My lab is looking for a PhD candidate in Modelling Predictive Error Responses:
https://www.ru.nl/en/working-at/job-opportunities/phd-position-in-computational-neuroscience-modelling-predictive-error-responses

For the NWO project ‘DBI2’ we are looking for a PhD candidate to study predictive error responses in the auditory cortex. The main goal is to design an experimental approach to distinguish between two alternative theories of predictive coding and processing.

Predictive coding and predictive processing are compelling theories to explain brain function. The idea that the brain continually maintains and updates an internal model of the outside world, and compares the incoming input with the expectations generated by this model, can explain many phenomena including adaptive behaviour and sensory effects such as oddball responses. However, until now there is no consensus on how such predictive coding could be implemented in real neural tissue. Importantly, there are two alternative theories on how error signals in predictive processing could be coded in neural signals: either as (1) top down signals from ‘higher order’ brain areas (hierarchical predictive coding, [2]) or (2) local signals, resulting in membrane potentials reflecting error signals ([3], for a review, see [1]). The goal of the research presented here, under the primary supervision of Dr Zeldenrust, is to design an experimental approach to distinguish between these two theoretical approaches.

Measuring error signals in neural tissue is experimentally challenging. Therefore, a direct exchange between theory and experiment is needed, so that hypotheses and specific predictions about which neurons to record from and stimulate and the results expected can be quickly updated for the design of optimal experiments. The student will work in close collaboration with the Englitz lab, so that there is a direct link between modelling, data analysis and experiment. As a PhD candidate you will use data on oddball paradigms [4,5], which provide the ability to directly observe predictions and distinguish them from prediction errors. The data are a combined approach of widefield imaging of the entire auditory cortex with local and layer-specific imaging using 2-photon recordings in the same animals. To directly test the top-down hypothesis, neurons in subareas of the prefrontal cortex will be transfected with an inhibitory opsin (eNpHR3.0) to modulate their top-down influence.

You will develop a model of the hierarchical interaction between the auditory cortex and the prefrontal cortex, in which error signals are either coded as top-down (theory 1) or local (theory 2). You will use this model to formulate testable predictions, distinguishing theory 1 from theory 2. These predictions will be tested by both analysing existing data from the Englitz lab and formulating new experimental paradigms that are suitable to distinguish between the local and top-down hypothesis.

[1] N’dri, A. W., Gebhardt, W., Teulière, C., Zeldenrust, F., Rao, R. P. N., Triesch, J., & Ororbia, A. (2024). Predictive Coding with Spiking Neural Networks: A Survey (arXiv:2409.05386). arXiv. https://doi.org/10.48550/arXiv.2409.05386
[2] Rao, R. P. N., & Ballard, D. H. (1999). Predictive coding in the visual cortex: A functional interpretation of some extra-classical receptive-field effects. Nature Neuroscience, 2(1), 79–87.
[3] Zeldenrust, F., Gutkin, B., & Denéve, S. (2021). Efficient and robust coding in heterogeneous recurrent networks. PLOS Computational Biology, 17(4), e1008673. https://doi.org/10.1371/journal.pcbi.1008673
[4] Nieto-Diego, J. & Malmierca, M. S. Topographic Distribution of Stimulus-Specific Adaptation across Auditory Cortical Fields in the Anesthetized Rat. (2016) PLOS Biol. 14, e1002397
[5] Lao-Rodríguez, A. B. ... Englitz B, (2023) Neuronal responses to omitted tones in the auditory brain: A neuronal correlate for predictive coding. Sci. Adv. 9, eabq8657

Profile

• You hold an MSc degree in computational neuroscience, mathematics, physics, computer science, AI or a similar computational field.
• You have experience in developing models (preferably in computational neuroscience) and can perform simulations, analytical derivations and advanced data analysis.
• You are a highly motivated, independent, critical, and creative researcher who wants to bridge the gap between real data and abstract theoretical models.
• You are a team player, ready to collaborate in a diverse, multidisciplinary research group.
• You have an excellent command of spoken and written English.
• You have experience in coding (Python and/or Matlab).

You will be appointed at the Biophysics of Neural Computation group, Donders Centre for Neuroscience, Donders Institute for Brain, Cognition and Behaviour. This vibrant and multidisciplinary institute studies the organisation and function of neural circuits across scales and techniques, from genes and proteins to behaviour, and from in vitro and in vivo experiments to computational modelling. It is an open and international environment, where young scientists from different backgrounds share a fascination for the interaction between brain structure and function. For PhD candidates, there is a lively graduate school and specialised support.