We study the transient dynamics occurring in a Wilson-Cowan type model of neuronal populations to explain psychophysical masking effects. Our model network consists of a layer of excitatory and inhibitory neurons to which spatio-temporal stimuli are presented. The neurons show dynamically emerging and decaying activity. A change in the timing or the spatial context of the stimuli yields dramatic differences in the strength and duration of these transients. The observed phenomena allow for an explanation of masking experiments in which a briefly presented vernier is folllowed by a grating of a variable geometrical configuration (shine-through and feature inheritance). We find a good agreement between the persistence of the transient activity and the visibility of the vernier as reported by the observers.