ABSTRACT
This chapter reviews some of the central questions pertaining to speech monitoring and the existing theories in the light of recent empirical studies. Temporal analyses indicate that self-repair can occur rather rapidly, demonstrating that speakers can use an internal monitoring channel, in addition to an external (auditory) channel. Both patient data and the comparison of error patterns to repair patterns suggest that there is an essential distinction between these internal and external monitoring channels. Some theoretical implications of these observations are discussed, in particular regarding the normal operation of the perceptual inner loop; the existence of fast, autonomous (production-based) monitoring devices; and the possibility of shifts in the division of labor between external and internal monitoring. That such shifts can occur follows from the changes in monitoring foci reported in the literature.
