Published in Nature on June 5, 2025, the Cogitate Consortium's landmark study subjected Integrated Information Theory (IIT), proposed by Giulio Tononi of the University of Wisconsin, and Global Neuronal Workspace Theory (GNWT), led by Stanislas Dehaene of the College de France, to what the authors called the first large-scale open-science adversarial collaboration in consciousness science. Funded by the Templeton World Charity Foundation as part of a roughly $20 million research program launched in 2019, the study tested IIT and GNWT with 256 participants using fMRI, magnetoencephalography (MEG), and intracranial EEG. The project traces back to a 2018 Allen Institute workshop where rival groups agreed to design a single, shared experiment; planning followed that year, data collection began in 2019, and analysis ran through 2024 before the paper's release. The adversarial collaboration model was famously advocated for by Nobel laureate Daniel Kahneman over 20 years ago and represents a fundamental shift in how science can be done.
The methodological design was engineered to eliminate the post-hoc rationalization that has long plagued consciousness research. Before the Consortium ran a single participant, the proponents of each theory sat across from each other and agreed on what their theory specifically predicted and what would count as evidence against it. More than 250 people participated in the consortium's study, which tested three predictions from GNWT and IIT: where in the brain content of conscious experiences is represented, how conscious experience is maintained over time, and how different regions communicate to generate conscious experience. Independent research teams conducted the study across seven laboratories worldwide. All researchers agreed beforehand to accept the outcome, whatever it may be.
The results vindicated neither camp. IIT's key prediction, that conscious perception depends on sustained synchronization in posterior brain regions, was not supported by the data. IIT correctly anticipated where information would be decoded but failed to show the predicted patterns of sustained internal connectivity. GNWT's key predictions about brief prefrontal "ignitions" fell short, especially when it came to maintaining awareness over time. Researchers did find information about conscious content in visual, ventrotemporal, and inferior frontal cortex, with sustained responses in occipital and lateral temporal cortex reflecting stimulus duration. That partial, theory-straddling pattern belonged to neither framework as written. Beyond challenging the theories, the authors present the study as an alternative approach to advance cognitive neuroscience through principled, theory-driven, collaborative research, and they highlight the need for a quantitative framework for systematic theory testing and building.
Skeptics and theorists have pushed back in distinct ways. The Cogitate results do not end IIT or GNWT as research programs; both theories have explanatory scope well beyond the specific predictions the Consortium preregistered, and both have proponents who argue that the operationalizations tested were imperfect representations of the theories' deepest commitments. The backlash against IIT ran further still: a letter signed by over 100 researchers declared IIT pseudoscientific, arguing that its core claims, particularly the identification of consciousness with integrated information and its panpsychist consequences, are unfalsifiable in practice. Tononi and colleagues published detailed rebuttals defending IIT's empirical commitments and mathematical rigor, while methodological commentary argued that calling a theory pseudoscientific was itself a category error. Correspondence author Lucia Melloni, of the Max Planck Institute for Empirical Aesthetics, offered the study's own verdict on all of it: "True progress comes from making theories vulnerable to falsification, not protecting them," Melloni said. "This wasn't about picking a winner; it was about raising the bar for how we test ideas." The Cogitate Consortium is now analyzing the results of a second large-scale experiment to further test GNWT and IIT, and is releasing its entire dataset to the scientific community so that researchers worldwide can retest, refine, and expand on the findings.
The most expensive, most methodologically airtight test consciousness science has ever run produced no winner, which means the field is now obligated to build better theories rather than defend old ones.