Neural Dynamics of Visual Cognition Group
I joined the lab in September 2018 as an intern student, and in October 2019 I started my PhD. Before that I did a BA in Philosophy (University of Trento, Italy) and a MSc in Cognitive Neuroscience (CIMeC, Italy).
Things in life that I appreciate are: movies (independent/arthouse), playing Frisbee, mountain hikes and backpacking, electronic music, spending time in nature and parks, learning, cooking and eating, sleeping, sunbathing.
General research interests
My research aims at understanding visual processing in the human brain, following a computational approach. For this, I model M/EEG and fMRI data through machine learning and deep learning algorithms. I am also interested in the intersection between biological and artificial vision, and in how this multidisciplinarity is advancing both fields of research.
I am funded by an Einstein Center for Neurosciences Berlin PhD fellowship. The funding started in October 2020, and will last until October 2023. I am also a member of the Bernstein Center for Computational Neuroscience Berlin.
I currently have three ongoing and one planned research projects.
Ongoing research projects
- Project 1: We collected a large and rich dataset of human EEG responses to naturalistic images (10 participants, each with 82,160 trials spanning 16,740 image conditions), validated it through several analyses, and released it as a tool to foster research in visual neuroscience and computer vision. Its largness enabled for the first time building encoding models end-to-end from EEG data. That is, we for the first time train artificial deep neural networks from scratch to predict brain responses directly from the pixels of the image. Check out the preprint with links to the dataset and code here.
- Project 2: What is the idiosyncratic contribution of each mid-level visual area (V2, V3, V4) to the processing of information along the visual ventral stream? How are these areas functionally and algorithmically characterized? We are using the Natural Scenes Dataset to tackle these underexplored questions.
- Project 3: How are the millisecond brain dynamics shaped by naturalistic videos? To investigate this, we are collecting the EEG responses to the Algonauts 2021 challenge stimuli videos.
Planned research projects
- Project 1: I will collaborate on the third edition of the Algonauts project. In this edition of the challenge participating teams will leverage the unprecedented size of the natural scenes dataset to build encoding models that predict the fMRI responses to natural scenes. More info coming soon!
Student supervision & opportunities
I am supervising the following student: Furkan Özkan Hüseyincan.
I have supervised the following student: Andrei Kitaitsev.
I am currently not looking for intern students. However, if you have research interests in line with mine and if you have a good computational background (Python, machine/deep learning), feel free to get in touch with me: if I see a good fit and have a project in mind I might consider supervising you!
If you are interested in doing an internship/thesis research project at the intersection of visual neuroscience and computer vision, and if you have a good computational background (Python, machine/deep learning), feel free to get in touch with me via email! There are several exciting research projects I don't have the time to work on myself, but you could work on under my supervision.