Testing micro-sessions vs. long reviews during crunch time

I’ve tracked my retention data over the last two weeks. During a busy stretch, I moved from hour-long blocks to 3-minute bursts. My recall for abstract terms like 如何 stayed stable, but concrete nouns like 叔父 actually peaked because of the context switches. Looking for more evidence on whether high-frequency 'sprints' beat traditional sessions when the schedule gets tight.