Theories of translation differ in the role assigned to the reformulation process. One view, the ''horizontal'' approach, considers that translation involves on-line searches for matches between linguistic entries in the two languages involved [Gerver, D. (1976). Empirical studies of simultaneous interpretation: A review and a model. In R. W. Brislin (Ed.), "Translation: Applications and research" (pp. 165-207). New York: Gardiner]. The second view, the ''vertical'' approach, assumes that on-line reformulation does not take place while reading: translation involves giving lexical expression to the meaning extracted after comprehension [Seleskovitch, D. (1976). Interpretation: A psychological approach to translating. In R.W. Brislin (Ed.), "Translation: Applications and research" (pp. 92-116). New York: Gardner]. In four experiments, translators or bilinguals read sentences for repetition or for translation. When participants read for translation, on-line and global comprehension was affected by lexical ambiguity and memory load (Experiment 1a and 1b). Furthermore, cognate words located at the final portion of the sentences facilitated performance (Experiment 2a and 2b). However, when participants were asked to understand and repeat the sentences, lexical ambiguity and the cognate status of the words did not have any effect. This pattern of results provides support for horizontal theories of translation.