This article examines the use of two different Spanish language learning tasks for students identified as high ability and average ability. Students were asked to read a short passage in Spanish on a computer screen and either use the glosses already provided for certain words to aid in comprehension (the control group) or create glosses for the same target words using a Spanish-English dictionary and annotation software (the experimental group). Next, they were asked to write a synopsis of the story in their own words in English and were also given an unannounced vocabulary recall test. They were retested 1 month later for vocabulary recall. High-ability students performed significantly better with the experimental task than with the control task on the vocabulary test, while average-ability learners showed no difference for recall between the two tasks. For reading comprehension, the complex nature of the experimental task seems to have created difficulties for average-ability learners, who scored substantially lower than the other groupings. Experimental average-ability students recalled significantly fewer words on the delayed recall test, as well, again suggesting a detrimental effect of the task on word retention for this group.