Thought Toys · Waves & rhythm · Exhibit 62

Ask which slit, lose the wave

Fire particles through two slits one at a time, each landing somewhere unpredictable — and yet, hit by hit, they pile up into a striped interference pattern with spots that are never hit at all. Learn which slit each one used, and those forbidden spots fill back in, exactly.

Particles land one at a time; watch the screen build up

Fired: 0Mode: no marker

Which-path information
Fire particles
your turn — add the marker and watch the stripes flatten

What you're seeing

Two narrow slits sit side by side in a barrier; particles are fired at them one at a time, and each one lands at a single, definite spot on the screen behind — there is nothing spread-out or fuzzy about any individual hit. What's strange is the shape the hits make together. Fire enough of them with no marker and the landing spots pile up into stripes: bright bands where hits pile up fast, and dark gaps in between where — no matter how long you run it — nothing ever lands. That's the signature of a wave: the two slits act like two coherent sources, and at the dark gaps their ripples arrive exactly out of step and cancel. A single particle, apparently, has no trouble taking both paths at once, in the sense that matters for where it can land — right up until you check.

Add a which-path marker and fire again: now there's a fact about which slit each particle used — a real, in-principle-readable record, whether or not anyone ever looks at it. The stripes don't fade gradually; they vanish outright. The new pattern is just the plain sum of what each slit would make on its own — the two single-slit humps added together, dark gaps included nowhere. Nothing about the slits or the particles changed. The only thing that changed is that the two routes stopped being alternatives that could interfere and became a fact with an answer.

The rule, exactly. Standard two-slit Fraunhofer diffraction (slit width a, separation d, screen distance L, wavelength λ; position x on the screen; this toy is a schematic model of the outcome statistics, not a full quantum-field simulation): β(x) = π·a·x / (λL)    δ(x) = π·d·x / (λL) no marker: I(x) = 4·sinc(β)²·cos(δ)²    marker on: I(x) = 2·sinc(β)² The no-marker curve hits exact zeros wherever cos(δ)=0 — true dark fringes, not just dim spots. Verified in node (improve/verify/62-double-slit.js): those zeros land at the predicted x to machine precision while the marker-on curve at those same x is clearly nonzero (a direct negative control); a 300,000-hit Monte-Carlo run of this page's own sampler shows the no-marker dark-fringe bins holding under a fifth of the neighbouring bright bins' count, while the marker-on run shows no such dip at all; and the marker-on curve is, within numerical tolerance, the no-marker curve averaged over one fringe period (removing the cross-term is exactly what a which-path fact does to the maths). Real double-slit experiments (electrons, neutrons, even large molecules) show precisely this trade — it isn't a metaphor.

Also in Waves & rhythm: Outrun your own sound →

All 7 in Waves & rhythm
  1. 12Fourier epicycles
  2. 19Standing waves & normal modes
  3. 21Lissajous figures
  4. 62Ask which slit, lose the wave — you are here
  5. 64Outrun your own sound
  6. 66Move, and your clock falls behind
  7. 70The angle past which light refuses to leave

← the cabinet · Thought Toys — a cabinet of explorable explanations. Exhibit 62.