No document available.
Abstract :
[en] Neuropsychological case-studies suggested that dates and encyclopedic numbers may be processed differently than unknown numbers. However, this issue was not yet investigated in reading in healthy participants, so that it is unclear if dates are read like words and processed as lexical items, or like numbers where each position strictly defines the digit value in a base-10 system.
Here, we compared processing of known dates to unknown numbers in a group of 26 experts (students and teachers in History). Participants performed an explicit recognition task on dates (e.g., 1789, 1945, …) and on acronyms (e.g., FNRS, HDMI, …), half known and half unknown. They were preceded by an identical prime (e.g., 1945-1945), a transposed-character prime (e.g., 1495-1945) or a substituted-character prime (e.g., 1635-1945).
Results show that for dates, there is a significant transposition gain (-57ms), while for unknown numbers as well as for acronyms (known and unknown), the transposed-character prime induced a cost (from +17 to +257ms) rather than a gain. The facilitation due to transposed characters found here on dates is similar to what is observed in studies of lexical decision on words. Therefore, it suggests that dates may be processed with similar types of orthographical mechanisms than words.