{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

NE101 Lecture Notes

Neighborly with short cuts interesting properties

Info iconThis preview shows pages 27–30. Sign up to view the full content.

View Full Document Right Arrow Icon
Neighborly with short-cuts interesting properties short path length & clustered Many real-world networks are small world networks. Why is this useful for the brain? 2. Data analysis Use computational tools to characterize observed brain activity. Use sophisticated tools to characterize EEG activity. The power spectrum Help us identify important rhythms Vital: we're overrun with data Ex: Multielectrode array 100 electrodes, 10 kHz, 24 hrs > 80 billion data points How do we make sens of these rich, overwhelming data? Will continue to become more vital Even higher density recordings 3. Modeling Mathematical equations to mimic an observed system. Ex: single neuron activity Make it simpler Ex: binary representation New model: a statistical model of spiking A rule we use to determine whether a neuron spikes at each instant of time (e.g., a coin flip) Idea: Abandon some biological realism for model simplification. Requires balance Make it complicated. Ex: Single neuron – We want to capture the biological details
Background image of page 27

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
December 5, 2012: Introduction to Computational Neuroscience (cont.) Application of computational neuroseicne: Multi-scale seizure dynamics Data Temporal scales Spatial scales Why do seizures stop? Why do seizures fail to stop? Macroscale Data: Invasive EEG or EcoG electrodes but directly on brain multivariate, high density 100 electrodes sampling 500 Hz purpose: localize seizure focus standard analysis: visual inspection How to characterize these data? EcoG Data: visual inspection onset: “increased activity” cessation: “no activity” Why do seizures stop? Temporal scales: rhythms one electrodes beginning: fast rhythmic patterning middle: convulsions – larger amplitude, intermediate rhythmic pattern end: Ictal chirp - slow rhythmic pattern quantify: time-frequency spectra color represents amplitude Hypothesis: rhythms slow during seizurse Spatial scales: coupling plot voltage of two electrodes Are they coupled? - Related in some way? Yes – draw a line (an edge) that display similar brain activity We employ cross correlation functional network Results Networks evolving in time Data overload: Succinct measure to characterize network evolution. Example: Components – groups of electrodes connected by edges trivial components – not connected to any node red components – largest one – with the most connections yellow components – smaller one onset: one dominant component → middle: dominant component fractures → end: one dominant component Microscale Data: local field potential, LFP microelectrode array (4mm x 4mm) records voltage from small neural populations Temporal scales: LFP rhythms example: one microelectrode Ictal chirp
Background image of page 28
time-frequency spectrum Hypothesis: rhythms slow at microscale during seizures Spatial interactions: waves Suggests microscale spatial organization What's happening dynamically at the end of the seizure? ASIDE: Critical transitions
Background image of page 29

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 30
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

Page27 / 32

Neighborly with short cuts interesting properties short...

This preview shows document pages 27 - 30. Sign up to view the full document.

View Full Document Right Arrow Icon bookmark
Ask a homework question - tutors are online