My research on the Overfitted Brain Hypothesis, which is on how dreaming promotes generalization, was given the cover of Patterns (May issue, 2021)
Scientific research.
I work broadly across a set of overlapping fields, and my research has been profiled in outlets like The Guardian, Wired, and Discover Magazine. Generally the type of science I like to do is abstract, as in modeling and theory building. I am often either the first or the last author of any paper I'm on, and really try to only put out only original papers that I believe in. And sometimes I have no funding for what I'm working on, but do it anyways. Mainly I focus on three areas of study.
I work broadly across a set of overlapping fields, and my research has been profiled in outlets like The Guardian, Wired, and Discover Magazine. Generally the type of science I like to do is abstract, as in modeling and theory building. I am often either the first or the last author of any paper I'm on, and really try to only put out only original papers that I believe in. And sometimes I have no funding for what I'm working on, but do it anyways. Mainly I focus on three areas of study.
a) Formal mathematical theories of consciousness.
In graduate school I helped develop aspects of Integrated Information Theory, which is probably the leading scientific theory of consciousness. Particularly, I worked on how integrated information is related to causation, and how and why consciousness peaks at a certain spatiotemporal scale of brain function. Since then I have been focused on issues around developing theories of consciousness, or what I like to call "meta-theory" problems, such as the difficulties, perhaps impossibilities, of constructing falsifiable theories of consciousness.
Selected papers.
In graduate school I helped develop aspects of Integrated Information Theory, which is probably the leading scientific theory of consciousness. Particularly, I worked on how integrated information is related to causation, and how and why consciousness peaks at a certain spatiotemporal scale of brain function. Since then I have been focused on issues around developing theories of consciousness, or what I like to call "meta-theory" problems, such as the difficulties, perhaps impossibilities, of constructing falsifiable theories of consciousness.
Selected papers.
- Kleiner, J., & Hoel, E. (2021). Falsification and consciousness. Neuroscience of Consciousness.
- Wenzel, M., Han, S., Smith, E. H., Hoel, E., Greger, B., House, P. A., & Yuste, R. (2019). Reduced Repertoire of Cortical Microstates and Neuronal Ensembles in Medically Induced Loss of Consciousness. Cell systems, 8(5), 467-474.
- Hoel, E., Albantakis, L., Marshall, W., & Tononi, G. (2016). Can the macro beat the micro? Integrated information across spatiotemporal scales. Neuroscience of Consciousness, 2016(1).
b) Formal mathematical theories of emergence and causation.
Emergence is not a philosophical issue, but a scientific one. In this vein, I am interested in creating theories that identify intrinsic scales of function, that is, system scales where information and/or causation peaks, and which help scientists make modeling and intervention choices. Additionally, I am interested in measures of causation, especially because I think causal relationships can be stronger at higher scales. Therefore, I and co-authors have proposed information-theoretic measures in things like Boolean networks and artificial neural networks. I am also interested in applying these sorts of analyses to real data to demonstrate the power of the theoretical framework.
Selected papers.
Emergence is not a philosophical issue, but a scientific one. In this vein, I am interested in creating theories that identify intrinsic scales of function, that is, system scales where information and/or causation peaks, and which help scientists make modeling and intervention choices. Additionally, I am interested in measures of causation, especially because I think causal relationships can be stronger at higher scales. Therefore, I and co-authors have proposed information-theoretic measures in things like Boolean networks and artificial neural networks. I am also interested in applying these sorts of analyses to real data to demonstrate the power of the theoretical framework.
Selected papers.
- Varley, T., & Hoel, E. (2021). Emergence as the conversion of information: A unifying theory. arXiv preprint 2104.13368.
- Mattsson, S., Michaud, E. J., & Hoel, E. (2020). Examining the causal structures of deep neural networks using information theory. Entropy 22(12), 1429. Editor's Choice Award winner.
- Klein, B., & Hoel, E. (2020). The Emergence of Informative Higher Scales in Complex Networks. Complexity, 2020.
- Evolution leads to emergence: An analysis of protein interactomes across the tree of life. Hoel, E., Klein, B., Swain, A., Grebenow, R., Levin, M. bioRxiv preprint.
- Albantakis, L., Marshall, W., Hoel, E., & Tononi, G. (2019). What caused what? A quantitative account of actual causation using dynamical causal networks. Entropy, 21(5), 459.
- Hoel, E. (2017). When the map is better than the territory. Entropy, 19(5), 188.
- Hoel, E., Albantakis, L., & Tononi, G. (2013). Quantifying causal emergence shows that macro can beat micro. Proceedings of the National Academy of Sciences, 110(49), 19790-19795.
c) Dreams and sleep.
I was lucky to work with some of the best sleep scientists on the planet during my PhD in neuroscience. This has also led to my recent development a theory of the neural purpose of dreaming, wherein the goal of dreams is to combat daily overfitting of the brain, which I also think just might be the purpose of fictions like TV shows or novels.
Selected papers.
I was lucky to work with some of the best sleep scientists on the planet during my PhD in neuroscience. This has also led to my recent development a theory of the neural purpose of dreaming, wherein the goal of dreams is to combat daily overfitting of the brain, which I also think just might be the purpose of fictions like TV shows or novels.
Selected papers.
- Hoel, E. (2020). The Overfitted Brain: Dreams evolved to assist generalization.
- A good scientific review of the Overfitted Brain Hypothesis (not by me) can be found here.
- Hoel, E., Albantakis, L., Cirelli, C., & Tononi, G. (2016). Synaptic refinement during development and its effect on slow-wave activity: a computational study. Journal of neurophysiology, 115(4), 2199-2213.
Samples of news media discussing my scientific work.
- The Guardian (2021) Weird dreams train us for the unexpected, says new theory
- Gizmodo (2021) Our Weirdest Dreams Could Be Training Us for Life, New Theory Says
- Medical News Today (2021) Does the weirdness of dreams help keep the brain flexible?
- Neuroscience News (2021) Our Dreams' weirdness might be why we have them
- Scitechdaily (2021) New AI-inspired theory of dreaming: Our dreams' weirdness might be why we have them
- AAAS Eurekalert! (2021) Our dreams' weirdness might be why we have them, argues new AI-inspired theory of dreaming
- Tufts Now (2021). A New Theory for Why We Dream
- Discover Magazine (2020). How Artificial Neural Networks Paved the Way For A Dramatic New Theory of Dreams
- Forbes (2018). 30 Under 30 profiles
- Wired (2017). New Math Untangles the Mysterious Nature of Causality
- Nautilus (2017). A Theory of Consciousness Can Help Build a Theory of Everything