The present studies examined the effects of varying degrees of unfamiliar vocabulary within written discourse on individuals' abilities to use linguistic context for the purposes of translation and comprehension (i.e., lexical inferencing). Prose varied in the number of foreign words introduced into each sentence (e.g., 0 through 7 content words per sentence). Furthermore, Krashen's Input Hypothesis and the Evaluation component of the Involvement Load Hypothesis were tested to determine the degree at which non-comprehensible input hinders the ability of a learner to successfully use linguistic context for translation and comprehension. Results indicated that, as the number of foreign words per sentence, i.e., non-comprehensible input, increased the ability to successfully translate foreign words and create situational models for comprehension begins to decrease especially beyond five unfamiliar words per sentence. This result suggests that there is an optimal level of effectiveness in the use of a linguistic context strategy for learning foreign language vocabulary, but also that there is a limit to the strategy's effectiveness. Implications and applications to the field of foreign language learning are discussed.