Not medical or therapy advice. If you or someone you know is struggling, see Talk to someone. In a crisis in the US, call or text 988.
Spot the bias
Short scenes from the garage, school, phones and money. Pick the thinking error or influence trick at work. Each answer says why, and links to the concept.
Quiz me
Ten questions from the concepts (from the filters on the Concepts tab, if any are set). Each answer tells you why and which concept to read.
How solid is the evidence? The four labels
Psychology studies people, and people are noisy: one study with 40 college students can find almost anything by luck. So every concept here carries a label saying how well it has held up when other scientists tried it again.
| Well replicated | Found again and again, by different labs, often in large pre-registered studies or in the real world. Safe to build on, though the size of the effect may be smaller than first reported. |
| Mixed or limited | Something real is probably there, but it's smaller than the famous version, works only in some conditions, rests on few studies, or the explanation is disputed. Use it with care. |
| Failed to replicate | Big, careful attempts to repeat it didn't find it (or the original study turned out to be flawed). Don't build on it, even though it's still in books, posters and talks. |
| Myth | A popular claim that research never supported in the first place. |
The replication crisis, in plain words
Around 2011 to 2015, psychologists started re-running famous experiments, carefully and with bigger groups of people. In the best-known check, the Open Science Collaboration (2015) repeated 100 published studies: 97 of the originals had reported a "significant" result, but only about 36 of the repeats did, and the effects that did show up were on average about half as big. The Many Labs projects ran the same studies in dozens of labs at once: about 10 of 13 held up in the first round (2014), and 14 of 28 in the second (2018).
Why it happened (no villain needed, mostly): small samples, so luck looks like a finding; journals that printed exciting "it worked" results and not the boring "it didn't" ones (publication bias); and researchers trying several ways to analyze the data and reporting the one that worked ("p-hacking", often without meaning to cheat). A few cases were outright fraud.
The shop version: an intermittent fault "fixed" once proves little. You want the fault to come back on demand, the fix to cure it every time, and ideally another tech to get the same result. Science works the same way: one study is a lead, not a verdict.
What changed: pre-registration (writing down the plan and the analysis before collecting data), registered reports (journals agree to publish before the results are known), much bigger samples, and shared data. Psychology is getting more trustworthy because it caught its own problems. The same thing has happened in medicine, economics and other fields.
The letter d in some evidence boxes is a standard size measure for an effect: about 0.2 is small, 0.5 medium, 0.8 large. An effect of 0.04 is essentially nothing.
How to judge a "study shows..." claim yourself
- How many people? Dozens is a hint; thousands is evidence.
- Has anyone else found it? Especially a lab that didn't invent it.
- Is the effect too good to be true? A two-minute pose changing your hormones, or a one-hour class changing grades for years: be suspicious of big results from tiny causes.
- Who's selling it? A book, a seminar, an app. That's not proof it's wrong, but it's a reason to check.
- Lab or real life? Something that works on a computer task in a lab may be much smaller at home, at school or on the job.
How to use this page
- Search by any word (it searches the whole concept), or narrow by group and by evidence label. "Teen-relevant" picks the ones about school, phones, sleep, peers and driving.
- A concept counts as read when you press Mark as read at the bottom. The quiz and Spot the bias remember what you got right.
- Progress is kept in this browser only. Nothing is sent anywhere.
- Good for two: read one together, then each find an example from your own week. The "why" is the part worth arguing about.
Sources
Each concept lists its sources: mostly the original studies, the large replication projects (Open Science Collaboration 2015, Science 349; Many Labs 1, Klein et al. 2014, Social Psychology 45; Many Labs 2, Klein et al. 2018, Advances in Methods and Practices in Psychological Science 1; Registered Replication Reports), meta-analyses, the American Psychological Association, and standard textbooks (e.g. OpenStax Psychology 2e, 2020; Myers & DeWall, Psychology). Checked 2026-10-24. Science moves: a label can change when new large studies come in.
Talk to someone
In a crisis (United States)
- 988 Suicide & Crisis Lifeline: call or text 988 (or chat at 988lifeline.org), any time, free and confidential, in English and Spanish. For anyone thinking about suicide, in emotional distress, or worried about someone else. Veterans: call 988 and press 1. Spanish: press 2.
- Crisis Text Line: text HOME to 741741.
- Danger right now (someone is hurt, has taken something, or can't be kept safe): call 911.
When to talk to a professional
Everyone has bad days and bad weeks. It's time to talk to someone (a family doctor, a school counselor, a therapist) when:
- sadness, worry, irritability or numbness lasts most days for two weeks or more;
- it's getting in the way of school, work, sleep, eating or friendships;
- there are panic attacks, or fears that keep you from doing normal things;
- you're using alcohol, drugs, vaping, gaming or the phone to get through the day and can't cut back;
- there are thoughts of self-harm or suicide (that one doesn't wait: 988);
- something bad happened and it keeps replaying, or you just don't feel like yourself.
Warning signs of suicide (get help now)
- talking about wanting to die, being a burden, or having no reason to live;
- looking for a way to do it (searching online, getting pills or a gun);
- giving things away, saying goodbye, or putting affairs in order;
- pulling away from friends and family; drinking or using drugs more;
- big mood swings, or a sudden calm after a very low stretch.
If you see these, ask directly, stay with them, and call or text 988 together. Lock up or remove guns and medications at home: putting time and distance between a person and the means is one of the best-supported ways to save a life.
For a teen
Telling a parent, a school counselor, a coach or a doctor is a smart move, not a weakness. If a friend tells you they're thinking about hurting themselves, don't keep it secret, even if they ask you to: tell a trusted adult, or call or text 988 together.
For a parent
Asking a teen straight out "are you thinking about suicide?" does not put the idea in their head: research shows asking doesn't increase suicidal thoughts, and many people feel relief at being asked. Listen more than you fix. The family doctor is a good first stop and can refer to a counselor.
Sources: 988 Suicide & Crisis Lifeline (SAMHSA, 988lifeline.org); Crisis Text Line; National Institute of Mental Health, "Warning Signs of Suicide" and "Depression" (nimh.nih.gov); Dazzi et al. 2014, Psychological Medicine 44, review finding that asking about suicide does not increase suicidal thoughts. Means safety: Harvard T.H. Chan School of Public Health, "Means Matter", and SAMHSA. Numbers checked 2026-10-24.