What does it mean that tight feedback loops accelerate learning?

The faster you get feedback on an action the faster you can adjust.

Example: You are learning to play guitar. In scenario one, you practice a chord progression for thirty minutes, record it, and listen back the next day. You hear mistakes but cannot remember which finger positions caused them. In scenario two, you play the same progression while watching a real-time tuner that lights up red the instant a note goes flat. Within five minutes you notice that your ring finger drifts every time you transition from G to C. You correct it on the next repetition. By the end of thirty minutes, the drift is gone. The information was identical in both scenarios — the same mistakes, the same corrections needed. What changed was the delay between action and signal. When that delay collapsed from twenty-four hours to milliseconds, learning accelerated by an order of magnitude.

Try this: Choose one skill you are actively practicing — writing, coding, speaking, cooking, anything with observable output. For the next five sessions, split each session in half. During the first half, practice as you normally would and review your performance afterward. During the second half, find a way to get immediate feedback during the practice itself: use a spell-checker as you type, run your code after every function, record yourself speaking and replay each segment immediately, taste your food at every stage of cooking. Log two things after each session: the number of errors you corrected in each half, and the subjective sense of how quickly you improved. After five sessions, compare the two halves. You are not testing whether feedback helps — you already know it does. You are calibrating your intuition for how much the speed of feedback matters.