The purpose of this study was to examine two scaffolding methods on the performance of students in computer-based concept linking and retention of comprehension. After training and practice in concept mapping and CmapTools--a computer-based concept mapping program, 116 undergraduate students were randomly assigned to one of four treatment groups to work on a computer-based concept mapping task. Students in the no scaffolding (NS) group did not receive any scaffolding. Students in the linking phrase scaffolding (PS) group received linking words or phrases as scaffolding. Students in the articulation hint scaffolding (AS) group received a hint question as scaffolding, which asked them to elaborate on relationships between concepts in full sentences. Students in the linking phrase and articulation hint scaffolding (PAS) group received both scaffolding while working on the computer-based concept mapping task. One week after the treatment, students took a concept linking posttest, in which they constructed a concept map in CmapTools based on a web-based instruction on the human heart. After another week, they took another posttest on retention of comprehension about the heart. Two 2 X 2 factorial Analysis of Variance (ANOVA) were conducted to examine the main effects of linking phrase scaffolding and articulation hint scaffolding and any interaction effect between them on the performance of students in computer-based concept linking and retention of comprehension. The results showed no significant difference in the performance of students in both tests. However, the Pearson's correlation analysis showed that there was a positive correlation between students' performance in computer-based concept linking and retention of comprehension (γ = 0.447, p < 0.01).