Signal encoding is the form visual information takes before it reaches a physical display or biological interface.
| Encoding | Description | Example |
|---|---|---|
| Analog waveform | Continuous voltage changes over time. | Broadcast TV brightness signal. |
| Raster scan | A time sequence mapped to screen position. | Scanned display lines. |
| Pixel framebuffer | Digital array of color and brightness values. | LCD, OLED, phone screens. |
| Pulse train | Information carried by pulse timing and amplitude. | Neural stimulation patterns. |
| Event stream | Only changes are transmitted. | Event cameras and motion-first sensing. |
| Feature code | Edges, orientation, motion, color, depth, or texture. | Vision models and biological receptive fields. |
| Semantic code | Objects, labels, hazards, spatial commands. | Assistive navigation systems. |
| Memory cue | Prompt that triggers recall rather than a direct image. | Photos, labels, narration, location cues. |
Natural displays use pixels because screens are surfaces. The visual system does not receive pixels internally. It receives transformed signals: contrast, timing, receptive-field activity, motion, color opponency, attention, and prediction.
A screenless interface may become more practical by sending less:
full image -> edges + movement + depth + hazards
or:
full image -> person, doorway, stairs, text, vehicle, motion direction
This is not as cinematic as a normal picture, but it may be more useful.
| Pattern | Purpose |
|---|---|
| Foveated encoding | Spend most bandwidth on the attended region. |
| Peripheral alerting | Use coarse cues for motion and hazards outside focus. |
| Edge-first encoding | Prioritize contours and object boundaries. |
| Depth-first encoding | Prioritize distance and collision risk. |
| Symbol encoding | Use learned glyphs instead of photorealism. |
| Task-aware encoding | Change what is sent based on reading, walking, faces, or tools. |
| Confidence encoding | Signal uncertainty so the user can distrust weak recognitions. |
| Adaptive encoding | Recalibrate to the user’s reported percepts over time. |
A perceptual codec is a hypothetical compression system designed around the user’s actual perceptual interface. It would not optimize for image quality on a monitor. It would optimize for:
The output might look like a small vocabulary of visual-like sensations, sounds, or tactile cues rather than a normal video image.