["The Journal of Physiology, EarlyView. ", "\nAbstract figure legend We delivered trains of brief mechanical pulses to the fingertips of subjects to investigate vibrotactile intensity perception. Some stimuli consisted of bursts of up to four pulses. Simulated responses of the tactile afferent population across the hand revealed that afferents innervating the skin directly under the probe reliably followed each pulse, whereas remote afferents often failed to respond to every pulse within a burst. Because large populations of remote afferents contribute to the perception, this reduced following limits the increase in perceived intensity despite the greater number of pulses delivered.\n\n\n\n\n\n\n\n\n\nAbstract\nSinusoidal vibratory stimuli are frequently used to study human sensory perception but have the limitation that changes in vibration frequency are accompanied by changes in the number and type of activated mechanoreceptive afferents. Here we used trains of brief mechanical pulses to investigate the neural coding of vibrotactile perceived intensity by grouping pulses into bursts. These pulse trains evoke the same perceived frequency, determined by the interval between bursts, as we have previously demonstrated, and held constant across conditions. Subjects rated the perceived intensity using a magnitude estimation task for stimuli varying in the number of pulses per burst (up to four) and stimulation amplitude (5–150 µm). In marked contrast to our previous findings using electrical stimulation, increasing the number of pulses per burst had only a minimal and inconsistent effect on perceived intensity. To explain this we simulated the responses of the afferent population across the hand using the TouchSim computational model. The model revealed that increasing pulse number, without changing amplitude, produced only a modest increase in total population spike count. This occurred because the population response was dominated by large numbers of afferents remote from the stimulation site, many of which failed to respond reliably to each pulse within a burst. In contrast increasing stimulus amplitude enhanced spatial recruitment, leading to greater population spike counts and increased perceived intensity. Together these results highlight the importance of both temporal and spatial summation in shaping tactile intensity perception and argue against a ‘hot zone’ model of intensity encoding.\n\n\n\n\n\n\n\n\n\nKey points\n\nElectrical stimulation of the finger has been shown to change perceived intensity when varying the number of pulses within a stimulus burst, indicating that touch‐sensitive nerve fibres encode intensity through the number of impulses they generate within bursts of activity.\nHere we used mechanical pulses applied to the skin to induce similar bursts of activity. In contrast to expectations, increasing the number of pulses within a burst did not consistently increase perceived intensity.\nWe simulated the neural responses of the entire population of tactile nerve fibres in the hand, revealing that the burst pattern had only a small effect on nerve activity, which was dominated by large numbers of remote fibres that did not reliably follow bursts.\nThese findings argue against a ‘hot zone’ model, where intensity is determined by nerve activity near the stimulus site, and instead suggest that the number of active fibres and overall activity are most significant.\n\n\n"]