I've been running what I'm now calling the passive window protocol for six weeks. The core change: for the first ninety seconds after recognizing the state, I have zero explicit objectives. No flying. No tests. No affirmations. No scene exploration.
Just: orient, breathe, observe, wait.
My previous approach involved immediately attempting something — usually testing the environment, sometimes trying to travel, occasionally just talking to dream characters. The success rate was low and the collapse rate was high. I was consistently losing the state within twenty to forty seconds of gaining it.
The passive window changes this. By deliberately deferring intention, I seem to allow the state to stabilize before it's stressed by goal-pursuit. The results over six weeks: average duration increased from roughly thirty seconds to approximately four minutes. Collapse rate dropped significantly.
I think what's happening is that wanting creates arousal, and arousal wakes me up. The lucid state requires something closer to relaxed attention — interested but not invested. The passive window creates that condition artificially by removing the content of wanting temporarily.
This isn't a new idea — LaBerge's stabilization techniques point in the same direction. But the specific mechanism of deferring intention for a fixed window seems to be more effective than spinning or rubbing hands, at least for me.