Resumen de Is morphology used to encode derivations when learning a foreign language?
Derived words suggest a very efficient mnemonic when they have to be learnt as items in a foreign language (FL). Thet could be remembered by tagging in semantic memory the property that is lexicalized by the stem and storing the particular affix. A learning experiment was designed to find out wheter students make spontaneous use of this encoding strategy. The results indicated that subjects' recall performance mas better for derived words than underived ones, even when the presence of steams was not pointed out to them by the experimenter. The error data were compatible with the use of the proposed mnemonic. Surprisingly, subjects who were only native language translations did better on the derivations than those who were provided additional comment on the morphological structure of these words. This finding proves that the method of giving translations for FL words is not so bad after all and that the memory representations subjects form in such conditions are not necessarily of the paired-associate type.
