This study aimed to investigate the effects of using online formative assessments on students' learning achievements. Using a quasi-experimental study design with one control group (no formative assessments available), and two experimental groups receiving feedback in available online formative assessments (knowledge of the correct response--KCR, or elaborated feedback--EF), it was investigated how feedback type in combination with learning content complexity will affect students' learning achievements when used in-vivo, in a digital signal processing university course. Data generated by the two experimental groups was additionally used to investigate differences in using online formative assessments based on the feedback type. Study findings suggest online formative assessments are a very efficient educational intervention for this domain. The acquired data suggests that students quickly recognized the value of the formative assessments and that more than 90% of students have used them extensively. Statistically significant improvements in learning achievements were observed in the KCR group compared to the control group (p < 0.01, Cohen's d between 0.691 and 1.080, depending on the learning content complexity), and KCR group compared to the EF group (p < 0.01, Cohen's d = 0.877, in the case of most complex of the three learning contents used). No statistically significant differences were found in formative assessment usage between the two experimental groups, aside from the difference in the time between consecutive formative assessment attempts, indicating students did make use of the available feedback. Reported results are significant for demonstrating the potential of online formative assessments in achieving the desired learning outcomes in higher education, as well as for gaining insights into students' habits of using them.