Summary

Theories of speech articulation planning must account for the coordinated movements of speech articulators that produce contrastive lexical items, as well as the ways these vary systematically in different contexts both within and across utterances. The systematic phonetic variability of phonologically equivalent forms relates to many factors including adjacent segmental context, rate and style of speech, local contextual factors, such as position in prosodic structure. This variability has been proposed to contribute to the transmission of meaning and to communicative efficiency.

Theories proposed to account for speech articulation behaviors differ in many key respects. Theoretical controversies relate to the nature of phonological representations used to distinguish words in the mental lexicon, as well as the processing components, representations, and mechanisms involved in planning and producing speech.

One of the most fundamental theoretical distinctions relates to assumptions about the spatiotemporal vs. symbolic nature of phonological representations. Distinctions in system architecture, including the number and type of processing components, derive from these assumptions. Other differences among theories relate to assumptions about the nature of speech production goals, movement targets, timekeeping mechanisms, dynamic parametric models of movement, movement coordination, the separate vs. integrated representation for spatial and temporal characteristics of movement, as well as proposed mechanisms to account for systematic phonetic variability in different contexts (coupled oscillators, optimization-based, vs. dynamic field theory approaches).