Synaptic high-frequency jumping synchronises vision to high-speed behaviour

During high-speed behaviour, animals must synchronise perception and action despite rapid environmental and self-generated motion. How neural systems achieve such precision remains unclear. Here we show how the housefly (Musca domestica) maintains visual accuracy during fast motion. Using intracellular and photomechanical recordings during saccade-like stimulation, we traced information flow from photoreceptors to large monopolar cells (LMCs). Visual neurons achieved record-high information sampling ([~]2,500 bits{middle dot}s-1) and synaptic transmission ([~]4,100 bits{middle dot}s-1), far exceeding previous estimates. We identify a previously unknown mechanism - synaptic high-frequency jumping - in which photoreceptor-LMC synapses dynamically shift transmission toward higher frequencies during saccades, extending visual bandwidth to [~]1,000 Hz, effectively eliminating synaptic delays, and quadrupling classical flicker-fusion limits ([~]230 Hz). Behavioural experiments show flies respond synchronously within [~]13-20 ms, even before photoreceptor responses peak. A biophysically realistic model reveals how photomechanical-stochastic-refractory quantal sampling co-adapts with saccadic behaviour: through self-motion, flies efficiently translate image motion into temporally-precise, predictive high-speed vision.