Springe direkt zu Inhalt

ERC PostDoc Call

Apply now for 3 Postdoc positions!

We are soliciting applications for three postdoc positions. The positions are funded through an ERC Starting Grant titled “CRACK- Cracking the Neural Code of Human Object Vision awarded to Radek by the European Research Council (ERC).

Funding is available immediately and maximally until 30.04.2024 (end of the funding period).

Overall goal of the project

The overall goal of the project is to explain how neural activity in the human brain gives rises to the cognitive phenomena of visual cognition. At the core there is a focus on visual object recognition. However, we investigate visual cognition more broadly and study its facets such as (conscious) perception, imagery, visual working memory, and its link to behavior.

To provide the pieces of evidence required for a new theory of visual cognition we strive for method innovation and technological integration. Specifically, the project will use advanced analysis methods (multivariate pattern analysis, RSA, encoding models) applied to fMRI (3T, and pushing the boundaries to layer-fMRI with 7T) and M/EEG data to know where and when the brain is activated during a particular task. To know what the activity represents, we will use computational modelling (in particular, deep learning).

Two research directions

We will pursue two research directions.

1) 7T Layer-fMRI for resolving information flow (Reference code: CICHY-ERC-7T)

We will distinguish, map and understand the role of feed-forward and feed-back information flow in human cortex for visual cognition. This will be done by resolving neural activity in space at the level of cortical layers using 7T-MRI, and in time through time-frequency decomposition of M/EEG data (e.g. Xie et al. 2020, Curr Biol), and their integration (Cichy et al., 2014 Nat Neuro; summarized in Cichy & Oliva, 2020 Neuron).

This work is in (already established and successful) collaboration with Nik Weiskopf (MPI for Cognitive and Brain Sciences) and Oliver Speck (LIN/OVGU).

2) Deep learning models of visual cortex (Reference code: CICHY-ERC-deep-learn)

We will develop, compare and use artificial neural networks as model to better understand the neural code of the visual system for visual contents. An example of primary work is Cichy at al., 2016 Sci Rep, a high-level perspective on our approach is summarized in Cichy & Kaiser 2020, Trends Cogn Sci.

This work is in time-tested and vibrant collaboration with Aude Oliva (CSAIL/MIT) and Gemma Roig (Goethe University Frankfurt).An example of collaborative work in this spirit is the Algonauts Project, which will be further conducted as part of CCN in 2021.

Whom we are looking for

We are looking for ambitious researchers that want to push the boundaries of acquisition and analysis of human neuroimaging data in service of theoretical insight. The ideal candidate fulfills the following criteria.

For both research directions:

  • a PhD in cognitive science, psychology, computer science (computer vision), biology or related subject
  • publications that prove your interest and expertise in vision science
  • very good programming skills (preferably in Matlab and/or Python
  • fluency in English (German not necessary)

For the 7T-layer-fMRI position:

  • experience with acquisition of fMRI data (7T fMRI experience is highly desirable, but not necessary)
  • proficiency in advanced analysis methods for fMRI (in particular MVPA, RSA, encoding models)
  • experience with acquisition/analysis of M/EEG data is desirable, but not necessary

 For the deep learning position:

  • a proven strong interest and track record of engaging in computational modelling of neural data, in particular with deep learning,
  • experience with computer vision is highly desirable

What else we offer

The lab is a vibrant, open, and diverse community of international researchers (currently from 11 countries: AR, CA, CL, CN, CU, DE, IN, IT, US, RU,TR).

We work together in the lab and in a wide network of collabrations (local, national, and world-wide). We will invite the successful applicant to join this team and to collaborate, to make use of our networks, and to actively enlarge them.

There is excellent access to research facilities the applicants will be free to use (MRI, EEG, neuro-navigated TMS, tactile (Braille) stimulation, auditory stimulation, eye tracking, ...) and computational ressources (high performance computing to be extended in discussion with the postdocs joining)

We are committed to open science and will support you to make your core code (github) and your data (e.g. osf) open.

We believe strongly in intellectual freedom. While the positions are offered in the context of a dedicated project, the goals are widely defined. You will be given opportunity and are expected to define studies and research directions on your own initiative.

The PI of the lab is dedicated to your scientific success and will provide strong support for your scientific career in the ways that you need.

Berlin is an excellent place for cognitive neuroscience, and a nice place to live.

How to apply

We encourage informal inquiries to Radek: rmcichy@zedat.fu-berlin.de.

To apply formally please send an email with a single pdf file that includes a motivation letter for one of the positions advertised, your CV (with publication list), and example publications (or manuscripts in preprint servers or in progress if you wish) to rmcichy@zedat.fu-berlin.de and cc to daniela.satici.thies@fu-berlin.de.

Please make sure to include the reference number for the position you are applying for in the email header and your pdf (layer-MRI/information flow: Cichy-ERC-7T; deep learning models: Cichy-ERC-deep-learn).

Deadline: September 27th, but applications are considered until the positions are filled.