Positing a morpheme-based route to complex word recognition entails not only access to morphological constituents, but also mechanisms for the initial segmentation of words into putative constituents, and compositional processes operating on these constituent representations. In this talk, I will present some recent findings on segmentation, morphological activation, and composition in complex words, including evidence from priming, lexical decision, and passive reading paradigms, in tandem with the electrophysiological brain imaging methods EEG and MEG. The results together suggest that complex word processing involves the decompositional and compositional processing of morphological-level primitives in the general case, counter to models proposing that putatively complex words are primarily or solely processed as undecomposed representations. This cross-method approach illustrates the potential for tracking the dynamics of segmentation, activation, and morpheme combination during complex word processing, toward informing a more precise neurocognitive model of word recognition.