University of Cambridge > > Department of Psychiatry & CPFT Thursday Lunchtime Seminar Series > What Makes Some Intelligent Agents Conscious

What Makes Some Intelligent Agents Conscious

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact .

Chair: Dr Rudolf Cardinal

Abstract Much of current empirical research on the neural mechanisms of subjective experience is biased by a common set of experimental confounds; when we compare between conscious and nonconscious conditions (e.g. under coma, interocular suppression, visual making), besides subjective experience there is usually also a large difference in terms of the depth or strength of information processing. This means that so long as a theory associates consciousness with some sort of strong, global, stable, complex, or deeply integrated signals, it would be trivial to obtain apparent empirical ‘support’ – even if the theory otherwise makes little sense. Using new methods like multivoxel neural-reinforcement, we can partially address this problem, by inducing strong nonconscious representations for the control conditions. However, before we can do these experiments on a larger scale, our theories need to be conservative, i.e. prima facie plausible, given what we know about brains and cognition. Given this context, I will present a ‘just so’ story, of how an intelligent agent capable of general reasoning and predictive sensory coding may come to enjoy qualitative experiences. For this to happen, the relevant sensory representations need to demonstrate the properties of smoothness and sparsity. This way some implicit metacognitive mechanisms can learn the relational structure of these representations, and allow the agent to know, without effort, what an experience is like, i.e. how exactly it is subjectively similar to all other experiences. In this view, consciousness is something considerably more specific than perception that functions well.

Biography Hakwan Lau directs the Laboratory for Consciousness at the Riken Institute in Japan. He was formerly a tenured full professor at UCLA , and prior to that, an associate professor at Columbia University. His research group employs tools of cognitive neuroscience to study the mechanisms for visual perception, attention, and metacognition. He also interacts closely with people working in the philosophy of mind and philosophy of science. The interdisciplinary nature of his research will be reflected in his upcoming book, In Consciousness We Trust, to be published by Oxford University Press early in 2022. For more information on Dr Lau, please visit:

This talk is part of the Department of Psychiatry & CPFT Thursday Lunchtime Seminar Series series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2024, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity