ac-1 stimulus type and onset

Hi, I'm trying to figure out which stimulus is played, how many times, and where the onset is. My aim is to compare the parameters of a statistical model on spontaneous vs. stimulus-evoked activity during playback of naturalistic sounds.

I have trouble finding this information in the provided files. Let me take the Asari dataset

"Results/20071010-ha2-001/20071010-ha2-001-002.mat"

as an example. In there, I have this info:

{"naturalsound", {{50569.}}, {{{20235.}}, "Sequence 1: 2  1  3  1  4"}}

Following the ReadMe file, the second number, 50569, should be the stimulus (trigger) onset in samples. I don't know how the other things are interpreted, i.e. what does 20235 and the Sequence mean? How do they relate to the stimulus files of Panthers and Birds etc.?

In the same file, I also get this info:

{"natural sound fragment 1", "diversity Track 25 (Greater Sac-winged Bat) From 1.805 to 6.145 sec", "class6/1.mat", {{{5.}}, {{4340.03}}, {{97656.}}}

Ok, here it tells me which file contains the sound, and the length of that audio file corresponds to the duration given: From 1.805 to 6.145 sec. However, these times do not correspond to the stimulus onset in samples, which would be 50.569 seconds with a sampling rate of 10 kHz.

So, is the sound only played during the time between 1.805 and 6.145 sec relative to the beginning of the recording, and if yes, what does the trigger onset number mean? Also, since the recording lasts almost 40 minutes, I wonder whether the stimulus was played more than once during that time, and if not, why is the recording so long? Can I assume that the remaining time is spontaneous activity?

Thanks for your help!

Thanks for your interest in our data sets. Analysis sample commands in the Contents.m will be helpful for you to figure out the format of our data.  But here is the quick explanation.

{"naturalsound", {{50569.}}, {{{20235.}}, "Sequence 1: 2  1  3  1  4"}}

This is the information on the first sound {triggers(1)} and reads as follows: "naturalsound" was presented at the onset time of 50569 sample points (i.e., 50569/10kHz = 5.0569 sec after the recording onset) for the duration of 20235 ms (= 20.235 sec), and this 20 sec sound consists of a sequence of stimuli in the order of 2 1 3 1 4. 

{"natural sound fragment 1", "diversity Track 25 (Greater Sac-winged Bat) From 1.805 to 6.145 sec", "class6/1.mat", {{{5.}}, {{4340.03}}, {{97656.}}}

This is the information on the stimulus 1 {param.stimulus(1)} and reads as follows: this "natural sound fragment 1" corresponds to the audio CD "Track 25 (Greater Sac-winged Bat) from 1.805 to 6.145 sec" (see Contents.m for the information of this CD). The path to the stimulus file is "class6/1.mat", and it was presented for 4340.03 ms at 97656 Hz with onset and offset ramp of 5 sample points. 

Below is the information on the other stimulus fragments from {param.stimulus}:

{"natural sound fragment 2", "neotropical-mammals-1 Track 06 (White-lined Sac-winged Bat) From 38.3 to 42.465 sec", "class6/2.mat" , {5, 4165.03, 97656.} }

{"natural sound fragment 3", "neotropical-mammals-1 Track 44 (Alston''s Pygmy Mouse) From 24 to 27.99 sec", "class6/3.mat" , {5, 3990.02, 97656.} }

{"natural sound fragment 4", "neotropical-mammals-2 Track 01 (Pygmy marmoset) From 33.2 to 36.6 sec", "class6/4.mat" , {5, 3400.02, 97656.} }

You will see that the duration of the stimulus fragments are, 4165, 3990, 3400 ms, respectively. Thus the first stimulus sequence (2 1 3 1 4) lasts for 4165 + 4340 + 3990 + 4340 + 3400 = 20235 ms in total.

During the 40 minutes of the recordings, we presented 96 sounds in total in this data set as specified in the struct array "triggers". The onset of the last sound was 2332.95 sec after the recording onset as in {triggers(96)}.

Please let me know if you have further questions.

Hiro