Representation, Transformation, Dynamics

General Discussion
1

Hi everyone,

I am VERY sorry to ask following questions.
VERY sorry!

I am quite new in the field of neuroscience.
Therefore, I am taking the coursera course “Computational Neuroscience” as well as reading the book “How to build a brain” by Mr. Eliasmith,

I am doing this for a practical course for my university, in which I should explain the work that need to be done to control a robotic system with body signals (EEG / EMG).

However, I am stuck at understanding the principle of NEF.
I have understood what encoding and decoding means.
But I am failing to capture the purpose of transformation and dynamics after encoding and decoding.

I read several explanation on different websites, but they confused me even more.
(I am feeling so silly…)

I think that I can work well with questions.

Encoding:
How does a stimulus cause a pattern of responses?
Goal: Mapping from a stimulus (e.g. audio) to the responds of neurons.

Decoding:
How can we reconstruct the stimulus?
Was it an audio or a light signal? And if so, how can the signal be described as?
Goal: Mapping from the responses of neurons to the stimulus.

What is the question of Transformation and Dynamics?
What is the purpose? What is the goal?

Hopefully, someone has got a second for answering me.
I would be very happy!

Best

2

Transformation:
How can we connect groups of neurons to compute functions for complex behaviour? Because neurons do more than just represent stimulus and we want to do more than reconstruct the stimulus from the neurons. We want to group neurons together so they can create behaviour.

Dynamics:
What about functions that change over time? Because behaviour often requires taking the past into account via integration or some other time-dependent operation. We can get time-dependent functions by connecting neural populations to themselves.

Does this help?

3

Yeah, that helped a lot!!!
Thank you!!!