About the influence of the presentation format on arithmetical-fact retrieval processes.

This article presents the results of two experiments. In Experiment 1, French-speaking participants were asked first to retrieve the product of two numbers presented in Arabic or verbal code, and then to perform a number-matching task on the same material to assess the encoding time difference between numerals in the two formats. Experiment 2 involved the same multiplication task with Dutch-speaking participants who name two-digit numbers in reverse order. The format effects obtained by Campbell and Clark (1992); Campbell (1994) for multiplication were replicated. However, several observations suggest that some of these effects may be due to encoding time differences between word and digit numerals. The same size-by-format interaction was found for the number-matching task as for the multiplication task, and the effect disappeared with practice in the multiplication task. Finally, despite the fact that the linguistic structure of number names differs between French and Dutch, the types of error produced in both groups were identical. The last result does not match with the hypothesis that operand intrusion errors are due to interference between reading processes and arithmetical-fact retrieval processes. Implications of these findings for the debate about the nature of arithmetical-fact retrieval are discussed.