Periodic Reporting for period 1 - MESO_AG (Mesoscale dissection of neuronal populations underlying cognition)
Berichtszeitraum: 2022-06-01 bis 2024-11-30
The brain is responsible for cognition, broadly defined as thinking, by combining mental processes such as sensory integration, perception, and working memory. One of neuroscience’s major challenges is understanding how the brain encodes cognition as a whole. The biggest obstacle to this goal is the complex nature of the brain, which contains billions of entangled neurons that form a dynamic, ever-changing network. We propose to use the mouse model to study cognitive processing streams across the brain. Our novel approach is to train mice on different cognitive tasks and use state-of-the-art imaging methods to record from as many brain areas in parallel. We aim to reveal and dissect brain-wide processing streams underlying cognitive processes such as sensory integration, working memory and perception. Next, we will zoom in to dissect a specific node or edge using multi-area two-photon microscopy, labeling techniques, and optogenetics. We expect to gain a significant impact in understanding how the brain encodes cognition in a brain-wide manner. Our aim is to obtain a brain-wide cognitive map that will aid in understanding cognition as a whole in both the healthy and the diseased brain.
We have already made substantial progress in all 3 described work packages.
for work package 1: We have obtained a complete dataset (n=10 mice) of cortex-wide imaging during both 'what' and 'where' tasks. We find two distinct pathways, a frontal and posterior, that are strongly related to the internal strategy of the mouse (i.e. either passive or active) rather to the type of task. This results validates our initial hypothesis that processing streams are governed by internal states. We have now also added additional datasets such as multi-fiber photometry (n=9 mice) and thalamo-cortico-amygdala network (n=4 mice) performing the same task and find similar posterior-frontal strategy-dependent divergence beyond the cortex. A manuscript is currently under consideration in Cell
Work package 2: We have managed to establish the behavioral paradigm for match-to-sample delayed interhemispheric task. We obtained a full data set (n=5 mice) of dual hemisphere cortex-wide dynamics during the task. These mice were also trained to match between two textures without a delay. In short we find that mice use different strategies to perform the tasks, active or passive, which governed the location of information transfer. In the simultaneous task, active mice transferred information between the two barrel cortices, whereas passive mice transferred information from posterior area P. In the delayed task, passive mice efficiently transferred working memory information between P areas. Active mice displayed a lot of impatient activity and used barrel cortex to transfer information but with a reduced performance. The results of this study are currently being consolidated and we aim to submit a full manuscript within 3 months.
Work package 3: We have established a 2AFC task to study perception and managed to obtain a psychometric curve from 3 mice. We have also imaged cortex-wide dynamics during task performance and found several brain areas, including barrel cortex, to contain perceptual related responses. We are now continuing to train and image more mice in order to substantiate the results.
for work package 1: We have obtained a complete dataset (n=10 mice) of cortex-wide imaging during both 'what' and 'where' tasks. We find two distinct pathways, a frontal and posterior, that are strongly related to the internal strategy of the mouse (i.e. either passive or active) rather to the type of task. This results validates our initial hypothesis that processing streams are governed by internal states. We have now also added additional datasets such as multi-fiber photometry (n=9 mice) and thalamo-cortico-amygdala network (n=4 mice) performing the same task and find similar posterior-frontal strategy-dependent divergence beyond the cortex. A manuscript is currently under consideration in Cell
Work package 2: We have managed to establish the behavioral paradigm for match-to-sample delayed interhemispheric task. We obtained a full data set (n=5 mice) of dual hemisphere cortex-wide dynamics during the task. These mice were also trained to match between two textures without a delay. In short we find that mice use different strategies to perform the tasks, active or passive, which governed the location of information transfer. In the simultaneous task, active mice transferred information between the two barrel cortices, whereas passive mice transferred information from posterior area P. In the delayed task, passive mice efficiently transferred working memory information between P areas. Active mice displayed a lot of impatient activity and used barrel cortex to transfer information but with a reduced performance. The results of this study are currently being consolidated and we aim to submit a full manuscript within 3 months.
Work package 3: We have established a 2AFC task to study perception and managed to obtain a psychometric curve from 3 mice. We have also imaged cortex-wide dynamics during task performance and found several brain areas, including barrel cortex, to contain perceptual related responses. We are now continuing to train and image more mice in order to substantiate the results.
Specifically for each objective, our outcomes are bound to have a high impact. For the first objective, we show that brain-wide networks are governed by internal states rather than external parameters. This outcome is a breakthrough on our understanding of brain-wide processing as it completely turns around some of the most basic theories regarding higher-order sensory processing. For the second objective, we are the first to show that working memory is transferred across hemispheres from area P, which is a very posterior and unstudied area. This result alone is very surprising since most of the field is concentrated in the frontal cortex and is bound to make a lot of impacts on the field. For the third objective, we find that perception is also encoded in lower-order areas such as barrel cortex and not only in higher-order areas. Importantly, the outcomes of each work package can now be merged together in order to gain wide insights onto how the brain encodes cognition as a whole. For example we find that area P is involved in passive sensory integration and also in maintaining and transferring working memory. Therefore area P may serve as a critical hub in encoding several different cognitive functions. We aim to further combine outcomes from different project in order to obtain a breakthrough in understanding brain-wide dynamics underlying cognition.