Active inference, a mechanistic framework, has been proposed as a foundation for a unified theory of consciousness, but current proposals need refinement to become a process theory. Computational modeling can help by implementing and validating theoretical concepts. Existing models show promise but remain preliminary; they require testing against new empirical neural data to improve predictive and structural validity. A key challenge is generalizing the model to cover the full range of consciousness phenomena, especially phenomenological aspects of experience. Despite these gaps, the approach is valuable for advancing theory and holds potential for future research.
Consciousness remains a hard problem: why does neural activity produce subjective experience, like the taste of chocolate or the feeling of a caress? Understanding this has medical and ethical implications, from assessing consciousness after brain injury to deciding whether nonhuman animals, fetuses, organoids, or advanced machines are conscious. A comprehensive theory is needed to determine which systems experience anything and to define ethical boundaries. An adversarial collaboration championing open science aims to test whether major current theories hold up.
Advancing scientific research on consciousness is important for addressing clinical and ethical issues in neurology and mental health. To support this field, funding priorities must be set carefully, and challenges such as job creation and media misrepresentation need to be addressed.