Recognition of briefly exposed digits, Latin letters, and Chinese characters: Evidence for language-specific differences in encoding and rehearsal

Abstract

German and Chinese students took part in four experiments. In Experiment 1, which was based on the Sperling paradigm, the participants were asked to report as many digits as possible from a briefly presented matrix. In contrast to previous experiments which had made use of Latin letters, the Chinese clearly outperformed the Germans this time. In Experiment 2, memory span for digits and Latin letters was measured. With Latin letters, the Germans achieved higher scores than the Chinese. The opposite was true for digit memory span. Taken together, the results indicate that performance in the Sperling task is not solely determined by encoding efficiency but also by rehearsal speed. In Experiment 3, the participants had to recognize only one item from briefly exposed stimulus displays. Besides Latin letters, language-specific stimuli were used consisting of combinations of letters for Germans and components of Chinese characters for Chinese. Based on the data from Experiment 3, two language-specific item sets comperable in terms of task difficulty, were chosen for use in the Sperling task in Experiment 4. Under these conditions, there was no significant difference in the overall level of performance between the two language groups. Performance was influenced, however, by the position of the items to be reported and this in a slightly different way in each language group. We discuss to what extent reading habits influenced the allocation of attention and adaptive processes caused differences in item recognition.