The effect you see in your example has many permutations. Albers walks the reader/student through them via a precise and well-crafted set of experiments. His book (now over 50 years old) remains the quintessential guide to understanding the relative nature of color.
As an introduction, Albers writes (emphasis mine):
In visual perception a color is almost never seen as it really is — as it physically is. This fact makes color the most relative medium in art.
Elsewhere, Albers quotes Kandinsky to emphasize the plasticity of color:
What counts is not the what but the how.
The material values of any color (not just value changes within a single tone) are deeply influenced by environment.
Elsewhere in the book he points to the fact that we can isolate things like musical notes to be heard (mostly) without comparative influence. Color is much more difficult to isolate and is, thus, always viewed within the context of relative interaction.
How do we make 1 color 2
In your example, the adjoining light/dark relationships relatively magnify each other. As your eye moves across the color plane it is next effected in the reverse by the value on the opposite side. In between, your eye blends the difference.
To make those shifts even across a space, you have to properly measure the steps. Albers also identifies the guiding principal behind this problem: The Weber-Fechner Law. In short, this tells us that even mathematical progression will yield an uneven gradation. In other words, if the first step moves from 10% brightness to 20%, choosing 30% next will cause your differential to drop off. The proper value should be selected by proportional increase, ie 40%.
The exciting part is that you can bend color in much greater ways by manipulating the same comparative factors. The original cover (above) perfectly articulates the basic concept.