The study adopts a cognitive psychology perspective in examining strategic differences in audiovisual text processing by two groups of tertiary-level learners of Japanese as an foreign language. The learners' protocol data were collected through a concurrent think-aloud procedure. These data revealed striking differences between the Proficient group (PG) and the Less-Proficient group (LPG) group. The PG identified the meaning of key terms and evoked instantly other top-down processing strategies such as 'elaborating' 'inferencing' or 'visualising'. The LPG identified key terms in isolation, without attempting linkage to other parts of the text or to aspects of the full text. Unlike the PG, the LPG could not access this information quickly to enable them to trigger other strategies, so their interpretation was confined mostly to single words. With extensive attention to individual words, the LPG's processing became overloaded and they generally could not remember what they had comprehended. In contrast, the PG related what they comprehended and processed larger chunks of information, enabling them to interpret the text more coherently. The excerpts presented in this study offer L2 teachers and learner further insights into why less-proficient learners achieve lower levels of comprehension than those achieved by proficient learners. Implications for L2 pedagogy and further studies are discussed