PURPOSE To determine differences in preferential looking (PL) acuities using stationary and temporally modulated stripe patterns in patients with various stages of retinopathy of prematurity (ROP). METHODS We measured the PL acuities of 134 patients (ages 4 mo to 13 yr) with various stages of ROP. Patients were divided into six subgroups according to PL vision measured with stationary stripes: (1) equal to or better than 20/200 (n = 24); (2) worse than 20/200 to 20/400 (n = 10); (3) worse than 20/400 to 20/800 (n = 15); (4) worse than 20/800 to 20/1600 (n = 13); (5) worse than 20/1600 to 20/6400 (n = 26); and (6) worse than 20/6400 (n = 46; no stationary vision). RESULTS In the group with PL acuity equal to or better than 20/200, no difference in vision was apparent between the two methods. In patients with acuities worse than 20/200 to 20/400, the temporally modulated PL acuities were 0.23 octave better than the PL acuities measured with the stationary stripes. The difference increased to 0.86 and 1.12 octaves in the groups with visual acuities worse than 20/400 to 20/800 and worse than 20/800 to 20/1600, respectively. The difference in the group with PL acuities worse than 20/1600 to 20/6400 was 1.69 octaves. The 46 patients with no stationary vision detected only the temporally modulated stripes. CONCLUSIONS The results suggest that the PL acuity difference between the temporally modulated and stationary stripes increases with visual impairment. Measuring PL acuity with temporally modulated stripes is an important addition to the evaluation of severely visually impaired subjects.