For one week I logged every hint I took, with one word next to it: why. The why vocabulary was tiny: stuck, bored, curious, out of time. Seven days later the log said more about my play than a month of scores.
Stuck was the honest category and the smallest one. Out of time was the largest, which was strange, because the puzzle takes two minutes and I have two minutes. What was actually happening was that I opened the hint page as soon as the solve stopped being effortless, and the stopwatch in my head was measuring patience rather than the clock.
Bored was the interesting category. A few hints were taken because the ladder had stopped being interesting, which usually meant I had already lost the frame and was now just filling in rungs. Those were the hints I regretted least.
Curious was the category that taught me something. Sometimes I wanted to see how the puzzle was built rather than how it ended, and I would take a hint with no intention of using it. That is not a hint; that is reading, and it should live next to the archive rather than the hints page.
The experiment cost seven words in a notebook and changed my behaviour for two reasons. The first is that naming the why stops the automatic click. The second is that the out-of-time entries were all emotional, not temporal, and knowing that made them easy to ignore.
If you want to run it, keep the four words handy while you play today's ladder, and write the reason before you open the hints page.
The log also explained why hints feel bad even when they help. The emotional cost is attached to the out-of-time entries, because those are the ones where the decision was not yours. Curious entries felt fine, stuck entries felt fair, out of time entries felt like a small defeat. Once the reasons were separated, so were the feelings.