One aspect of second language teaching via computerized language learning programs that has received great attention over the past few years is helping vocabulary learning through link technology. The present study was conducted to examine the effect of two types of link (repetitive textual and in-text reference) on vocabulary learning and retention and reading comprehension. The design of this quasi-experimental study was pretest-posttest with two independent groups in which 35 undergraduate students in two experimental groups participated in the study and were asked to read a computerized text. Two different versions of the reading text were developed for the experimental groups involved in the study. In one version of the computerized texts, the repetitive textual links, and in the other, in-text reference links were used. Having read the text, the participants took an immediate vocabulary posttest as well as a reading comprehension test. Three weeks following the exposure, the participants in the two groups were asked to complete a delayed vocabulary posttest in order to measure their retention level of the target words. Data obtained from pre-test and post-tests were analyzed using t test at p<0.05 significance level for two independent groups. The mean score of in-text reference links group in learning vocabulary posttest was 16.11 (SD=2.45), in vocabulary retention posttest 13.66 (SD=1.86) and in reading comprehension was 10.11 (SD=1.41).The mean score of repetitive textual links group in learning vocabulary posttest was 12.06 (SD=2.14), in vocabulary retention posttest 8.29 (SD=1.53) and in reading comprehension was 9.29 (SD=2.02).The result of the study indicated that the participants in the group with in-text reference links outperformed those in the group with repetitive textual links in terms of vocabulary learning and retention; however, no statistical significance in the participants’ reading comprehension level was detected. Based on levels-of-processing theory, processing target words at different levels would lead to a differential rate of learning and retention of those words. Put another way, those target words that are deeply processed would be learned better and retained longer in contrast to those which are only treated at shallow levels of processing