The Screenless Sight Codex

Research Questions

This page collects questions that can drive future encyclopedia expansion.

Display Transduction Foundations

• Which principles from CRTs and other display systems remain useful after a conventional screen is removed?
• Is scanning a better interface pattern than pixel arrays for low-channel artificial vision?
• How much of a visual field can be reconstructed from timed point events?
• What biological or perceptual effects behave like persistence?
• Can temporal integration be used as a design principle for prosthetic cues?
• How do raster, spiral, edge-trace, and symbol-trace scanning compare?
• What does color-channel separation in display systems suggest about feature-channel separation in artificial perception?

Optical And Near-Eye Interfaces

• When does “screenless” mean no screen, and when does it mean a screen moved into the optical path?
• Can retinal projection be more useful as an overlay than as full replacement?
• How should a display use eye motion instead of fighting it?
• Can foveated rendering reduce bandwidth enough to help low-power systems?
• What tasks benefit most from retinal projection: reading, navigation, object-finding, or augmented memory?
• Can adaptive optics make personal retinal displays clearer?

Retina And Retinal Prostheses

• Which retinal layer is the most useful target for artificial input?
• How should disease state change the encoding strategy?
• Can a retinal prosthesis use a symbol vocabulary rather than crude pixels?
• Which features should be encoded first: brightness, edges, motion, depth, or object labels?
• How stable are perceived phosphenes over time?
• Can retinal stimulation be paired with audio labels during training?
• Can a personal percept diary improve calibration?
• What would a retina-matched encoder look like for different retinal diseases?

Optogenetic Vision

• Should optogenetic interfaces aim for natural images or artificial visual symbols?
• Which modified cell classes provide the most useful response?
• How does response speed change what visual features can be encoded?
• Can optogenetic systems exploit high-contrast edge maps better than full images?
• What role should external goggles or processors play?
• How should training differ from electrode-based retinal prostheses?

Optic Nerve, LGN, And Cortex

• Is optic-nerve stimulation better treated as a cable problem or a code problem?
• What makes LGN attractive despite being a difficult target?
• How display-like is V1 retinotopy in practice?
• Can cortical form tracing outperform simultaneous dot stimulation?
• Could a cortical system use a learned alphabet of phosphenes?
• How should a cortical interface represent confidence?
• Can dynamic lines, sweeps, or motion cues bind better than static dots?
• What should be routed to V1 versus higher semantic systems?

Magnetic, RF, And Ultrasound Carriers

• Is the carrier best used for stimulation, power, synchronization, or local transducer control?
• When does a remote-field concept become more concrete by adding a local receiver?
• Can magnetic or ultrasound methods provide useful low-symbol cueing?
• Can these carriers function as gating systems paired with another output channel?
• What is the difference between evoking sensation and carrying information?
• How should “addressability” be defined for noninvasive methods?

Sensory Substitution

• Which visual features translate best into sound?
• Which translate best into touch?
• Can audio carry object identity while touch carries spatial urgency?
• How much training turns substitution into perception-like experience?
• Can event-based inputs reduce sensory overload?
• Should substitution systems be designed as languages, maps, or instruments?
• What is the right cue grammar for space, motion, confidence, and urgency?

Semantic And AI-Assisted Interfaces

• When is a label more useful than an image?
• What should an AI scene compressor discard?
• How should the system represent uncertainty?
• Can task modes prevent overload?
• What should be prioritized during walking, reading, cooking, social interaction, or tool use?
• Can a semantic display become a “second perception” with training?
• How should user preferences shape what the system considers important?

Perception As Medium And Project Tracks

• Which middle layers can be removed without reducing usefulness?
• When does a cue become something the user perceives through rather than an alert the user must consciously decode?
• How much active movement or control is needed for a cue language to become fluent?
• Can a five-cue scene description outperform a noisy low-resolution image?
• Which cue grammars best split information across vision, audio, touch, and symbols?
• How should personal percept maps track drift, fatigue, context, and learning?
• What should count as success: recognition, navigation, confidence, speed, comfort, or image likeness?
• Can a perception-first prototype be tested safely before making any neural or medical claim?

Memory, Imagery, And Conceptual Interfaces

• What is the difference between perception, imagery, recall, and belief?
• Can visual cue systems strengthen recall without distorting it?
• Which cues improve accuracy versus only vividness?
• How do images, narration, places, music, and emotion tags differ as memory anchors?
• Can a memory lantern system help users reconstruct scenes voluntarily?
• Can dream rehearsal be studied as cue timing and imagery practice rather than direct dream control?
• What would a true memory-state interface need to represent?

Toy Models Worth Building Later

• Low-resolution phosphene grid simulator.
• Dynamic shape-tracing simulator.
• Phosphene font trainer.
• Event-based scene stream visualizer.
• Scene skeleton annotator.
• Cross-modal cue trainer.
• Memory lantern card format.
• Retinotopic address-bus diagram.
• Neural shadow-mask crosstalk simulator.
• Semantic subtitle track demo using ordinary video.
• Perceptual codec bench.
• Active scan trainer.
• Confidence grammar test.