The interface between language comprehension and language production Pienie Zwitserlood For many decades, language production and comprehension have been studied separately, although both modes are part of one language faculty and evident in individual language users. This separation is more pronounced in experimental research with healthy subjects than in work on language disorders. It is particularly striking since experimental paradigms often combine production and comprehension tasks. This holds for comprehension studies using naming responses and for production studies of syntactic priming, in which sentence primes are presented for comprehension (Potter & Lombardi, 1998). It also evidently applies to the picture-word paradigm, the paradigm par excellence to study speech production "from intention to articulation" (Levelt, 1989). The paradigm combines pictures - the targets for speech production - with visual or spoken "distractor" words, which are processed by the comprehension system. It closely resembles the priming paradigm, with which many aspects of language comprehension are investigated. Models of language production and comprehension are often indifferent with respect to the question of shared or separate representations and processes. Some models make clear statements as to what is shared and what not (Levelt, Roelofs, & Meyer, 1999). Contrary to this position, our leading hypothesis is that as much as possible is shared between comprehension and production. This applies first and foremost to information about words stored in memory: the phonological specification and robustness of word forms, the syntactic properties of words, their connections or projections into conceptual memory. The idea is that stored knowledge is shared between the two language modes. (cf. Zwitserlood, 1994; Pickering & Garrod, 2003). Drawing on old and new data, the following issues will be addressed: (1) the status of morphemes in lexical processing; (2) the impact of featural mismatch between primes / distractors and targets in speaking and perception; (3) the processes of lexical selection in comprehension and production; (4) the relation between conceptual processing - as measured by means of eye movements - and particular types of information in produced or perceived utterances. The picture that arises is that the bulk of evidence is in favour of shared representations at multiple levels, but that processes operating on this information might well be different. This goes against models of production (possibly also of comprehension) that view certain representations as particular to speaking (Levelt et al., 1999). References Levelt, W.J.M. (1989). Speaking: From intention to articulation. Cambridge, MA: MIT Press. Levelt, W.J.M., Roelofs, A., & Meyer, A. (1999). A theory of lexical access in speech production. Behavioral and Brain Sciences, 22, 1-75. Pickering, M., & Garrod, S. (2004). Toward a mechanistic psychology of dialogue. Behavioral and Brain Sciences, in press. Potter, M., & Lombardi, L. (1998). Regeneration in the short-term recall of sentences. Journal of Memory and Language, 29, 633-654. Zwitserlood, P. (1994). Access to phonological form representations in language comprehension and production. In C. Clifton, L. Frazier, & K. Rayner (Eds.) Perspectives on sentence processing. Hillsdale, NY: Lawrence Erlbaum.