The inspector who can't sign off
Why can't I stop checking my work for mistakes?
Your perfectionism isn't vanity. It's care that forgot to include a stopping point.
Short answer
Scanning for mistakes under pressure is common, and it often starts as care. Lab studies show being evaluated raises the stress response, and stress narrows attention, which may be part of why checking gets louder when work feels watched. Our model's view is that the problem is a missing stop rule, not the standards; that is untested. Writing good-enough criteria before you start is a reasonable experiment.
Part of a bigger question: Why can’t I switch off after work?
The work is finished. You know it's finished. And you're reading it through one more time, because there might be a comma, a tone, a thing someone could point at, and if there's one more thing to catch, you haven't earned the right to stop yet.
You think this is just being thorough. Often it is. But notice when the scanning gets loud: under pressure, when the stakes feel watched. That's when thoroughness can quietly turn into something that doesn't finish.
The dial that only turns up
Here's what may be running. Under pressure, the error-scanner turns up the gain, hunting harder for the flaw that might get seen, the mistake that might invite criticism. Being judged is one of the more reliable stress triggers studied: in lab tasks, having an audience evaluate you raises the body's stress response more than doing the same task unobserved. Strong evidence Stress also tends to narrow attention toward whatever feels like a threat, which may be part of why the scanner locks on to flaws. Limited evidence Often that is care, pointed at the work and the people who'll judge it. Sometimes it is fear of one particular person. Sometimes the stakes are high and the extra pass is right.
Our model's hypothesis is structural. A scanner set to find flaws can keep finding one more, because "flawless" isn't a state that exists. So the work may struggle to be finished not because the standards are wrong, but because nothing told them when to stop. Speculative That is our reading, and a testable one; the research above shows the pressure, not the missing off-switch.
The hidden cost
Look at where your best energy goes. It goes into preventing mistakes, including ones nobody else was going to notice, or make, or mind. Some of those hours go into catching flaws that would have caused no harm if they'd slipped through. How many is worth counting rather than guessing.
So the cost isn't poor work. It's brilliant, exhausting over-delivery on the parts that didn't need it, paid for out of the energy you no longer have for what does.
What helps, what backfires
What tends to backfire is vague excellence ("make it great," "as good as it can be"), because a standard with no edge gives the scanner infinite room. So does praise that only rewards perfection; it can teach the dial there's no safe place to stop.
What tends to help is deciding what "done" means before you begin. Set the good-enough criteria up front, in writing, while you're calm. Then when the work meets them, done is allowed to be done. This acts a bit like an if-then plan (if it meets these three things, I send it), and plans made in advance tend to help people do what they decided while calm, though no study has tested this exact use. Limited evidence You're not lowering your standards. You're giving them a finish line, so care has somewhere to rest.
What to watch
Try comparing. For the next five pieces of work, count the full read-throughs and note what the last two actually caught. On two of them, write three done criteria before you start. If the later passes rarely catch anything that matters, and the criteria cut the passes without anyone noticing a drop in quality, the missing stop rule fits you. If the late passes keep catching real errors, your checking is doing its job.
If the extra rounds of checking are adding up across the week, the fatigue measure looks at how small repeated loads accumulate, and the work-stress material quiz maps how work pressure tends to land for you.
Decide what done means before you begin, and let done be done.
Your perfectionism isn't vanity. It's care that forgot to include a stopping point.
Keep investigating
Where this fits: the mechanism behind it, a material that responds differently, and the Field Guide it belongs to.
What the evidence can't tell you Lab studies show evaluation raises stress and narrows attention; they cannot set the right stopping point for your work or rule out a clinical checking pattern.
Take it further
Measures this touches
- How repeats wear on you (Fatigue) How much many small, repeated stresses add up for you, from a drip that gradually wears you down to small things rolling off.
- How much it takes to move you (Stiffness) How strongly you hold to a set plan, route or expectation once it is in place, and how much changing it costs.
Situations
Claims and evidence
Each claim in this article, as an observed pattern, a plausible mechanism and the strength of the evidence.
-
Observed in studies Strong
- What studies found
- In lab studies, tasks where people are judged by others raise the stress response more than the same tasks done without evaluation.
- One explanation
- One explanation is that social-evaluative threat is among the more reliable triggers of the stress response, which may be part of why checking intensifies when work feels watched.
- What it does not show
- Short lab tasks measuring cortisol; they did not measure checking behaviour or perfectionism.
- Evidence
- Large reviews or meta-analyses agree on the pattern.
Dickerson & Kemeny 2004Meta-analysis
- Who
- 208 laboratory studies
- Conditions
- Short psychological stressors in the laboratory
- Method
- Pooled cortisol changes across studies
- Observed
- Cortisol rose most when the task was uncontrollable and others could judge performance, and then took longest to recover.
- Design
- Meta-analysis
Acute stressors and cortisol responses: a theoretical integration and synthesis of laboratory research. Psychological Bulletin. doi:10.1037/0033-2909.130.3.355
-
Observed in studies Limited
- What a few studies found
- Under stress, attention tends to narrow toward what feels threatening.
- One possible explanation
- Narrowed attention may lock the scanner onto possible flaws; much of the underlying evidence is older or from animal work.
- What it does not show
- Older human work and animal studies; not a direct study of error-checking at work.
- Evidence
- A few studies, often small or correlational. Plausible, not settled.
Easterbrook 1959Theory
- Design
- Theory (summarised from the title; no abstract available)
The effect of emotion on cue utilization and the organization of behavior. Psychological Review. doi:10.1037/h0047707
Arnsten 2009Narrative review
- Conditions
- Acute and prolonged stress
- Method
- Review of brain signalling research
- Observed
- Even mild uncontrollable stress quickly impaired the prefrontal cortex, the brain's planning area, and prolonged stress changed its structure.
- Design
- Narrative review
Stress signalling pathways that impair prefrontal cortex structure and function. Nature Reviews Neuroscience. doi:10.1038/nrn2648
-
Observed in studies Limited
- What a few studies found
- In studies across many goals, people who made specific if-then plans in advance followed through more often.
- One possible explanation
- Pre-set done criteria may work as an if-then plan for stopping; this application has not been tested directly.
- What it does not show
- The studies tested follow-through on intended actions; none tested pre-set done criteria for ending a checking loop.
- Evidence
- A few studies, often small or correlational. Plausible, not settled.
Gollwitzer & Sheeran 2006Meta-analysis
- Who
- 94 independent tests
- Conditions
- Making if-then plans for when, where and how to act
- Method
- Pooled effects on reaching goals
- Observed
- If-then plans had a medium to large positive effect on reaching goals.
- Design
- Meta-analysis
Implementation intentions and goal achievement: a meta-analysis of effects and processes. Advances in Experimental Social Psychology. doi:10.1016/S0065-2601(06)38002-1
-
Calmistry hypothesis Speculative
- What our model proposes
- Our model predicts that error-scanning without a defined finish line tends not to end on its own.
- Why we think so
- A Calmistry hypothesis: the trouble is a missing stop rule, not the standards themselves.
- Evidence
- An idea or metaphor that fits what people describe, with little or no direct research behind it yet.
Sources
- Arnsten 2009. Stress signalling pathways that impair prefrontal cortex structure and function. Nature Reviews Neuroscience. doi:10.1038/nrn2648
- Dickerson & Kemeny 2004. Acute stressors and cortisol responses: a theoretical integration and synthesis of laboratory research. Psychological Bulletin. doi:10.1037/0033-2909.130.3.355
- Easterbrook 1959. The effect of emotion on cue utilization and the organization of behavior. Psychological Review. doi:10.1037/h0047707
- Gollwitzer & Sheeran 2006. Implementation intentions and goal achievement: a meta-analysis of effects and processes. Advances in Experimental Social Psychology. doi:10.1016/S0065-2601(06)38002-1
Situation quiz
Your work-stress material
You know how many times you checked this piece of work. Does the same need to check show up in other parts of your life?
About 12 minutes. It describes a tendency across situations, not a fixed identity, and it can change.