Every evening, in the city archive, the attendant places new folders on the shelves. If too many arrive during the day, the shelves sag, and the next morning the archive responds to requests more slowly. A black hole's memory works in much the same way: it doesn't just store information, it physically feels its weight. And when two black holes collide, their informational cargo is not a silent witness; it changes the very dance of the merger.

For a long time, it was thought that a black hole is defined by three numbers: mass, charge, and angular momentum. Information about what fell into it seemed to disappear behind the horizon. But in recent years, physicists have realized that information is not passive. It can accumulate in special 'gapless' modes that store data almost without energy cost, like an archive where new folders find empty shelves on their own. This mechanism is called auxiliary gaplessness. The new work shows that when a black hole is perturbed, for instance during a merger, this memory activates quickly and changes the classical dynamics.

The key parameter is the 'memory burden', denoted by μ. It is the fraction of occupied memory: how many 'shelves' of the archive are already filled. The smaller μ, the stronger the response. For astrophysical black holes born from the collapse of ordinary matter, μ is extremely small, less than 10⁻⁹. This means the memory effect is significant for them. For primordial black holes from the collapse of a Hubble volume, μ can drop to the minimum value 1/√S, where S is the Bekenstein–Hawking entropy. During a merger, the perturbation spectrum shifts to the infrared, and the amplitude is suppressed. This leaves a characteristic trace in gravitational waves.

The work introduces a new macroscopic quantum characteristic of a black hole: the memory burden parameter, which complements mass, charge, and angular momentum. Unlike previous approaches, where the memory burden set in gradually over macroscopic time, here the effect acts instantaneously upon an external classical perturbation. This makes it observable for black holes of arbitrary mass, including supermassive ones. Moreover, the effect is universal: it applies to any system with high information storage capacity, from black holes to laboratory condensates.

The practical payoff is in black hole spectroscopy and the analysis of data from gravitational wave observatories LIGO and Virgo. If the memory burden affects the signal shape, future detectors could distinguish how much information each merging black hole carried. This opens a new channel for testing quantum gravity, not only in space but also on the tabletop: a laboratory test is proposed using a ring of cold attracting bosons. There, one can reproduce the same mechanics and see how information load changes the system's behavior.

An archive that changes the gait of its keeper is not a metaphor but an accurate description. A black hole with full memory and a black hole with empty memory dance differently. And soon we will learn to hear this dance.