Forsøksdyr: Neural computation of space and time addressed by electrophysiology in vivo


Godkjenningsdato 01.11.2018

A major goal for 21st century neuroscience is to understand neural computation in high-end association cortices. This is challenging because neural codes in these systems are multiplexed and intermixed. A rare exception is the spatial-representation system of the hippocampal and entorhinal cortices. In these regions, neural firing correlates are so evident that cells have been given simple, descriptive names, such as place cells, grid cells, head direction cells, and border cells. The presence of functionally dedicated cell types opens a window for studies aiming to identify computational algorithms at the peak of the cortical hierarchy. Taking advantage of this opportunity, we have uncovered many of the functional elements of spatial computation, at the nuts-and-bolts level. However, how the firing patterns are generated, integrated and read out is poorly understood. Since such processes likely take place in distributed neural circuits, we need parallel data from many hundreds of entorhinal-hippocampal cells, of multiple categories, to decipher mechanisms. The aim of the research program for which we seek approval with the present application, is to introduce new recording tools to obtain simultaneous single-cell-resolution activity data from up to 1,000 entorhinal-hippocampal neurons in freely behaving rodents, and to use these tools to understand local-circuit mechanisms of pattern formation and pattern organization. The proposed studies will pioneer our understanding of neural-circuit computation in non-sensory association cortices. Because representation of spatial location has appeared as one of the first cognitive functions to be understood in mechanistic detail, the project opens possibilities for understanding general principles of computation in cortical microcircuits, which we see as a prerequisite for comprehending and treating any disease that affects the brain – along the entire spectrum of neurological and psychiatric diseases, but with Alzheimer´s disease as a prime example, given the early damage of hippocampal and entorhinal systems in this disease.

The project is an extension of past applications (most recently project 7163 from 2014) but introduces state-of-the-art electrophysiology techniques for collecting parallel data from hundreds of labelled neurons simultaneously, an achievement that we expect will put us on the track of some of the core mechanisms of brain computation. All major new technologies have been piloted and are already covered by approved ongoing FOTS projects.

We expect to use 1,080 rats and 2,880 mice over a period of 4 years. The animals will receive implants of ultrathin electrodes (voltage sensors) for recording electrical activity in the hippocampus or cortex of the brain, using previously approved fast-recovery anaesthesia and anaelgesia protocols. Over the subsequent weeks, activity is recorded while animals rest or search for food rewards, with minimal food deprivation. No pain or discomfort is expected during behavioural testing; discomfort would be detectable instantly from the animal's behaviour. Numbers of animals are reduced by recording as many cells as possible at the same time. Animals are housed in enriched environments and receive daily handling. Experiments are supplemented by computational modelling in order to minimize needs for experiment and maximize conceptual steps for each experiment.