Toward an Experimental Device-Independent Verification of Indefinite Causal Order

This paper presents the first experimental implementation of a device-independent protocol to verify indefinite causal order by violating a Bell-like inequality with a value of $1.8328 \pm 0.0045$, which exceeds the definite causal order bound by 18 standard deviations, despite the presence of experimental loopholes.

The Gist

Imagine you are watching a movie. In our everyday world, the story follows a strict timeline: Scene A happens, then Scene B happens, then Scene C. The past causes the future, but the future never changes the past. This is what physicists call a "definite causal order."

However, quantum physics suggests that at the tiniest scales, the rules of time can get fuzzy. It's as if a movie could play Scene A and Scene B at the same time, or in a superposition where the order is both "A then B" AND "B then A" simultaneously. This is called Indefinite Causal Order (ICO).

For years, scientists have built machines (called "quantum switches") to create this weird time-soup. But there was a catch: to prove the switch was actually doing something magical, they had to trust the machines themselves. It was like asking a magician to prove they aren't using hidden wires, but then having to trust the magician's own explanation of how the wires work.

This paper describes a major step toward proving that this "time soup" is real, without having to trust the machines.

The Big Idea: A Game of "Blind Trust"

To prove the order is truly indefinite, the researchers used a clever trick borrowed from a famous test called a "Bell Test." Think of it like a high-stakes game show with four players: Alice 1, Alice 2, Bob, and Charlie.

1. The Setup: Alice 1 and Alice 2 are inside a "black box" (the quantum switch). They perform actions on a photon. Bob and Charlie are outside.
2. The Challenge: The goal is to see if Alice 1 and Alice 2 can influence each other in a way that defies a fixed timeline, while Bob and Charlie play a separate game to prove they are "entangled" (connected in a spooky, quantum way).
3. The Rule: If the universe follows normal rules (where time has a fixed order), there is a hard limit to how well the team can win this game. The paper calls this limit the "Definite Causal Order Bound."
4. The Result: The team played the game and scored 1.83. The maximum score allowed by normal, fixed-time rules is 1.75.

Because they broke the "fixed time" limit, they proved that the events inside the switch did not happen in a single, definite order. The "black box" was doing something that no classical machine could do, even if we don't know exactly how the machine works inside.

The Experiment: A Photon's Journey

Here is how they did it, using a metaphor of a traveling messenger:

• The Messenger: They used a single particle of light (a photon).
• The Control Switch: Imagine the photon has a "brain" (a control qubit) that decides which path it takes.
  • If the brain is in state "0," the messenger visits Alice 1 first, then Alice 2.
  • If the brain is in state "1," the messenger visits Alice 2 first, then Alice 1.
• The Superposition: The researchers put the messenger's brain into a state where it is both 0 and 1 at the same time. This means the messenger visits Alice 1 then 2, AND Alice 2 then 1, simultaneously.
• The Entanglement: They also linked the messenger's brain to a second messenger (Bob) who is far away. This link is so strong that what happens to one instantly affects the other.
• The Test: Bob and Charlie play a game to check if their link is strong enough to break the rules of normal physics. At the same time, they check if Alice 1 and Alice 2 can "signal" to each other in a way that only works if the order is mixed up.

The Scoreboard

The researchers measured the results and found:

• The Limit for Normal Time: 1.75
• Their Score: 1.8328
• The Gap: Their score was 18 standard deviations higher than the limit. In the world of physics, this is a massive, undeniable victory. It's not a fluke; it's a clear signal that the causal order was indefinite.

The "Loophole" Caveat

The paper is very honest about its limitations. While they proved the logic of the test works without trusting the machine's internal theory, their physical setup still had some "loopholes" (shortcuts in the rules) that a perfect experiment would need to close.

• The Distance Problem: In a perfect "device-independent" test, Bob and Charlie should be so far apart that no signal could travel between them fast enough to cheat. In this experiment, they were on the same optical table, less than a meter apart.
• The Timing Problem: The experiment relied on the first photon being detected to trigger the second part of the machine. In a perfect test, the choices of what to measure should be made randomly and independently, without one event triggering the other.

The authors admit these loopholes exist. However, they argue this is a crucial "proof of principle." They showed that the math works and the violation is possible. It's like building a prototype car that drives 100mph but still has a handbrake on; it proves the engine works, even if the car isn't ready for the highway yet.

Why This Matters

This experiment is a giant step toward confirming that indefinite causal order is a real physical phenomenon, not just a mathematical trick or a simulation.

• It moves us from "We think this is happening because our machine says so" to "Nature itself is behaving in a way that defies a fixed timeline."
• It opens the door to using this "time soup" for future technologies, such as better communication or computing, though the paper focuses strictly on proving the phenomenon exists rather than building those devices yet.

In short: The researchers successfully played a game of quantum logic against the universe, and the universe admitted that, in this specific setup, time doesn't always flow in a straight line.

Technical Summary

1. Problem Statement

In classical physics, events follow a definite causal order (past influences future). Quantum theory, however, allows for Indefinite Causal Order (ICO), where the order of operations exists in a superposition. While the Quantum Switch is the primary experimental realization of ICO, previous verifications have been device-dependent or semi-device-independent. These methods rely on assumptions about the internal workings of the devices (e.g., trusting the state preparation or measurement settings), leaving them open to "loopholes" where observed correlations could theoretically be explained by classical causal models with hidden variables.

The core challenge is to verify ICO Device-Independently (DI). A DI verification would prove that nature allows correlations violating causal inequalities without assuming any specific quantum theory or trusting the experimental hardware. However, the standard Quantum Switch does not violate standard causal inequalities. Therefore, a new theoretical framework was needed to test the Quantum Switch in a DI manner.

2. Methodology

Theoretical Framework: The VBC Inequality

The authors implement a protocol based on a recently proposed inequality by van der Lugt, Barrett, and Chiribella (VBC). This inequality involves four parties:

• Alice 1 ($A_1$) and Alice 2 ($A_2$): Operate within the Quantum Switch.
• Charlie ($C$): Acts in the future light cone of the Alices.
• Bob ($B$): Space-like separated from the others.

The VBC inequality combines two distinct "games":

1. The Causal Order Game: Bob chooses a setting ($y=0$) to measure the control qubit. If the causal order is definite, Bob's outcome ($b$) should perfectly correlate with the hidden variable $\lambda$ determining the order ($A_1 \to A_2$ or $A_2 \to A_1$). The Alices attempt to signal to each other based on this order.
2. The CHSH Game: Bob and Charlie play a standard Bell test (CHSH) to check for entanglement.

The Inequality:
$$p(b=0, a_2=x_1 | y=0) + p(b=1, a_1=x_2 | y=0) + p(b \oplus c = yz | x_1=x_2=0) \leq \frac{7}{4}$$

• Definite Causal Order (DCO) Bound: $\leq 1.75$.
• Quantum Maximum (ICO): $\approx 1.8536$.

To violate this, the system must simultaneously win the Causal Order Game (requiring ICO) and the CHSH Game (requiring entanglement), which is impossible if the causal order is fixed.

Experimental Implementation

The team used a photonic Quantum Switch with the following architecture:

• Source: A Type-0 Spontaneous Parametric Down-Conversion (SPDC) source generates polarization-entangled photon pairs ($|\Phi^+\rangle$) at 1550 nm.
• Encoding:
  • Control Qubit: Encoded in the time-bin degree of freedom (early vs. late arrival) of Photon 2.
  • Target Qubit: Encoded in the polarization of Photon 2.
  • Ancilla: Photon 1 (sent to Bob) remains in polarization encoding.
• Entanglement Transfer: An imbalanced Mach-Zehnder-like interferometer converts Photon 2's polarization entanglement with Photon 1 into time-bin entanglement.
• The Switch:
  • Ultra-Fast Optical Routers (UFORs): Synchronized by the detection of Photon 1, these route the early time-bin to $A_1 \to A_2$ and the late time-bin to $A_2 \to A_1$.
  • Operations: Inside the switch, $A_1$ and $A_2$ perform projective measurements in the computational basis and re-prepare the target state based on random settings ($x_1, x_2$).
• Measurement:
  • Bob: Measures Photon 1 in the Z or X basis.
  • Charlie: Interferes the time bins of Photon 2 (using a reverse interferometer) and performs polarization tomography to measure the control and target qubits.

3. Key Contributions

• First DI Protocol for Quantum Switch: This is the first experimental implementation of a device-independent protocol specifically designed to verify ICO in a Quantum Switch.
• Loophole Identification: While the experiment does not close all standard Bell loopholes (e.g., locality, detection efficiency), it explicitly identifies and discusses new loopholes specific to ICO, such as the "time-delocalized events" and the "closed lab" assumption, which are critical for future loophole-free ICO experiments.
• Noise Characterization: The study provides a detailed analysis of how different noise types (depolarizing vs. dephasing) affect the violation of the VBC inequality, distinguishing between the loss of entanglement and the loss of causal correlations.

4. Results

• Violation of the Inequality: The experiment measured a VBC value of $1.8328 \pm 0.0045$.
• Statistical Significance: This result exceeds the Definite Causal Order bound of $1.75$ by 18 standard deviations.
• Fidelity: The measured value is close to the theoretical quantum maximum of $\approx 1.8536$. The deviation is attributed to:
  • Input state purity ($0.98036 \pm 0.00034$).
  • Quantum switch process fidelity ($F_{switch} = 0.9816 \pm 0.0069$).
• Noise Analysis:
  • Depolarizing Noise: Reduced all terms of the inequality, destroying both entanglement and causal correlations.
  • Dephasing Noise: Reduced the CHSH term (entanglement) but left the causal order terms ($p_1, p_2$) largely intact, demonstrating that classical correlations can persist even when quantum coherence is lost.

5. Significance

• Foundational Impact: This work represents a major step toward a loophole-free verification of ICO. It moves the field from "device-dependent" demonstrations (which rely on trusting the apparatus) to a framework where the causal structure itself is the only assumption being tested.
• Resolving the Debate: The results support the view that photonic Quantum Switches realize genuine ICO (via superposition of spacetime events) rather than just simulating it, countering arguments that ICO requires gravitational superposition.
• Resource for Quantum Technologies: By confirming ICO as a distinct quantum resource independent of entanglement, this verification strengthens the theoretical basis for ICO-based applications, such as:
  • Enhanced channel discrimination.
  • Reduced communication complexity.
  • Noise mitigation and thermodynamic advantages.
• Future Directions: The paper outlines the specific requirements (e.g., space-like separation, high-efficiency detection, and local readout of events) necessary to close the remaining loopholes and achieve a fully loophole-free DI verification of indefinite causal order.