30 31 32 33 Keywords: binocular rivalry; double pass consistency; internal noise; computational 34 modelling; bistable stimuli. 35 36 37 2 38 39 40 41 Abstract 42 43Neural systems are inherently noisy, and this noise can affect our perception from moment 44 to moment. This is particularly apparent in binocular rivalry, where our perception of 45 competing stimuli shown to the left and right eyes alternates over time in a seemingly random 46 fashion. We investigated internal noise using binocular rivalry by modulating rivalling stimuli 47 using dynamic sequences of external noise of various rates and amplitudes. As well as 48 measuring the effect on dominance durations, we repeated each external noise sequence 49 twice, and assessed the consistency of percepts across repetitions. External noise 50 modulations with standard deviations above 4% contrast increased consistency scores above 51 baseline, and were most effective at 1/8Hz. A computational model of rivalry in which internal 52 noise has a 1/f (pink) temporal amplitude spectrum, and a standard deviation of 16%, 53provided the best account of our data, and was able to correctly predict perception in 54 additional conditions. Our novel technique provides detailed estimates of the dynamic 55properties of internal noise during binocular rivalry, and by extension the stochastic processes 56 that drive our perception and other types of spontaneous brain activity.