Consciousness is not separate from the physical world, but is a property of certain forms of organized matter. Understanding how matter becomes conscious is the scientific challenge of our time. As AI grows ever more capable, one question becomes impossible to ignore: whether such systems can experience anything at all. To answer this, we must look to biological brains, the only conscious systems we know. We believe systems neuroscience is finally ready to reveal how the brain generates conscious experience.
Our concrete hypothesis is that consciousness is the process by which the brain builds symbols — stable, persistent representations of the world and the self — and composes them into an internal model of reality. We seek to understand the neural representation of these persistent symbols. Furthermore, because experience is composed from symbols, it can be engineered to move neuroscience from observing correlates of experience to precisely generating specific percepts, thoughts, and states of mind.
Astera Neuro will work closely with Astera's AI team. The principles we uncover in the brain can guide the creation of AI world models, and running these brain-inspired models in diverse, real-world situations is the sharpest test our neural theories can face. We believe understanding and engineering intelligence are one endeavor: neuro-AI.
The physical world model. Conscious experience begins with a stable, coherent model of the physical world comprising persistent objects, and the relations between them. We are mapping the brain’s object-tracking machinery, and the factorization of what an object is from where it is, that we believe lies at the foundation of all conscious experience.
Thought. Thought is built from the same machinery: abstract symbols, flexibly composed. We study how the brain forms concepts and recombines them into new configurations, exploring the hypothesis that the generativity of language and reasoning is inherited from the compositional structure of the physical world itself.
The self. Among all the objects the brain models, one is special: the self, a body sensed from within with a felt sense of agency. We ask how the brain constructs the self, how the self acquires motivations, and how the self can be extended, merged, or transformed.
The architecture of the brain. Percepts, thoughts, and the self arise from interactions among many regions, so the world model cannot be understood without knowing how the brain is wired. We are building a functional and anatomical atlas of what each region represents and how regions connect, to guide where to look and what to test.
Reading and writing experience. We are building the tools to test these ideas causally — large-scale, high-density, chronic read/write at single-neuron resolution, across model systems and, ultimately, in people, so that we can write sequences of symbols into the brain to evoke specific experiences. The same capability carries a direct therapeutic promise. Tools precise enough to write experience are tools that can help restore it: to give back sight or movement lost to injury, and to relieve the disordered perception, memory, and mood at the root of so much neurological and psychiatric disease.
The horizon. Understanding the fundamental mechanism for generating consciousness opens the deepest questions we can ask: whether a mind can be extended, merged, or carried onto a new substrate, and how to know whether any system, biological or artificial, is conscious at all. The ultimate mission of Astera Neuro is to take consciousness from evolved to engineered.
MEDICAL IMPACT. Deep understanding of the brain’s internal model of the world and the self, together with the next generation brain-machine interfaces we are developing, will open fundamentally new approaches to treating neurological and psychiatric disease. Brain disorders including Alzheimer’s, depression, schizophrenia, PTSD, OCD, and insomnia, cost over $1.5 trillion annually in the US alone, equivalent to 9% of GDP. More importantly, they are among the greatest sources of human suffering, because they alter perception, thought, memory, emotion, and ultimately the sense of self.
These diseases arise due to aberrant neural representations and dynamics. By uncovering the fundamental principles underlying the brain’s internal model, including how consciousness is turned on and off and how thoughts are generated, while simultaneously developing the ultra-high bandwidth and resolution read/write technologies necessary to test these principles, we will be able to treat these diseases at the root.
Doing science that can meaningfully reduce human suffering is one of our greatest sources of motivation. If we can understand and restore the neural representations that underlie healthy cognition and experience, it will not only transform medicine, it will provide powerful evidence that we genuinely understand the brain’s internal model.
Our ambition is so large that we will not be able to realize it alone; this is work that we must all do together. Everything we do at Astera is in the open, in line with Astera’s Open Science Policy. We will be delighted if others take our ideas and build on them , because this will help us reach our north star goals faster. This work is too consequential to lock behind closed doors, so we will share our results, data, code, methods, and technology as they come. We hope they will be useful to the neuroscience community, and as they roll out, we warmly welcome feedback. We are also planning a postbac program for college graduates and an internship program for graduate students who would like to learn more about our new way of doing science. Stay tuned!