Imitative compatibility, or automatic imitation, has been used as a measure of imitative performance and as a behavioral index of the functioning of the human mirror system (e.g., Brass, Bekkering, Wohlschlager, & Prinz, 2000; Heyes, Bird, Johnson, & Haggard, 2005; Kilner, Paulignan, & Blakemore, 2003). However, the use of imitative compatibility as a measure of imitation has been criticized on the grounds that imitative compatibility has been confounded with simple spatial compatibility (Aicken, Wilson, Williams, & Mon-Williams, 2007; Bertenthal, Longo, & Kosobud, 2006; Jansson, Wilson, Williams, & Mon-Williams, 2007). Two experiments are reported in which, in contrast with previous studies, imitative compatibility was measured on both spatially compatible and spatially incompatible trials, and imitative compatibility was shown to be present regardless of spatial compatibility. Additional features of the experiments allowed measurement of the time courses of the imitative and spatial compatibility effects both within and across trials. It was found that imitative compatibility follows a different time course from spatial compatibility, providing further evidence for their independence and supporting the use of imitative compatibility as a measure of imitation. (Contains 3 tables and 4 figures.)