Vision and perception
Perception: how the brain creates images
The eye delivers only a flat, two-dimensional image. Yet we see a rich 3D world. The brain fills in the rest, using clever rules of thumb. That explains why we effortlessly read separate LED dots as letters.
The image that lands at the back of your eye is actually flat, like a photo. Yet you see a world with depth, in which things stand in front of and behind one another. How is that possible?
Your brain does the work! It fills in what the eye cannot see, and usually guesses right. That is also why you can effortlessly read separate dots of light on a screen as letters and words: your brain connects the dots all by itself.
The optics of the eye are only the ‘mechanics’ of seeing. The real work happens in the brain, which interprets the information from the eyes. A classic example is the inverse problem: the eye forms a two-dimensional image, but we experience a three-dimensional world. Three poles of different lengths at different distances can produce exactly the same retinal image, and yet we effortlessly see them as three distinct objects.
The brain fills in the gaps
The prevailing theory, constructivism, holds that our visual system adds information to what the retina supplies, in order to choose the most likely explanation from the infinitely many possible ones. In doing so, the brain uses rules of thumb and ‘unconscious inferences’. One such rule is the minimum principle: among several possible interpretations, the brain chooses the simplest.
Why we read separate LEDs as letters
Characters on an LED screen consist of separate LEDs or clusters with space between them. That space is called the pixel pitch. Thanks to its ability to fill in information, the brain ‘connects’ those dots into legible characters. Two things help: a smaller pitch and sufficient brightness (light bridges the gaps between the LEDs). If the pitch becomes too large for the viewing distance, the text falls apart into separate dots.
The inverse problem is mathematically ill-posed: one 2D projection corresponds to infinitely many 3D scenes. The visual system resolves this by adding priors (expectations about the world), the core of constructivism. The minimum principle and other ‘unconscious inferences’ select the simplest, most likely interpretation. In addition, a ‘what’ system (object recognition) and a ‘where’ system (localisation) are distinguished.
Perceptual grouping and legibility
Merging separate LED dots into a character is a form of perceptual grouping (comparable to the Gestalt principles). The angular distance between neighbouring LEDs on the retina determines whether the brain still groups them as a single character or sees them as separate dots. That angular distance follows from pitch divided by viewing distance; brightness (luminance) lowers the threshold because light visually bridges the gaps.
Design consequence
Legibility is therefore not a purely optical property, but a perceptual one. The design choices (pitch, character height, luminance, contrast) directly steer the perceptual grouping process. This connects the psychology of perception with the concrete pixel-pitch and viewing-distance choices.