Computational framework for exploring retrocausality in quantum mechanics.
Can quantum mechanics let us send information backward in time? No. But the boundary between what retrocausality means and what it cannot do is fascinating -- and this project lets you explore it computationally.
The delayed-choice quantum eraser (Kim et al., 1999) appears to show that a future measurement choice can retroactively affect a past result. Does this mean we can build a time machine?
This framework demonstrates three key results:
-
The quantum eraser does NOT enable backward signaling. The total D0 distribution is always featureless -- interference only appears in post-selected subsets via coincidence counting. (Phase 1)
-
Retrocausal effects are real in post-selected ensembles. The Two-State Vector Formalism (TSVF) produces anomalous weak values that lie outside the eigenvalue spectrum -- experimentally measurable "retrocausal" signatures. (Phase 2)
-
Retrocausal models can be local AND violate Bell's inequality. The Price-Wharton zigzag model reproduces all quantum predictions while being strictly local, at the cost of future-input dependence. But no-signaling is always respected. (Phase 3)
- Quantum Eraser Simulation -- Full Kim et al. (1999) setup with SPDC, coincidence counting, and no-signaling verification
- TSVF Engine -- First open-source Two-State Vector Formalism simulator with weak value calculator, ABL rule, and the three-box paradox
- Retrocausal Toy Models -- Executable implementations of the Price-Wharton zigzag model and Wharton-Argaman boundary-value approach
- Bell Test Comparator -- Side-by-side comparison of QM, retrocausal, and classical models
- No-Signaling Audit -- Rigorous verification that all models respect the no-communication theorem
- Advanced Experiments -- GHZ/W state analysis, decoherence effects, Tlalpan phase-transition simulation, Castagnoli speedup-retrocausality link
pip install -r requirements.txt
python main.pysrc/
core/ # Quantum states, operators, density matrices
eraser/ # Phase 1: Quantum eraser + no-signaling
tsvf/ # Phase 2: Two-State Vector Formalism engine
retrocausal/ # Phase 3: Retrocausal hidden variable models
advanced/ # Phase 4: GHZ, decoherence, phase transitions
analysis/ # Statistical tools, Bell inequalities
visualization/ # Plotting functions
tests/ # Unit tests
notebooks/ # Interactive Jupyter notebooks
| Paper | Year | Relevance |
|---|---|---|
| Kim, Yu, Kulik, Shih, Scully -- Delayed-Choice Quantum Eraser | 1999 | Phase 1: the experiment we simulate |
| Aharonov, Bergmann, Lebowitz -- Time Symmetry in Quantum Measurement | 1964 | Phase 2: foundation of TSVF |
| Aharonov, Albert, Vaidman -- Weak Values | 1988 | Phase 2: anomalous weak values |
| Wharton & Argaman -- Bell's Theorem and Local Reformulations | 2020 | Phase 3: retrocausal Bell models |
| Leifer & Pusey -- Time Symmetry Without Retrocausality? | 2017 | Theoretical motivation |
| Castagnoli -- Quantum Speedup and Retrocausality | 2025 | Phase 4: speedup connection |
- numpy, scipy, matplotlib, plotly, pandas
- pytest (for testing)
MIT