CBT thought record

What a thought record actually does

A feeling that arrives “out of nowhere” usually arrived through a sentence. Something happened, your mind narrated it in about a fifth of a second, and the narration is what you reacted to. The narration is the automatic thought, and the reason it is hard to argue with is that you did not experience it as an argument — you experienced it as the situation.

The thought record is a way of slowing that sequence down until the sentence is visible as a sentence. It does that by splitting one blurred moment into columns you have to fill in one at a time: what happened, what you thought, what you felt and how strongly, what shape the thinking took, what the actual evidence is on both sides, and what a fairer sentence would be. The columns are the whole method. There is no clever trick hiding behind them.

The target is not a nicer thought. It is a more accurate one. “Everything will be fine” is not a balanced thought, it is another unevidenced claim pointing the other way, and minds tend to reject it just as fast. “A short reply probably means she's busy; I have no evidence I'm being fired, and I'll know more tomorrow” is a balanced thought, because you could show someone the reasons for it.

Where the seven columns come from

The form descends from the Daily Record of Dysfunctional Thoughts, published in Cognitive Therapy of Depression by Aaron T. Beck, A. John Rush, Brian Shaw and Gary Emery in 1979 — a five-column sheet with situation, emotion, automatic thought, rational response, and outcome. David Burns's Feeling Good (1980) attached the now-famous list of cognitive distortions to it. Dennis Greenberger and Christine Padesky's Mind Over Mood (1995) is where the seven-column shape used above — with evidence for and evidence against as separate columns — became the standard teaching version.

That is a description of where the worksheet comes from, not a claim about what filling one in will do for you. The twelve distortion labels in step 4 are the common set from that literature; they are shorthand for shapes of reasoning, not diagnoses, and a thought can carry two or three of them at once.

A worked example

Using the placeholders already in the form:

1. Situation. “Got a one-line reply from my manager.” Facts only, as a camera would record them. Not “got a cold reply” — cold is already an interpretation and belongs in column 2.
2. Automatic thought. “She's angry with me. I'm about to be fired.” Write it in the words your head actually used, including the melodramatic ones. Sanding it down here removes the thing you are trying to look at.
3. Feeling and intensity. Anxious, 70 out of 100. One word for the feeling; the number is your own read, right now.
4. Distortions. Mind-reading (you assigned her an emotion you have no access to) and fortune-telling (you assigned a future). Naming them is not disproving them. It is noticing which move your mind made.
5. Evidence for. “The reply was short.” Be fair here — a record where column 5 is empty is a record you were never going to believe.
6. Evidence against. “She's always terse over email. My last review was good. Short ≠ angry.” This is the column that does the work, and the one people skip.
7. Balanced thought and re-rating. “A short reply probably means she's busy, not furious. I have no real evidence I'm being fired.” Then re-rate the same feeling. The second number is not a score. It is a prompt to check in.

Ten minutes is a realistic first pass. Written records are slower than doing it in your head, and that slowness is the point — a thought you have to type out in full is much harder to skate over.

Honest limits

• The 0–100 ratings are self-report, not measurement. Same rater, minutes apart, who knows what the exercise is supposed to do. Read the pair as a check-in, not as a result, and expect the number to move for reasons that have nothing to do with the worksheet.
• The number sometimes goes up. Writing out a thought you have been half-avoiding can raise the intensity before anything else happens. That is unremarkable, not a sign you filled it in wrong.
• It does not work on true thoughts. If the evidence-for column is strong and the evidence-against column is thin, the thought may simply be correct. A real deadline, a real conflict and a real diagnosis are not distortions, and they want a plan, not a reframe.
• It is not a mood tracker. There is no chart, no streak and no trend line here, on purpose. Saved entries exist so you can re-read what you wrote last time, not so you can be scored.
• Written records are the slow version. Practitioners generally treat the worksheet as training wheels for doing the same thing in your head. If you are filling in six a day, that is worth mentioning to a professional rather than optimising.
• It is not treatment. This is a self-help worksheet, not therapy, not a diagnosis, and not a substitute for care you are already getting.

What happens to what you type

Nothing you type is uploaded. When you press Save thought record, the entry is written to your browser's own localStorage, under the key na.records, and it is stored only in this browser — there is no server copy, no account, and no sync to your phone or to another browser on the same machine. Reload the page, close the tab, restart the computer: the entries are still there. Clear your browsing data or site data for this domain and they are gone, permanently, with no backup and no export. Private or incognito windows discard the storage when the window closes, and if your browser blocks site storage entirely the save will fail silently and the list will keep saying it is empty.

The newest 50 entries are kept; beyond that the oldest one drops off each time you save. Clear all saved asks for confirmation and then empties the key. Individual entries have their own delete button. The saved list shows the situation, the automatic thought, the distortions, the balanced view and the before-and-after numbers — the two evidence columns are stored but not re-displayed, to keep the list scannable. The one network request this page makes is a self-hosted page-view counter that records the URL, never your text.

Related tools and reading

Weighing the evidence for a thought is cognitive restructuring, from classical CBT. The other move — leaving the content alone and loosening your grip on it instead — is cognitive defusion, from ACT, and it has its own page: the thought defuser. Both are useful, and doing one while believing you are doing the other tends to end in an argument with your own head. The tools page also has a box-breathing pacer and a focus timer, and the field guide explains the psychology underneath all four in plain language, including where the evidence is thin.