Summary: A careful methodological review challenges the scientific foundations of much contemporary consciousness research. The authors argue that commonly used neuroscientific approaches are not reliably measuring subjective experience, and instead often capture general information processing. This ambiguity undermines many high-profile claims about sentience in AI systems, animals, fetuses, and lab-grown organoids.
The paper shows that widely used experimental paradigms can unintentionally conflate conscious awareness with non-conscious cognitive processing. The research team calls for clearer concepts and stronger methods to separate subjective experience from broader perceptual and computational capacities.
Key Facts
- The Measurement Crisis: The study reframes the debate by asking whether current neuroscientific tools actually measure consciousness itself, rather than only the brain’s capacity to process information or perform computations.
- The Methodological Loop: Common paradigms—such as visual masking, binocular rivalry, and perceptual threshold tasks—do more than alter reported awareness. They also disrupt baseline information-processing capacity, so experimental markers may reflect general cognitive function instead of subjective experience.
- The Non-Human Entity Boom: This conceptual ambiguity contributes to increasingly bold claims about animal consciousness, conscious AI, fetal sentience, and organoid awareness. Many published “markers” used to support these claims may primarily indicate processing capacity rather than first-person experience.
- The Threat of Behaviorist Backlash: The authors warn that repeating poorly grounded assertions risks a historical repeat. In the late 19th and early 20th centuries, weakly supported claims provoked a scientific backlash that helped spawn behaviorism and stalled mind research for decades.
- The Dissociation Roadmap: The paper highlights neuropsychological dissociations—like blindsight and hemispatial neglect—where conscious experience separates from perception and behavior. These cases demonstrate that subjective awareness and information processing can be biologically distinct, offering a path to more precise measurements.
- High-Stakes Ethical Grounding: Senior author Hakwan Lau emphasizes that scientific claims about consciousness now influence animal welfare, AI ethics, and biomedical regulation. Given these ethical stakes, the field needs conceptual clarity and rigorous methods before making policy-relevant assertions.
Source: Institute for Basic Science
As artificial intelligence systems grow more sophisticated, questions once confined to philosophy have entered public and scientific debate: Can AI be conscious? Do animals, fetuses, or organoids have subjective experiences?
A team led by Hakwan Lau at the Center for Neuroscience Imaging Research, Institute for Basic Science (IBS), together with colleagues from the Université de Montréal and New York University, has released a critical analysis arguing that current experimental and analytical methods are insufficiently precise to answer these questions. Their paper examines how consciousness is operationalized in neuroscience and shows many common approaches fail to clearly separate subjective experience from general information processing.
Rather than taking a position on whether specific non-human entities are conscious, the authors ask a more fundamental question: Do our methods actually measure subjective awareness? Lau explains, “Many findings cited in support of consciousness theories may instead reflect general information processing. That makes it difficult to claim those theories truly explain conscious experience.”
The review scrutinizes experimental paradigms such as visual masking, binocular rivalry, and threshold detection, noting these manipulations change both subjective reports and the brain’s ability to process sensory information. Consequently, neural or behavioral “markers” attributed to consciousness could instead index overall perceptual or cognitive capacity.
The authors caution that this methodological ambiguity has real-world implications. Recent debates have produced strong assertions about sentience in animals, AI systems, fetuses, and lab-grown brain organoids. If the criteria used to support those claims primarily track processing power, then policy and ethical decisions based on those criteria risk being premature or misguided.
To advance the field, the paper recommends focusing on cases where subjective awareness dissociates from processing and behavior. Classic neuropsychological conditions like blindsight—where individuals can respond to visual stimuli without conscious perception—and hemispatial neglect—where space and attention are disrupted while some processing persists—provide natural experiments. Studying such dissociations can help isolate neural signatures that more closely correspond to conscious experience.
The researchers argue that developing methods capable of isolating subjective experience is essential for assessing future claims about consciousness across biological and artificial systems. As Lau notes, “When scientific claims begin to shape ethical, legal, and regulatory frameworks, the standards for evidence must be exceptionally rigorous.”
The team hopes their analysis will prompt more stringent methodological standards and clearer conceptual distinctions throughout consciousness science, fostering progress grounded in reliable evidence rather than conflated markers.
Key Questions Answered:
A: Current tools often cannot distinguish a system that is subjectively experiencing something from one that is only processing information. Experimental manipulations used to study awareness can also change how much information the system handles, creating a confound in which measurements reflect computational capacity rather than first-person experience.
A: Blindsight shows that the brain can process visual data and guide behavior without conscious perception. This dissociation proves that processing and awareness are separable. By examining where and how these capacities diverge, researchers can design experiments that better test whether an advanced AI genuinely has subjective experiences or is simply performing complex information processing.
A: Overstated or poorly supported claims can provoke a defensive backlash that stalls research. A century ago, similar controversies contributed to the rise of behaviorism and a long period of skepticism toward consciousness studies. Today, with ethical and policy consequences at stake, the community must avoid repeating that mistake.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- Journal paper reviewed in full.
- Additional context added by staff.
About this AI and consciousness research news
Author: William Suh
Source: Institute for Basic Science
Contact: William Suh – Institute for Basic Science ([email protected])
Image: The image is credited to Neuroscience News
Original Research: Open access. “The Ethical Impasse of Current Consciousness Science” by Vincent Taschereau-Dumouchel, Jun Seo Hwang, Hakwan Lau, and Joseph E. LeDoux. Neuron. DOI: 10.1016/j.neuron.2026.04.007
Abstract
The Ethical Impasse of Current Consciousness Science
Growing public and scientific attention to consciousness in animals, fetuses, organoids, and AI has produced strong claims that often rest on markers of information processing rather than on measures of subjective experience itself. This analysis argues that such markers have limited value for adjudicating theories of consciousness. To move forward, the field must develop conceptual clarity and experimental tools that can better isolate the neural correlates of subjective awareness from the broader computations that support perception and behavior.