An international group of quantum physics researchers reckons it's come up with an experimental validation of the reality of the wave-function.

It's one of quantum mechanics' most elusive concepts, ever since Erwin Schrödinger crafted the famous “Schrödinger's cat” thought experiment.

In the current experiment, led by Martin Ringbauer of the University of Queensland, photons seem to genuinely exist in multiple states until they're measured.

There's still wriggle-room in their result, reported in *Nature Physics* here (Arxiv pre-print in full here), since the experiment does not yet rule out all models where the wavefunction is only partly real.

The experimental setup (described in detail below) allowed Ringbauer and coworkers to prepare different states of photons, and measure those states to estimate “how large the overlap of the probability distributions can be”.

In short – because this is going to be a long read – an experiment at Andrew White’s Quantum Technology Lab at the University of Queensland suggests the cat might genuinely obey the wave-function. Until you measure its state, it's dead-and-alive in superposition rather than “there's an 80 per cent chance he's dead, let's look”.

The long version starts with the question “How can you test this?”

### Dead and alive

The paradox offered by “Schrödinger's cat” - put a cat in a box, put it in the presence of a lethal quantum event like radioactive decay triggering a gas release, come back and check whether the cat's dead or alive - is that the wave function described in quantum mechanics suggests the cat exists in a superposition of dead/alive states simultaneously, only resolved when you open the box and make your observation.

This put an emphasis on the observer that some great minds of the 20th century (Einstein among them) found objectionable – hence the thought experiment.

Ringbauer said the group's current work seeks to rule out one of three overarching approaches to interpreting quantum mechanics:

- One interpretation – not addressed in this experiment – says quantum systems “have no objective properties”. Their properties depend on the observer;
- One “realist” interpretation is that the wave function is an objective property;
- Another “realist” view contends that the wave function is a statistical description of what is knowable about a quantum system– the “partial knowledge” view.

### Build it!

An approach to testing the reality of the wave function was proposed by Dr Eric Cavalcanti (a contributor to this paper) and collaborators in 2014, and was refined by another participant in this research, Dr Cyril Branciard.

Ringbauer's experiment was designed to work out whether either of the “realist” views could be favoured (and the other ruled out).

How? By measuring whether the combinations of the photons' quantum states (polarisation and path) showed a statistical distribution, or whether they appeared sufficiently discrete to support the idea that the wave function is real.

Ringbauer explained that the experiment used four dimensions – vertical vs horizontal polarisation provided two; and which path the photons followed provided the other two.

“You cannot perfectly distinguish two quantum states unless they are orthogonal. Vertical vs horizontal polarisation are orthogonal; but diagonal polarisation isn't orthogonal to vertical”, he explained.

However, using only polarisation is not enough to test the "knowledge" interpretation.

Adding a path choice – did the quanta take path A or B, or a superposition of both? – creates a four-dimensional model, which allows discrepancies between the interpretations to be observed.

Martin Ringbauer's experimental setup: vertical and horizontal polarisation and different paths

provided four dimensions of measurement. Photons travelled from the source (SPDC) through the experiment,

and measurement suggests they existed in a true superposition until measured.

This experimental setup allowed Ringbauer's team to prepare different states of photons, and measure those states to estimate “how large the overlap of the probability distributions can be".

The group was looking for “the discrepancy between what we see, and what theory says is needed” [to support either the knowledge (probability) model, or suggest the waveform is real].

If the wavefunction is merely probabilistic, then the photon was always in a definite polarisation state and path, from the time it was prepared to the time it was measured. If the wavefunction is real, then the photon was really in many polarization states and locations at once.

The outcome of those measurements is best put by *New Scientist*: “not enough information could be gained about the polarisation of the photons to imply they were in particular states before measurement”.

### So now what?

Imagine for a moment that the wave function *is* real. “So what?” is a legitimate question, asked by impatient readers before *The Register*'s attempted explanation, and asked by patient readers now.

If the wave function is real, it's more than merely a vindication of the idea that the universe is weird. It has implications for the future of how we in the macro world interact with the quantum one.

Don't ever discount the practicality of quantum mechanics: you're reading this article thanks to bits transmitted by lasers, and lasers were merely an early practical application of the strange world of quanta.

For example, Ringbauer says, it helps us to understand our limits when simulating quantum systems. In the knowledge system, “where there are physical properties of a system, and the wave function is just a statistical description, that is a classical model, at heart. So you would expect to be able to simulate it.”

If the wave function is an objective property of quantum systems, it's much harder to simulate on classical computers: “This result puts some bounds on how well, or how resource-efficiently, you can simulate a quantum system.”

Simulation is necessary if we want to understand how quantum effects (like molecules absorbing photons) affect macro systems (like photosynthesis).

Regrettably, a real wave function isn't the answer simulation researchers would want, because it means “we can't easily simulate quantum systems on a classical computer.”

*The Register* notes that we've discussed how one important application of quantum computers might be to simulate quantum systems. ®