Little red dots, or LRDs, are one of the most intriguing classes of objects JWST has revealed since it began scientific operations in 2022. They are extremely small on cosmic scales yet shine with a characteristic red color in the infrared, and researchers interpret them as black holes that are collecting matter from their surroundings and growing at a furious pace. Because their light has traveled for more than twelve billion years before reaching the telescope, each dot offers a snapshot of a supermassive black hole in the act of assembling itself during the universe’s first billion years or so of cosmic history.

The central puzzle motivating the study is one of the biggest open questions in modern astrophysics. Astronomers agree that a supermassive black hole sits at the center of essentially every large galaxy in today’s universe, with masses ranging from millions to billions of times that of the Sun. These objects are orders of magnitude more massive than ordinary stellar-remnant black holes, yet no one knows for certain how they grew so large so quickly. One possible explanation is that black holes merge with one another over time, building up mass in jumps rather than through steady feeding alone. Testing that idea requires finding black holes caught in the act of approaching one another, something that had never been convincingly achieved for LRDs until now.

The obstacle, ironically, may have been the search methods themselves. When two LRDs lie very close together on the sky, conventional analysis techniques, which rely largely on the overall brightness and color of a source, can mistake the pair for a single, complex object. Two growing black holes on the verge of a merger could therefore be hiding in plain sight within telescope data, misclassified as one source rather than counted as two. Tanaka’s team suspected that this blending problem was masking exactly the population of close pairs that theorists most wanted to find.

To get around the limitation, the researchers developed a new pixel-by-pixel color selection method. Rather than treating each LRD as a single point of aggregated light, the technique examines the color of every individual pixel in the JWST infrared images. Because the two components of a close pair imprint subtly different color signatures across the image, a pixel-level analysis can separate sources that traditional photometry would smear together. By fine-tuning conventional search procedures with this approach, the team looked specifically for objects cataloged as single LRDs that were in fact two black holes in the process of closing in on one another.

The search paid off. The team uncovered four sets of dual LRDs in the universe between 12.5 and 12.8 billion years ago, with each member of every pair lying extremely close to its companion. The separations ranged from a few thousand to a few tens of thousands of light-years, a distance far smaller than the roughly 100,000 light-year span of our own Milky Way galaxy. In other words, these growing black holes are packed within a fraction of a single galaxy’s width of each other, precisely the configuration expected if two galactic nuclei are being drawn together by gravity.

Of course, apparent closeness on the sky is not the same as true closeness in three-dimensional space. Two objects at very different distances from Earth can happen to line up along the line of sight, creating an illusion of a pair. To rule out this possibility, the researchers calculated the probability that such close alignments would appear by chance, based on the number and distribution of LRDs in the surveyed area. The odds made it unlikely that all four pairs were mere coincidences of projection. Instead, the analysis pointed to something more interesting: LRDs appear to cluster strongly on scales of several thousand light-years, meaning these rapidly growing black holes genuinely prefer the same cosmic neighborhoods.

That clustering carries a deeper implication for how galaxies and their central black holes co-evolved. Galaxy mergers are thought to be one of the mechanisms that transport large amounts of gas toward the centers of galaxies, driving matter inward and stimulating the black hole at the core to feed and brighten. The new results could indicate that galaxy mergers are linked to the rapidly growing black holes seen in the early universe, with the dual LRDs representing the smoking gun of galactic collisions in progress. If each dot in a pair indeed harbors a growing supermassive black hole, then the two black holes themselves might eventually spiral together and merge once their host galaxies fully combine.

If those black hole mergers do occur, they would not stay hidden forever. The collision of two supermassive black holes would churn the fabric of spacetime itself, producing gravitational waves, ripples that could be detected by future gravitational-wave observatories now being planned and built. The four candidate pairs identified in this study thus represent not only a snapshot of black hole growth in the infant universe but also potential future sources for an entirely different kind of astronomy, one that listens to the universe rather than looks at it. Confirming the pairs and tracking their evolution could connect the optical and infrared picture of early black hole growth with the gravitational-wave signals that mergers are expected to leave behind.

The team’s next step is to push the analysis further. The researchers plan to study the dual LRDs in more detail and to improve their pixel-by-pixel method by applying it to a larger sample of objects, turning a first handful of candidate pairs into a statistically meaningful census. By continuing the search for black hole mergers with this technique and measuring what fraction of massive black holes form close pairs, astronomers hope to quantify the role that mergers played in building the first supermassive black holes during their earliest evolutionary stages. For now, the four close pairs tucked inside JWST’s infrared images stand as a reminder that some of the universe’s most dramatic events can remain invisible until someone looks at the data one pixel at a time.

Subject of Research: Discovery of dual little red dots, candidate merging supermassive black holes, in the early universe using JWST infrared imaging

Article Title: Researchers have uncovered a growing black hole heading toward a merger in the early universe

Article References: Researchers have uncovered a growing black hole heading toward a merger in the early universe. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: James Webb Space Telescope, little red dots, supermassive black holes, black hole mergers, early universe, galaxy mergers, Kavli IPMU, infrared astronomy, gravitational waves, Publications of the Astronomical Society of Japan, black hole growth, quasars

The scale of the problem is easy to state but hard to solve. In the local Universe, virtually every large galaxy hosts a central supermassive black hole, with masses ranging from millions to billions of times that of the Sun, and these black holes are tightly correlated with the properties of their host galaxies, a relationship established through decades of work beginning with the demography of massive dark objects in galaxy centers in the late 1990s. Quasars, the luminous beacons produced when gas falls onto these black holes, had already been detected at redshifts above six, corresponding to epochs less than a billion years after the Big Bang, long before JWST launched. Those discoveries already strained theory: a black hole of a billion solar masses, shining as a quasar when the Universe was barely 700 million years old, must have grown extraordinarily fast from some initial seed, whether the remnant of a massive first-generation star or the direct gravitational collapse of a pristine gas cloud.

What JWST has done is transform the census from a handful of exceptional beacons into a genuine population study. Using its Near-Infrared Spectrograph, NIRSpec, and its mid-infrared instrument MIRI, the telescope has identified broad-line active galactic nuclei at redshifts greater than five, including objects seen just 570 million years after the Big Bang and a broad-line AGN confirmed at redshift 8.5. Deep surveys such as CEERS, JADES, UNCOVER, and RUBIES have collectively revealed that faint, actively accreting black holes are far more abundant in the early Universe than extrapolations from bright quasars had suggested. Many of these objects appear as the now-famous little red dots, compact red sources whose spectra show broad hydrogen emission lines, a signature of fast-moving gas in the vicinity of a black hole. The review emphasizes that this abundant population of faint AGN provides anchor points to constrain theoretical models of black hole formation and growth in a way that individual luminous quasars never could.

The technical basis for these discoveries lies in spectroscopy. Broad emission lines, particularly the hydrogen alpha and beta lines, are broadened by the Doppler effect as gas orbits close to the black hole at thousands of kilometers per second, and the width of the line, combined with the luminosity of the emitting region, yields an estimate of the black hole mass through so-called virial or single-epoch relations calibrated in the local Universe. JWST’s sensitivity in the near-infrared is crucial because light emitted at ultraviolet and optical wavelengths in the early Universe is redshifted into the infrared by cosmic expansion. At the same time, diagnostic line ratios of the kind first formalized in the Baldwin-Phillips-Terlevich scheme, extended to ultraviolet and auroral lines such as [O III] 4363, allow astronomers to distinguish nuclear activity from intense star formation, a nontrivial task in galaxies whose nebulae are metal-poor and ionized by young massive stars. These methods have revealed not only broad-line AGN but also a large population of obscured, narrow-line active nuclei that broad-line selection alone would miss.

Among the most consequential surprises is the apparent overmassiveness of early black holes relative to their hosts. In the local Universe, the mass of the central black hole is roughly one-thousandth of the stellar mass of its galaxy, a relation that likely reflects their coupled growth over cosmic time. JWST spectra, however, frequently yield black hole masses of ten million to a hundred million solar masses in galaxies whose stellar masses are comparable to or even smaller than the black hole itself, ratios hundreds of times higher than the local norm. Some studies argue that selection effects and the difficulty of measuring stellar masses in compact, AGN-dominated systems may soften the discrepancy, and evolutionary models in which black holes grow first and galaxies catch up later can reconcile much of the data. Nevertheless, the review highlights that the observed rapid evolution of the black-hole-to-stellar-mass relation at redshifts above three is now one of the central observational facts that any theory of black hole and galaxy coevolution must explain.

Equally puzzling are the physical properties of the little red dots themselves. Their spectra show extremely high gas densities, Balmer breaks that appear too strong to be produced by ordinary stellar populations, and a striking deficit of X-ray emission despite their enormous inferred bolometric luminosities. Chandra stacking analyses confirm that these sources are X-ray weak, prompting theoretical proposals ranging from mildly super-Eddington accretion onto slowly spinning black holes, which geometrically thickens the accretion flow and suppresses X-rays, to dense ionized cocoons or black hole envelopes that reprocess the radiation. Recent work has even identified a non-stellar Balmer break in a black-hole-dominated little red dot and rest-frame Balmer absorption features, suggesting that the observed light may emerge from an ultra-dense gaseous atmosphere surrounding the black hole rather than from stars at all. Whether these objects represent a brief, dust-enshrouded phase in the growth of the first supermassive black holes, or something more exotic such as late-stage quasi-stars, remains one of the field’s most actively debated questions.

Measuring black hole masses at these distances is fraught with uncertainty, and the review is candid about the status of the mass scale. Single-epoch virial masses rely on calibrations derived from local active galaxies and on assumptions about the geometry and kinematics of the broad-line region that may not hold at high redshift, where metallicities are low, densities are extreme, and scattering processes may reshape the line profiles. Independent checks are scarce but growing: gravitational lensing has enabled a high black-hole-to-host mass ratio measurement in one early AGN, spatially resolved integral-field spectroscopy is beginning to test the single-epoch method, and dynamical measurements of gas kinematics in quasar hosts at redshifts up to about seven offer an alternative route. The stakes are high, because if the masses are systematically overestimated, the overmassiveness problem diminishes; if they are robust, then the seeds of the first black holes must have been heavy from the start, favoring direct-collapse scenarios over the remnants of ordinary massive stars.

The review also surveys the environments and feedback of these early engines. Integral-field observations with NIRSpec have revealed offset AGN, dual and even triple active nuclei, and merging quasar hosts, indicating that black hole growth in the first billion years is intimately connected with the hierarchical assembly of galaxies in overdense regions. Fast AGN-driven outflows have been directly detected in little red dot host galaxies, and kiloparsec-scale shells of gas around some high-redshift quasars point to early episodes of feedback that may regulate both star formation and further black hole growth. At the same time, the scarcity of cold dust and molecular gas in many little red dots, constrained by deep ALMA observations, hints that these systems differ fundamentally from the dusty, gas-rich quasar hosts seen at slightly later epochs. Multiwavelength follow-up, from radio detections of jet candidates to mid-infrared imaging with MIRI, is gradually assembling a coherent picture of how these objects radiate across the spectrum.

Looking forward, the review identifies clear observational targets. Larger spectroscopic samples will sharpen the black hole mass function at high redshift and test whether the abundance of faint AGN is sufficient to seed the quasar population seen later. Deeper X-ray observations, ultimately with the proposed NewAthena mission, will test the X-ray weakness of the little red dots and search for heavily obscured nuclei. Pulsar timing arrays have already detected a gravitational-wave background consistent with a cosmic population of supermassive black hole binaries, and the future Laser Interferometer Space Antenna, LISA, will directly detect the mergers of massive black holes at high redshift, providing an entirely independent probe of early black hole growth. For now, JWST continues to deliver anchor points at ever greater distances, and each new spectrum of a faint red dot or a broad-line nucleus adds a constraint on one of astronomy’s oldest questions: how did the Universe build its monsters so quickly, so soon after the Big Bang?

Subject of Research: Massive black holes in the first billion years of the Universe observed with JWST

Article Title: Massive black holes in the first billion years with JWST

Article References: Übler, H. (2026). Massive black holes in the first billion years with JWST. Nature Astronomy, 10(9), 1273-1282. https://doi.org/10.1038/s41550-026-02966-8

Image Credits: AI Generated

DOI: 10.1038/s41550-026-02966-8

Keywords: supermassive black holes, JWST, early Universe, active galactic nuclei, little red dots, high redshift, quasars, black hole seeding, galaxy coevolution, NIRSpec, cosmic dawn, black hole mass