The double-slit experiment
Send electrons through two tiny openings, one at a time. Why do scattered dots turn into stripes?
Top view · waves are a sketch, not a particle’s track.
Start with one dot. Then let the picture build.
Both slits are open. No device records which one the electron passes through.
Changing the setup clears the screen. Speed only fast-forwards the trials. This is a simulation.
Experiment paused. 0 electrons detected.Every arrival is a dot.
An electron hits the screen in one place. The next arrives only after the previous one has been detected, so they cannot be bumping into each other.
Yet the dots pile up in some places and avoid others. The stripes emerge from many individual arrivals. This happens in real experiments, too.
The possibilities behave like waves.
To predict where an electron might land, quantum physics uses a wavefunction. With both slits open and no record of the path, contributions from both openings combine.
Like ripples, they can reinforce or cancel each other. This changes the chance of landing at a spot. It doesn’t split the final dot into two.
Through slit A
Through slit B
Together
More likely to land here
Chance relative to a bright band
A wave analogy for probability amplitudes, not electrons wobbling up and down.
Each electron has the same striped map of chances. One arrival looks random. Thousands reveal the map. No other electrons are needed to create interference.
What if we check which slit?
Try “Measure the path” above. A device records which opening each electron uses. This removes the interference between the two paths. A broad spread replaces the fine stripes.
“Observation” means a physical interaction that leaves a path record. No human needs to read it. The final screen records where an electron landed, not which slit it used.
The physics behind the explanationA small note about this simulation
The dots are random samples from an ideal far-field probability distribution for equally illuminated slits. This is an illustration, not measured data or a full simulation of an electron’s motion. The wave sketch is schematic; sizes and flight times are not to scale.
Two unmeasured paths: P(x) ∝ sinc²(αx) cos²(βx). One open slit, or two fully distinguishable paths: P(x) ∝ sinc²(αx). Each mode is normalized over the visible screen. Here the two single-slit envelopes overlap, so recording the path gives one broad envelope, not two separate piles. A single slit can still diffract.
The graph compares the histogram of screen positions with the predicted curve; each is scaled to its own peak. Vertical dot positions are randomized for legibility. Each run stops at 6,000 detections. Changing the setup starts a new run; existing hits never move.
The wave slider shows how equal-amplitude contributions reinforce or cancel at a selected position. Its bar shows the chance relative to a bright-band maximum, not the probability of hitting an exact point.