Cookies on this website
We use cookies to ensure that we give you the best experience on our website. If you click 'Continue' we'll assume that you are happy to receive all cookies and you won't see this message again. Click 'Find out more' for information on how to change your cookie settings.


Michał J. Wójcik



Cross-domain abstraction and generalisation of latent structures within the environment


My research focuses on how the brain gives rise to representations of complex relations that can be encountered in the environment. Furthermore, a special focus of my DPhil is the process of generalising them across different domains to facilitate learning.

To probe the neuronal activity for the emergent properties necessary for generalisation, I use human electroencephalography and high resolution neural population recordings from animals. Novel machine learning algorithms and representational geometry are employed to open a window into the computational processes underlying generalisation and its link to flexible, inteligent behaviour. 

Some artificial systems like, for example, recurrent neuronal networks, are also able to learn an abstract representation and achieve an acceptable level of generalisation. However, training them requires more resources than human learning. Understanding what makes human learning so energy-efficient and fast could provide useful insights in such fields like machine learning and artificial intelligence. 

I am supervised by Mark Stokes and Kia Nobre, and funded by the Clarendon Fund Scholarship in partnership with the Saven European Scholarship.