Beijing, China — An international research team led by Prof. Fangzhou Jiang at the Kavli Institute for Astronomy and Astrophysics (KIAA) at Peking University has unveiled a new theory explaining one of the biggest astrophysical mysteries revealed by the James Webb Space Telescope, the origin of the enigmatic “Little Red Dots”. These tiny but extraordinarily bright objects at cosmic dawn appear to host supermassive black holes far larger than their surrounding galaxies should allow. Until now, their existence has posed a severe challenge to standard models of galaxy and black-hole formation and coevolution.
In a paper just published by The Astrophysical Journal Letters, the KIAA team shows that these objects can arise naturally from the collapse of self-interacting dark-matter halos, offering the first cosmological, statistical explanation of the Little Red Dot population and establishing one of the leading formation scenarios for these objects from the perspective of cosmic structure formation.
Black holes born from dark matter
Figure: In a self-interacting dark-matter universe, a halo that formed early and was exceptionally dense at the center can undergo core collapse before reionization, giving rise to a supermassive black hole seeded directly from dark matter.
The research reveals a radical possibility: the first massive black holes may have formed directly from dark matter itself. Unlike ordinary cold dark matter, self-interacting dark matter (SIDM) can scatter and conduct heat. This process can eventually drive a runaway gravothermal collapse, causing the dark-matter core to shrink catastrophically and form a massive black hole. In rare, dense halos forming in the early universe, this process can proceed rapidly before any stars exist.
The team shows that this process occurs preferentially in rare, high-concentration halos at redshifts above ~8, exactly when Little Red Dots are observed to appear. The model considers existing results from core-collapse studies that about 0.1 to 1 percent of a dark-matter halo’s mass collapses into a black hole, naturally producing black holes of ∼106 solar masses in halos of ~109 solar masses at the moment of black-hole formation. This is the same black-hole mass scale inferred for Little Red Dots.
Matching the observed Little Red Dot population
Figure: Black-hole mass functions of the Little Red Dots at redshift of 5 predicted by the model of the core-collapse of self-interacting-dark-matter (SIDM) halos and black-hole growth. The fiducial case (red) adopts a velocity-dependent SIDM cross section with σ0m = 30cm−1g and ω = 80kms−1 -- values consistent with independent constraint from nearby galaxy kinematics. Observationally inferred BHMFs for LRDs at z = 4.5–6.5 are shown as diamonds with error bars or arrows indicating upper limits. Two alternative cross-section models (blue and green) illustrate the sensitivity of the predicted BHMF to SIDM parameters: increasing the low-velocity cross section σ0m boosts the low-mass end, while increasing ω enhances the massive end.
Using cosmological Monte-Carlo merger trees combined with SIDM physics, the researchers constructed a full population model of black holes at cosmic dawn. They show that with dark-matter interaction strengths consistent with independent constraints from nearby galaxies, the model reproduces the observed abundance and mass distribution of Little Red Dots at redshifts 5–8, when the Universe was less than one tenth of its current age. This makes SIDM-driven dark collapse one of the first frameworks that explains both why Little Red Dots host over-massive black holes, and why they lack normal galaxies around them. Because the black hole forms before stars, the galaxy arrives late.
A new way to test dark matter
Perhaps most importantly, the work shows that the statistics of early black holes now probe the particle physics of dark matter. The team demonstrates that changing the dark-matter cross section shifts the predicted black-hole mass function in distinctive ways—meaning that future JWST and Roman Space Telescope surveys can constrain dark-matter properties using the demographics of cosmic-dawn black holes. This opens a new observational window on dark matter that is independent of traditional galaxy-rotation and gravitational-lensing tests.
A paradigm shift for cosmic origins
“In this new paradigm, the Little Red Dots are not merely possible -- they are inevitable,” the lead author of this study, Prof. Fangzhou Jiang, concludes, “they may be the visible tips of the earliest density peaks formed by dark matter.”
“The dark-matter scattering cross section required to explain these distant black holes turns out to be consistent with constraints from nearby galaxy rotation curves”, says Zixiang Jia, a first-year graduate student at Peking University, who is leading a companion study that uses local galaxies to constrain the strength of dark-matter self-interactions.
“Our standard theory of structure formation predicts a distribution of dark-matter halo structures in the early Universe,” says Dr. Haonan Zheng, a KIAA and Boya Fellow at KIAA. “At sufficiently early times, this distribution is remarkably insensitive to epoch or mass scale, always producing a rare tail of unusually dense halos—the ones we believe host the Little Red Dots.”
By linking dark-matter microphysics, halo assembly, and black-hole formation into a single predictive framework, the KIAA team has provided a cosmological, statistical origin theory for the Little Red Dots.
Publication
Fangzhou Jiang, Zixiang Jia, Haonan Zheng, Luis C. Ho et al. (KIAA Peking University)
Formation of the Little Red Dots from the Core Collapse of Self-interacting Dark Matter Halos
The Astrophysical Journal Letters, 996, L19 (2026)
