Page 1 of 1

design afferent stimulus events with phasic bursts

Posted: Sun Jan 12, 2020 8:20 pm
by breakwave922
I have a big network model with over ten thousand cells. I want to deliver a stimulus to each cell with discrete events.
Those events to each cell should present phasic burst style, with decreasing event interval among bursts (intra-burst), i.e. increasing event frequency.
For example, if I want to have ten burst events with 1000ms duration, the first burst should have 1000ms intra-burst intervals, then the second one is 900ms, up to the last one with 100ms; The interval between each burst (inter-burst) is 1000ms. Each cell receive the same style of burst, but may have different randnomness in event noise, so in this way, the network will not receive the exact same event for all cells at the same time, to avoid high synchronization.

I couldn't figure this out for days on how to design this efficiently with less burdens on computing resources. If for each burst with specific intra-burst freq, I design a netstim() for it, that would be ending up with too many netstim()...

Any help would appreciate!
Thanks in advance.

Re: design afferent stimulus events with phasic bursts

Posted: Tue Jan 14, 2020 11:45 am
by ted
If for each burst with specific intra-burst freq, I design a netstim() for it, that would be ending up with too many netstim()
Really? What operating system are you using, and how much memory does it have? Assuming your code allows you to specify how many cells the network has, what is the largest model that you can run?

Re: design afferent stimulus events with phasic bursts

Posted: Tue Jan 14, 2020 1:29 pm
by breakwave922
ted wrote:
Tue Jan 14, 2020 11:45 am
If for each burst with specific intra-burst freq, I design a netstim() for it, that would be ending up with too many netstim()
Really? What operating system are you using, and how much memory does it have? Assuming your code allows you to specify how many cells the network has, what is the largest model that you can run?
Dear Ted,

Thanks for the reply. My network has 15,000 cells. What I was thinking was, for example, if I need to deliver 100 bursts with different ISI for each cell, that would end up with 15000*100 netstim() additionally to the network, isn't it?

Re: design afferent stimulus events with phasic bursts

Posted: Tue Jan 14, 2020 9:17 pm
by ted
. . . if I need to deliver . . .
So? Have you tried to do it? How do you know there is a problem if you haven't tried it?