are processed differently from other shapes in a visual search task wh
ere subjects have to decide whether or not a predesignated target symb
ol is present in a subsequently presented string of five such symbols.
If the M-shaped letter search function, which relates correct reactio
n time to target position, reflects an efficient strategy used in word
recognition, it should be produced by skilled readers of English who
also read a logographic script. A cross-linguistic study of biscriptal
Mandarin/English and monoscriptal English readers (Expt 1) provided e
vidence of the generality of a basic search strategy for alphabetic ta
rgets. Hand-of-response affected the search function in an asymmetric
fashion for both groups of readers, and although case differences betw
een target and string increased reaction times overall, the classic M-
shaped function remained. In Expt 2, we used a within-subjects design
and examined the extent to which biscriptal Mandarin/English readers p
roduced different search functions for letters and logographs. Consist
ent with expectation, these readers showed an M-shaped function for le
tters but a more U-shaped function for logographs. Hand-of-response ex
erted a consistent effect for both types of material. Taken together,
these experiments support the view that skilled readers develop script
-specific procedures.