Approximation of Visual Angles with binocular viewing

When viewing is binocular, which a very common experimental situation, the visual angle subtended at each of both eyes (the monocular angle) is often approximated by applying formula (1) to a virtual eye O located midway between the two eyes. This is schematically represented in Figure 1. A point O is displayed midway between the eyes separated by a typical interpupillary distance of 6.5 cm.
In this example, the exact visual angle subtended at right eye (brown angle) and left eye (angle not shown) is 19.97°.
This angular value is usually approximated by the exact angular value (here 20°) subtended at the virtual eye (point O). The difference between these two exact values is usually considered as a valid approximation for most purposes.

The difference between these two exact values becomes larger as the point M gets further from the point SO on the tangent screen. An example is shown in figure 2 where M moves along the local x axis of the tangent screen. It is the responsibility of the researcher to assess whether approximating the monocular visual angle by the angle subtended at O is still valid with large eccentricies of M.

Figure 1: Static view showing the exact visual angles subtended at virtual point O (orange) and at the right eye (brown) by the BA segment. Note the small difference between these two visual angles.	Figure 2: Animation with the midpoint M of vertical segment BA moving on the local X axis of the tangent screen. Note that the difference between both angles increases with the eccentricity of M.

😉 Remark: In the methods section of most papers using binocular viewing, the viewing distance is defined as the euclidean distance between O and SO (it is usually measured with a ruler) - and this value is assigned to the D paramter of formula 1 (see section 'Basic definition of visual angles') to measure angular values of any segment lying on the tangent screen, even when the segment's center is not coinciding with SO.