Previous work provided corpus evidence for structural priming for specific syntactic constructions. The present paper extends these results by investigating priming effects involving arbitrary syntactic rules in spoken dialogue corpora. We demonstrate the existence of within- and between-speaker priming in both spontaneous conversation (the Switchboard corpus) and task-oriented dialogue (the Map Task corpus). We also find that between-speaker priming is stronger in the Map Task corpus. This supports the hypothesis that in task-oriented dialog, low-level priming is linked to higher-level alignment of situation models.