How Black Hole Hunters Searches for Hidden Black Holes

Black Hole Hunters asks volunteers to help search the brightness records of stars for a subtle clue: a dark object passing in front of a companion star can bend and magnify its light. Researchers use data from the Transiting Exoplanet Survey Satellite (TESS), while volunteer classifications and algorithmic tools help sift through the observations. A promising signal needs further analysis; spotting a pattern is not the same as confirming a black hole. 1718

What Black Hole Hunters is looking for

The project focuses on gravitational self-lensing in binary star systems. When a binary is viewed from a suitable, nearly edge-on angle, light from the visible star can be bent and magnified by its dark companion as that object passes in front. Unlike a one-off microlensing event, the brightening can repeat with the system’s orbit. 18

That repeating change in brightness is the signal researchers hope to find in stellar light curves. The project uses simulated examples to help volunteers learn what patterns to look for. Its FAQ cautions that these events are rare and that no conclusive black-hole self-lensing example had been found at the time described there.

How volunteers and algorithms work together

Black Hole Hunters runs on Zooniverse, where people can classify light curves after web-based training. The project draws on TESS observations; its researchers have also described plans to examine data from other surveys. 1718

Volunteer classifications are one part of the search, not a substitute for analysis by researchers. In an April 2026 update, researcher Adam McMaster said that volunteers had submitted 7.6 million classifications. He described combining those classifications with more sophisticated simulations to develop algorithms that pre-filter light curves and help narrow the search to promising candidates.

The project launched on Zooniverse on October 26, 2021. Its listed team includes Adam McMaster, Hugh Dickinson and Matthew Middleton, with researchers from the University of Southampton, the Open University and the Flatiron Institute among its collaborators. Southampton later described a 2024 grant supporting work by Middleton and McMaster with Dickinson and the recruitment of volunteers. 17

What the separate Higgs Hunters project can—and cannot—show

Higgs Hunters tackled a different problem. Volunteers examined images from the ATLAS experiment at the Large Hadron Collider, looking for off-centre vertices: particle tracks that appear to begin away from the main collision point. Such features can be clues to particles that travel before decaying, but an unusual mark in an image is not, by itself, evidence of a new particle. 10

An analysis of the project examined how non-expert volunteers identified long-lived-particle decays and other unusual features. It reported that volunteers’ efficiency and false-positive rate were comparable to those of the ATLAS algorithms on the material studied. The paper also described volunteers flagging events with unexpected features for further attention. 3

Oxford reports that Higgs Hunters reached 42,280 participants from more than 170 countries and involved UK school students in a subsequent analysis stage. Those participation figures describe that project; they should not be confused with Black Hole Hunters’ classifications or treated as evidence that volunteers outperform algorithms in every scientific task. 4

The lesson for citizen science

Together, the projects illustrate a practical role for public participation: trained volunteers can help examine large collections of data and draw attention to patterns that researchers can investigate further. Higgs Hunters provides a specific comparison with algorithms for one image-classification task; Black Hole Hunters is applying volunteer classifications alongside simulations and algorithmic filtering to a different kind of astronomical data. Neither approach makes a candidate discovery conclusive without scientific follow-up. 318