Common-onset masking (COM) refers to a methodology where a mask can impair awareness of an object if the mask’s offset is delayed relative to the offset of the object. This method has classically been used to understand how discontinuities in visual input lead to the discrete removal of object representations before they reach conscious awareness. However, COM has recently been shown to reduce the precision of conscious object representations (Harrison, Rajsic, & Wilson, Psychonomic Bulletin & Review, 23(1), 180–186, 2016). As a result, Harrison et al. proposed that COM shortens the temporal window for perceptual sampling of an object’s representation, an account consistent with interruption-based theories of masking. In the present study we modified the standard COM methodology to assess the impact of a delayed mask offset on the temporal perception of an object’s representation. Across two experiments we provide novel evidence that a delayed mask offset can impair temporal perception of a conscious percept, such that it reduces the percept’s perceived duration (Experiment 1), and prematurely terminates updating of the percept’s dynamic orientation (Experiment 2). We refer to these results as temporal trimming, and suggest that the mechanism responsible for COM operates during the sustained perception of an object.