Quantum mechanics, the theory governing the microscopic world, is famously counterintuitive. A particle can exist in a superposition of multiple states, such as different positions, until a measurement is performed. At that point, the wavefunction describing that particle appears to 'collapse' to a single outcome. This puzzle lies at the heart of the measurement problem, famously illustrated by Schrödinger's cat, suspended between life and death until observed. The XENONnT detector, which was designed to be sensitive to rare physics events, has tightened constraints on one family of possible solutions to the measurement problem, known as 'collapse theories.' The work, which was partially funded by FQxI, was reported in Physical Review Letters in March 2026.
"This is exactly where foundations meet experiment," says Catalina Curceanu, a quantum physicist at INFN Frascati, in Italy. "Our study, partially funded by FQxI, allowed us to push sensitivity to new levels and turn the measurement problem from a philosophical question into a testable feature of nature and many proposed solutions are now under real pressure."
"Our study, partially funded by FQxI, allowed us to push sensitivity to new levels and turn the measurement problem from a philosophical question into a testable feature of nature and many proposed solutions are now under real pressure," says Catalina Curceanu.
What actually causes this collapse? And is it a real physical process?
Over the past century, physicists have proposed a wide range of answers. Some interpretations suggest that all possible outcomes occur in parallel worlds (the many-worlds interpretation). Others treat the wavefunction as a tool for predicting observations, rather than a description of reality itself. Still others assume that particles have well-defined trajectories guided by, crudely-speaking, a quantum wave—at the price of introducing nonlocality (that is, allowing information to appear to travel at faster-than-light speeds, contradicting Einstein's relativity). Most of these ideas, however, are difficult, if not impossible, to test experimentally.
Triggering Collapse
A notable exception is a class of theories known as collapse models. These propose that wavefunction collapse is a real, physical process triggered by interactions with a continuous noise field, or even by gravity. In such models, quantum superpositions are continuously and spontaneously destroyed, explaining why the macroscopic objects that we see around us do not display quantum behavior. Crucially, collapse models make testable predictions. The collapse process should induce a tiny but measurable effect: the emission of faint electromagnetic radiation, known as 'spontaneous radiation,' produced as charged particles are randomly accelerated.
"They move the measurement problem from interpretation into the realm of experimental physics," says Kristian Piscicchia.
"This is what makes collapse models so compelling," says Kristian Piscicchia, a physicist at the Enrico Fermi Research Center (CREF) and INFN Frascati National Laboratories, in Italy. "They move the measurement problem from interpretation into the realm of experimental physics."
In previous work within the VIP collaboration, Piscicchia, Curceanu (a spokesperson for VIP) and Simone Manti at INFN-LNF in Italy, developed detailed predictions for the radiation expected from the most widely-studied collapse models. These include the Continuous Spontaneous Localization (CSL) model and the Diósi-Penrose (DP) model, which link wavefunction collapse to gravitational effects.
Underground Collaboration
Building on these predictions, the team has now collaborated with the XENONnT experiment to search for this elusive signal. Located deep underground at the Gran Sasso National Laboratory (INFN-LNGS) in Italy, XENONnT uses a large volume of ultra-pure liquid xenon to detect extremely rare interactions in an environment shielded from cosmic radiation, usually to search for signs of dark matter.
If collapse-induced radiation were present, it would interact with xenon atoms and produce detectable light signals. However, the analysis revealed no such excess. This absence of a signal places the most stringent constraints to date on key collapse models. The limits on the CSL model improve upon previous bounds by two orders of magnitude, while constraints on the Diósi-Penrose model are tightened by about a factor of five.
"XENONnT demonstrates that detectors designed for dark matter searches can also probe fundamental questions in quantum physics, thanks to its extremely low background levels," says Jingqiang Ye.
These results significantly narrow the range of viable parameters for collapse models, guiding future theoretical developments while motivating the design of even more sensitive experiments. "XENONnT demonstrates that detectors designed for dark matter searches can also probe fundamental questions in quantum physics, thanks to its extremely low background levels," says Jingqiang Ye, a physicist at the Chinese University of Hong Kong, Shenzhen, and a member of the XENON collaboration. "Our next-generation detectors, with larger target masses and even lower backgrounds, are capable of extending these tests even further."
Bridging Quantum and Gravity
Understanding whether wavefunction collapse has a physical origin would have profound implications. In particular, some collapse models offer a possible bridge between quantum mechanics and gravity, two fundamental frameworks that remain notoriously difficult to reconcile. "I believe that a complete theory unifying quantum mechanics and gravity will ultimately explain the measurement process as a natural physical phenomenon," adds Piscicchia.
"I believe that a complete theory unifying quantum mechanics and gravity will ultimately explain the measurement process as a natural physical phenomenon," says Kristian Piscicchia.
For now, the search continues. With increasingly sensitive experiments, questions once considered purely philosophical are becoming accessible to empirical investigation, opening a new window on one of the deepest mysteries in physics.
This work was partially supported through FQxI's Consciousness in the Physical World program
