When people read narratives, they often need to update their situation models as the described events change. Previous research has shown little to no increases in reading times for spatial shifts but consistent increases for temporal shifts. On this basis, researchers have suggested that spatial updating does not regularly occur, whereas temporal updating does. The current study looked more deeply into this reading time pattern for spatial updating. If the prior interpretation is correct, then the absence of a reading time increase reflects a failure to update the situation model. Two experiments evaluated this claim by assessing whether other indicators of updating, namely memory probes, converge on a similar interpretation as that derived from the reading time data. Our results showed that, in contrast to previous accounts, although there was no change in the pattern of reading times, spatial updating was occurring and was extensive. As a comparison, we also looked at temporal updating. Unlike spatial updating, the temporal shifts had an influence on reading time but did not have as extensive an influence on memory probe performance. (PsycINFO Database Record (c) 2009 APA, all rights reserved).