Behavior-dependent interactions between frontal and somatosensory cortices

Project Information

AG-GF

Grant agreement ID: 659719

DOI

10.3030/659719

Project closed

EC signature date 20 October 2015

Start date 1 August 2016

End date 31 July 2019

Funded under

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Total cost

€ 278 674,20

EU contribution

€ 278 674,20

Coordinated by

THE HEBREW UNIVERSITY OF JERUSALEM
Israel

Periodic Reporting for period 2 - AG-GF (Behavior-dependent interactions between frontal and somatosensory cortices)

Reporting period: 2018-08-01 to 2019-07-31

Summary of the context and overall objectives of the project

Problems being addressed:
The mammalian brain integrates behaviorally relevant sensory information by recruiting large parts of the neocortex to enable precise perception, apt decisions and appropriate actions. However, we still poorly understand which brain regions are activated during specific cognitive functions and how they relate to behavioral parameters. For example, it is unknown which cortical areas maintain short-term memory, or which areas are involved in learning a new task.

Overall objectives:
For the outgoing phase: To study the large-scale cortical dynamics as mice perform a tactile memory task. Specifically, we were interested in finding the cortical locations of sensory integration and especially short-term memory maintenance.
For the return phase: To study the population dynamics in the auditory thalamus during learning of an auditory go/no-go discrimination task.

Importance for society:
This findings highlight the dynamic plasticity of neuronal networks at the mesoscale level, which greatly differs from artificial hard-wired networks. This project expands our understanding of how the healthy brain flexibly maintains memory, and may aid in understanding the disease-related brain. Several neurodegenerative diseases such as Parkinson and Alzheimer are characterized by memory impairment which is one of the most common complaints from patients. The frontoposterior network studied in this project has been linked to neurodegenerative disorders and our work adds on to its role in short-term memory. Having the ability to also manipulate the network is a big advantage that is not enabled in human studies. In addition, we show the complex neuronal dynamics during learning of a new task. Learning involves many cortical areas which eventually leads to a refinement of relevant association cortex that may underlie gaining expertise. The comprehensive understanding of the neuronal correlates of learning may aid in diagnosing and improving subjects with learning disorders, e.g. dyslexia.

Conclusions of the action:
For the outgoing phase: We found multidimensional cortical dynamics involved in both sensory integration and short-term memory and also related strongly to the behavioral parameters of each mouse. Mice either deployed an active or passive strategy. Independent of strategy, whisker-related posterior areas encoded choice early after touch. During the delay, in contrast, persistent cortical activity was located medio-frontally in active trials but in a lateral posterior area in passive trials. Perturbing these areas impaired performance for the associated strategy and also provoked strategy switches. Thus, depending on behavioral strategy, cortical activity is routed differentially to hold information either frontally or posteriorly before converging to similar action.
For the return phase: We found that thalamic responses in expert mice encoded the choice of the mouse, as we could rapidly discriminate between hit and miss trials. In addition, we observed opposite effects on the go and no-go responses. In recordings sites preferring the ‘go’ sounds, responses to ‘go’ on hit trials increased. In recordings sites preferring the ‘no-go’ sounds, responses to ‘no-go’ sounds decreased. There was a strong correlation between the time in which the mouse crossed behavioral threshold to expert level and the time in which thalamus responses displayed the strongest modulation (either go enhancement or no-go suppression). These results show that the auditory thalamus encodes task- and learning-related information.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

Work performed during the project:

Outgoing phase:
1) Setting up wide field imaging in behaving mice. Duration of this part was 2 month.
2) Training protocol for texture discrimination with a delay component. Training duration was around 2 month per mouse.
3) Wide-field imaging and data analysis. For each mouse we localized areas of interest for further analysis and single cell imaging. Duration - approximately 12 months.
4) In parallel to wide-field imaging and data analysis, in 5 out of the 8 mice we performed a craniotomy over areas of interest and used two-photon microscopy to image single cells. Duration - approximately 8 months.
5) Optogenetic experiments to assess whether the areas of interest detected in the wide-field imaging are behaviorally relevant. Duration - approximately 6 months.
6) Results were finalized and paper were written. One paper containing most of the results was published in Neuron on the 22nd of August 2018. Duration - approximately 6 months.
Return phase:
1) Setting up fiber photometry in behaving mice. Duration of this part was 2 month.
2) Imaging in the thalamus and training the mice - 6 months
3) Data analysis - 2 months
4) Results were finalized and paper is under final preparation.

Exploitation and dissemination: Results of the outgoing phase are published in a peer-reviewed journal (Neuron, Aug 2018). Results from the return phase are being prepared for publication. The project was also presented in numerous conferences, among others the Neuroscience SFN conference, and the FENS Forum.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

State-of-the-art results and impacts:

1) Short-term memory is maintained in different areas in the same task. Thus different cortical routes can lead to a similar behavioral action.

2) Area P is a newly identified area where we are the first to show its functional and behavioral relation. In cases where area P maintains short-term memory, we show that frontal cortex does not carry any information concerning the future action. This contradicts many previous studies claiming that frontal cortex maintains information necessary to plan a future movement.

3) Optogenetic perturbation of one area in some cases impairs performance, but in other cases provokes strategy switches. Thus, some mice can overcome the perturbation and switch to the alternative strategy. These results undermine the notion that one area is causally related to one process. Rather, we propose that cortical dynamics may flexibly reroute information to eventually reach the same outcome.

4) M2 or P hold a different type of short-term memory (motor or sensory related) which has not been shown before.

5) Learning involves an orchestrated refinement of many association areas to highlight task-related areas.

6) Learning related neuronal changes occur much before the actual learning and much before the stimulus is resented, highlighting the complexity underlying learning.

These results may have a strong socio-economic impact. It would be interesting to similarly investigate the location of working memory in the human brain in subjects that come from different socio-economic backgrounds. It may be that subjects from a high socio-economic background may display higher frontal activity compared to subjects from a low socio-economic background which would show higher posterior activity. Thus fronto-posterior interactions within each individual may promote our understanding of the neural correlates of socio-economic factors. In addition, these studies shed light on how the brain is modulated during learning, which may aid in understanding the neuronal mechanisms underlying learning deficits in many human subjects, e.g. dyslexia. This may have a wide societal implication by identifying and possibly treating specific cortical areas to eventually improve learning capabilities.

Graphical abstract for Gilad et. al 2018.