A research team has published a sharp challenge to one of neuroscience’s most fraught questions: whether current methods can reliably tell us if something is conscious. Led by Hakwan Lau, the group argues that common experimental markers may be reading out general information processing rather than subjective experience itself.
The timing matters because the debate is widening fast. Artificial intelligence systems are becoming more sophisticated, and scientists are publicly wrestling with claims about consciousness in animals, fetuses and laboratory-grown brain organoids. Lau, director of the Center for Neuroscience Imaging Research within the Institute for Basic Science, said many current theories of consciousness appear to be supported by a range of experimental findings, but those findings may actually reflect general information processing rather than consciousness itself, making it hard to say the theories truly explain consciousness.
The analysis was published with collaborators from the Université de Montréal and New York University, and it does not try to decide whether any of those entities are conscious. Instead, it attacks the tools. The paper criticizes visual masking, binocular rivalry and perceptual threshold detection as paradigms that can alter both conscious experience and the brain’s broader ability to process information, leaving researchers with markers that may look persuasive without settling the question they are meant to answer.
That warning lands because the field has been here before. The authors point to the late nineteenth and early twentieth centuries, when strong but poorly grounded claims about consciousness helped fuel a major backlash and the rise of behaviorism. They also highlight blindsight and hemispatial neglect, conditions in which awareness can be separated from perception and behavior, as reminders that the mind does not always reveal itself cleanly through performance alone.
For researchers studying conscious AI, animal consciousness, fetal consciousness or organoids, the problem is practical as much as philosophical. If widely used markers are mostly tracking information processing, then some of the most public claims in the field may be resting on uncertain ground. The study does not offer a replacement method, and that is the central gap left open: if the current ones are not enough, what measurement would be?
That question is likely to hang over the next round of debate, whether the subject is a mouse, a fetus, a brain organoid or a system built with artificial intelligence. For now, the paper’s message is blunt: neuroscience may be asking the right question about consciousness with tools that are not yet sharp enough to answer it.
