000952: Improper use of generic TimeSeries type

[bendichter]

https://dandiarchive.org/dandiset/000952/0.240731.1609

These files have two neurodata objects in /acquisition, both of which are type TimeSeries. One is described as "behavioral data" and the other is described as "electrophysiological data".

1. The "behavioral data" is a time series of some sort of analog signal, but there is no description that would allow us to interpret it in the context of the task.

https://neurosift.app/?p=/nwb&url=https://api.dandiarchive.org/api/assets/0faa4f25-81fa-4b29-ae67-a7d21577043a/download/&dandisetId=000952&dandisetVersion=0.240731.1609&tab=neurodata-item:/acquisition/R125_BrainER_raspberrypi-102|TimeSeries&tab-time=1862.888772800483,2078.1895050721146,undefined

My guess is this signal should be parsed into events in a Trials table but it's not 100% clear.

2. The LFP signal should use the ElectricalSeries data type. It is currently using the TimeSeries type. The ElectricalSeries type will provide places to describe the electrodes and locations of those electrodes. The Dandiset descriptions states that the data was recorded using OpenEphys, so this should be pretty easy to convert properly using NWB GUIDE.

3. There's lots of weird stuff going on in the LFP signal itself:

https://neurosift.app/?p=/nwb&url=https://api.dandiarchive.org/api/assets/231cf19c-e0ca-4421-85a9-15d53230a548/download/&dandisetId=000952&dandisetVersion=0.240731.1609&tab=neurodata-item:/acquisition/R161_LFP_neatlabs-HP-Z2-Mini-G3-Workstation|TimeSeries&tab-time=8848.551241054083,8849.306390130074,8848.928815592079

The description does not describe any sort of stimulation that would explain these artifacts.