A pair source sends one photon to Alice and one to Bob. Rotating their polarization analyzers changes joint correlations without changing either local 50:50 marginal.
• An entangled-pair source, Alice and Bob analyzers, paired detection records and a joint-statistics display. • Controls for Alice analyzer angle (°), Bob analyzer angle (°), singlet visibility and detected pairs per second. • Four tabs: Quantum bench, Probabilities & measurements, Experiments (with built-in model checks), and Learn & assess.
For the polarization singlet, E(a,b) = −V cos[2(a−b)], with P(A=+) = P(B=+) = 1/2 and P(same) = (1+E)/2. P(++) = P(−−) = (1+E)/4 and P(+−) = P(−+) = (1−E)/4. The optimized four-setting |S| = 2√2 V, against a local-hidden-variable bound of 2.
The model has ideal paired detections with a depolarizing visibility parameter. It has no detector inefficiency, loophole analysis or causal propagation simulation. The displayed S is a theoretical four-setting prediction, while the measured E uses only the current analyzer pair. Packet travel is a slowed illustration, not a signaling mechanism.
No. Bob’s marginal stays 50:50; only comparing the records reveals correlations.
No. It is the predicted value for an optimized four-setting experiment.
With aligned analyzers results are always opposite while each local stream remains random; at a 45-degree separation the correlation vanishes and outcomes are neither preferentially same nor opposite.
Correlations weaken, and the predicted optimized S falls below two at a visibility of 0.5 (the "Noisy pair source" preset).