The universe rang piercingly in its first moments. During the inflation epoch, when space expanded faster than light, energies were monstrous — trillions of times higher than in the Large Hadron Collider. This unimaginable "ringing" — oscillations of the primordial plasma born from quantum fluctuations — froze into the cosmic microwave background and the distribution of matter. If superheavy particles existed back then, they would have left a characteristic rhythmic imprint, like ripples from a stone. Seeing this pattern in the large-scale structure means probing a "collider" that operated at temperatures of 10²⁹ K. The idea of such a cosmological collider was proposed by Juan Maldacena, building on the theory of Alan Guth.
Recreating the signal in computer simulations was not easy. Standard methods decompose three-point correlations — the bispectrum — into pre-built templates, but for models with rapid oscillations they fail. The team developed a workaround: they split the wavenumber space into cells and stitched the values with cubic polynomials, achieving accuracy better than one percent. The entire work relied on standard cosmology with Hubble's law and Friedmann equations. Then, on a supercomputer, they ran an ensemble of cubes with side 3 Gpc/h (nearly 10 billion light-years), filled with 10⁹ particles of dark matter. The evolution was tracked to the present epoch (z=0). Halos (giant clouds of dark matter in which galaxies are born) were identified, and their propensity to cluster — the bias — was compared to the distribution of all matter.
For the first time, clear non-monotonic oscillations of the bias (a coefficient showing how much stronger halos are clustered compared to matter) appeared. For the baseline model, the wave is clearly visible on scales around 10⁻² h/Mpc. Increasing the halo mass by a factor of ten shifted the characteristic scale $k_c$ from 0.015 to 0.0068 h/Mpc, almost exactly following the change in Lagrangian radius. The oscillation amplitude also increased. Even more interesting is the concentration effect: when halos were divided by internal density, the phase shifted by a factor of two — for more compact objects, peaks move to noticeably smaller scales. As frequency ω increases, the signal is rapidly suppressed: if in the primordial bispectrum oscillations have constant amplitude, in the halo bias they average out and drop to a few percent at ω≈8.
The discovery provides a powerful new test of inflationary models. Unlike classical local non-Gaussianity, whose signal is monotonic and can be mimicked by instrumental effects, here we have a periodic structure with predictable frequency and phase. Collider signals could offer an even cleaner window into early-universe physics. The practical takeaway: bias parameter calibrations previously developed for local PNG work for these models as well, meaning existing analysis pipelines (DESI, Euclid, Rubin) can be easily extended.
Thus the pulse of the universe, born in the first moments, echoes in the arrangement of modern clusters. The next step is to extract these rhythmic oscillations in real catalogs of millions of galaxies. If the signal is confirmed, we will learn not only about particles that have never been produced in laboratories but also about the very nature of inflation — the epoch that opened our world.
The cosmological collider isn't science fiction: it's the Universe itself at $10^{-32}$ seconds old. The temperature then reached $10^{29}$ kelvin, and collisions were 10 trillion times more energetic than at the Large Hadron Collider.
In Carl Sagan's 'Contact,' an extraterrestrial intelligence encodes information in a pulsar — here the Universe itself has left a 'message' in the distribution of galaxies, which we are only beginning to decode.