PhD Project 14
Connectivity-based fingerprinting of memory and language network dynamics
(last update 2019-06-27)
The acquisition of language is an immense mnemonic achievement as it requires linking thousands of verbal codes with semantic meaning. This project will provide fine-grained functional segmentations across the language and memory domains where overlapping dynamics support the acquisition of language by mnemonic processes. rs-fMRI and task-fMRI data will be utilised together with novel multivariate data analysis approaches to disentangle different mnemonic processes such as episodic encoding, hippocampal consolidation and neocortical long-term storage as well as characterize the interaction between different memory processes involved in language acquisition.
Connectivity gradients across language and memory systems.
Team members: Przezdzik, Haak (DCCN), Fernández, and Beckmann
The aim is to provide fine-grained functional segmentations across the language and memory domains where overlapping dynamics support the acquisition of language by mnemonic processes. Resting-state fMRI and task-fMRI data was used to investigate connectivity gradients and their behavioural correlates across language and memory domains.
The main findings of this PhD project are as follows:
(i) The gradual organisation of the hippocampus can be reliably mapped.
(ii) Inter-individual differences in the hippocampal gradient are behaviourally relevant: the spatial organisation of the gradient along the anterior-posterior axis at an individual level predicts recollection.
Findings (i) and (ii) are an important step forward in understanding the functional organization of the human hippocampus.
(iii) The anterior-temporal lobe connectivity gradients predicts distinct aspects of semantic cognition on an individual level. The inferior-superior gradient is predictive of differences in story comprehension whereas the anterior-posterior gradient maps onto differences in picture vocabulary naming.
Connectopic mapping in the language and memory regions and their association with behaviour.
Top panel: anterior-posterior connectivity gradient in the hippocampus. This gradient is related to recollection. Bottom panel: two connectivity gradients estimated in the ATL; the inferior-superior gradient is predictive of differences in story comprehension whereas the anterior-posterior gradient maps onto differences in picture vocabulary naming.
Multiple interdisciplinary approaches were used, including: resting-state fMRI, task-fMRI as well as behavioural measures acquired outside of the MRI scanner. All these data were analysed using novel data analysis frameworks such as Connectopic Mapping and Trend Surface Modelling in order to answer crucial questions regarding cross subject variability of the language processing as well as its interaction with the supporting systems, in particular the memory system.
The project involved collaboration of multiple researchers with different backgrounds and expertise. The methodological and theoretical input of every person involved was crucial to progress and come to the exciting conclusions and findings. This collaboration and cooperation at multiple levels shows how important and fruitful the “team science” is.
Further analysis focused on the gradual organisation of the hippocampus on individual level, and inter-individual differences in this organisation. Findings have been presented at OHBM 2018 meeting and are now under the review at the journal Cortex.
Anterior-temporal lobe connectivity gradient was also investigated. Findings have been submitted to the 2019 OHBM meeting, and the manuscript including these results is in preparation for submission.
This project integrates new fMRI data analyses methods with cognitive neuroscience of memory and language interaction. Assessing individual maps of neural representations (finger prints) is a novel method that has been developed by Christian Beckmann, which offers entirely new insights that allows to asses common features that represent linguistic information in more or less all of us and it assesses individual features that explain differences.