In this experiment participants memorized letter strings, formed using
a finite state grammar, and subsequently discriminated between new ''
grammatical'' items and similar ''nongrammatical'' items. Test items w
ere presented briefly in either the left or right visual field, and fo
rmed either from the same letter set as during training or a different
set. Only when items were formed from the same letters as at study, a
nd presented in the right visual filed, was discrimination performance
better than chance. These results are consistent with the concept of
an abstract visual form system that operates more effectively in the l
eft hemisphere than in the right (Marsolek, 1995).