Adding Light:
Photic Stimulation

The strongest signal in the entrainment literature does not come from audio. It comes from light. Rhythmic visual stimulation produces a steady-state visually evoked potential (SSVEP) — a measurable electrical response in visual cortex that follows the stimulation frequency with a reliability audio-only methods do not match. Photic driving has been observed on EEG since the 1930s and is robust enough that clinicians use it as a standard provocation test.

The most cited recent work is the 40 Hz line of research (MIT's GENUS studies), where combined light and sound at 40 Hz produced measurable gamma-band entrainment in humans using broadband white light. That research targets cognitive decline and is not what Phantas does or claims — but it established the delivery mechanism: synchronized light and sound entrain more strongly than either channel alone.

Phantas has always been honest that audio-only entrainment evidence is mixed. Adding the visual channel moves the protocols onto the stronger branch of the literature.

A display can only produce clean flicker at rates that divide its refresh rate exactly. A 60 Hz screen can flash at 10, 12, 15, 20, or 30 Hz. True 40 Hz requires a 120 or 240 Hz display. Anything in between produces beat artifacts against the refresh cycle — a degraded stimulus and a less comfortable one.

So Phantas measures your display first. Every induction begins with a calibration pass that measures the real refresh rate and offers only the rates your screen can produce exactly. If your display cannot serve a protocol's frequency — 40 Hz GAMMA on a standard 60 Hz screen, for instance — the session says so plainly and runs audio-only. Nothing is approximated, ever.

The stimulus itself is a full-field square wave at 50% duty cycle, frame-counted against the display clock, with luminance ramps at entry and exit. The engine tracks its own dropped frames and reports delivery quality per session.

You cannot watch a flashing screen and work at the same time — and you do not need to. Sessions split into two phases: a short induction (five minutes by default, eyes closed — the stimulus works through closed eyelids) with synchronized light and audio, then an open-ended sustain phase where the audio continues alone while you work.

The induction is deliberately fixed and brief: a clear boundary between "preparing the state" and "using it." Start with the default, keep your own judgment: the session log records what you notice.

Flashing light carries a real risk for people with photosensitive epilepsy. Phantas treats that as a hard engineering constraint, not a disclaimer:

• — A screening runs before the first induction. Any history of seizures, family photosensitive epilepsy, or prior discomfort from flashing light disables the stimulation channel entirely — the full audio product remains.
• — Any tap or keypress cuts the light to black on the next frame.
• — Luminance always ramps in and out; saturated red — flagged in broadcast and accessibility flash guidelines — is excluded from the stimulus palette.
• — Induction length is capped, and the light never starts without an explicit begin action.