereotype threat — with effects still detectable two years on.
Cohen, Garcia, Apfel & Master, “Reducing the racial achievement gap: a social-psychological intervention,” Science, 2006.The proposed mechanism is not belief installation. It is that reminding yourself of something you genuinely value restores enough footing that a threat stops consuming attention. You are not persuading yourself of anything new; you are standing on ground you already occupy.
Note what those exercises never ask. They do not ask you to assert something false about your circumstances. They ask you to write about something already true about your priorities.
The popular version can backfire
The distinction matters because the other kind has been tested too, and the result is uncomfortable.
Wood, Perunovic and Lee had participants repeat “I am a lovable person.” Those with high self-esteem felt slightly better. Those with low self-esteem felt worse — worse than a control group that repeated nothing at all.
Wood, Perunovic & Lee, “Positive self-statements: power for some, peril for others,” Psychological Science, 2009.The explanation is straightforward once stated. Repeating a claim you do not believe does not overwrite the belief; it invites the rebuttal. Say I am lovable to yourself and some part of you produces the counter-evidence, in detail, unprompted. You have not installed a belief. You have run a rehearsal of the case against it.
An affirmation you do not believe is an argument you keep losing.
This is worth sitting with, because it inverts who the practice is sold to. Affirmations are marketed hardest at people who feel worst about themselves, and that is precisely the group the finding says is harmed.
What to write instead
The repair is not to abandon the practice. It is to stop asserting things you do not believe and start asserting things you do.
Affirm the value, not the outcome
“I care about doing work I am not embarrassed by” rather than “I am a wildly successful entrepreneur.” The first is already true and needs no defending. The second starts an argument.
Use the evidence you already have
“I have handled harder weeks than this one” beats “I am unstoppable.” A specific true memory cannot be rebutted; a superlative can be, instantly.
Move it from being to doing
“I send the difficult email before I open anything else” is a statement about behaviour, which you can make true today. Statements about identity have to be argued for.
Write, do not chant
The studied version involves a few sentences of writing about why something matters, not high-repetition recitation. The writing is doing the work.
The honest limits
Three, briefly. The self-affirmation literature is largest in education and health contexts, not in career or money — extending it is reasonable but is an extension. Effect sizes are modest, and the field has had its share of replication scrutiny. And nothing here suggests a values exercise substitutes for treatment where treatment is what is needed.
What it does support is narrower and still worth having: a short, honest, written reminder of what you actually care about, at a moment when you are under pressure, helps more than repeating a sentence you would not defend out loud.
Finding out for yourself
The interesting thing about affirmations is that the effect depends on you — specifically on whether you believe the sentence — which means the average result across a study genuinely may not be your result.
So run it. Write down in advance what would count as this helping, in terms you could check: days you did the practice, steadiness ratings, whether you took the action you were avoiding. Thirty days. Read your own sentence back at the end.
Claims and gradings are on the sources page. If you arrived here because affirmations were not working, that piece covers the other common reasons, and if-then plans are the technique with the strongest support on this site.
If you want to find out for yourself
Run the thirty days as an experiment
Write down what would count as it working — before you start. Log six things a day. On day thirty you read your own sentence back and decide. Not us, and not a score.
It takes about ninety seconds a day. Nothing is shared, nothing is scored, and you can take everything with you at any moment.
Start the experiment See what the evidence actually says first