This study presents the results of four experiments in the framework of processing instruction conducted with four language learner groups (Spanish, "n"?=?43; German, "n"?=?46; Russian, "n"?=?44; and French, "n"?=?48; discussed in that order). In each experiment, the processing problem was held constant (the First-Noun Principle) although the structure was unique to the language. The independent variables were explicit information prior to treatment and grammatical sensitivity as measured by the Modern Language Aptitude Test. Two dependent measures were included: (a) trials-to-criterion (how long it took participants to begin processing correctly) and (b) improvement over time (pretest/posttest). Results show that explicit information may be useful for the processing of some structures but not others; however, explicit information is not necessary. Results also show that grammatical sensitivity does not significantly correlate with any of the measures under any conditions except for the German group that received explicit information. Our conclusion is that when instructed SLA is viewed as processing, variables such as explicit information and grammatical sensitivity may not play the same role as when instructed SLA is viewed as rule learning. (Contains 13 tables and 7 notes.)