Background: The impact of aniseikonia on stereopsis has been studied for decades, however, inconsistency which may be partly attributed to the method of inducing aniseikonia exists among these findings. This study aimed to induce overall and meridional aniseikonia using a three-dimensional (3D) computer and then evaluate the effect of induced aniseikonia on distance stereopsis using contour-based and random-dot-based patterns.Methods: A 3D laptop was used to produce all of the test symbols. Unlike the usual method of creating aniseikonia with size lenses, which would change not only the size but also the relative position of the test symbols in the two images as seen by the two eyes, a new test system was designed to evaluate the aniseikonia, which only modified the size while maintaining the relative position of the test symbols. This new method reduced the disparities induced by location changing of the paired test targets when inducing aniseikonia. Aniseikonia was induced overall or in one of the meridians. The induced meridional aniseikonia included 180°, 30°, 45°, 60°, and 90°, respectively. The range of induced aniseikonia was 0-30% with an increment of 5%.Results: Overall magnification affected stereopsis more than magnification in any one of the meridians.The stereoacuity differences between oblique meridians (30°, 45°, and 60°) were not significant until the aniseikonia increased up to 20%. The difference between 180°, 45°, and 90° was significant when the aniseikonia increased up to 10% in the contour-based test and over 20% in the random-dot-based test. The stereoacuity trend was improved gradually, coupled with the angle changing from 180º to 90º in the contourbased pattern, and deteriorated gradually in the random-dot-based pattern.Conclusions: Overall aniseikonia affected stereopsis more than meridional aniseikonia. The stereoacuity of the contour-based pattern was superior to that of the random-dot-based pattern in the 90° meridian, while the opposite result was obtained in the 180° meridian.