A proposal - Video Record All Behavioral Assays and Submit with Publication

Seeing is believing. In recent months, I’ve begun recording all of my behavioral assays using either an iPhone or GoPro camera. I regret not having done this sooner.

Assays I’ve recorded include:

  • Von Frey assay
  • Pin prick
  • Dynamic Brush
  • Real-time place aversion.
  • Optogenetics

The major advantages of recording these assays are:

  • Permanent record of what actually happened at time of assay
    • Anyone who wants to can see what happened during the assay and how it was performed
    • Others can verify the results and perform secondary analyses on the data if they’d like.
  • Full capture of all behavior of animal
    • If one wants to look for a secondary or tertiary response, the video may have the answer in it. this is best used as pilot data, since these outcomes were not part of the initial design, but still, this is much better than having to redo the assay to look for other features.
  • Ability to dissociate data acquisition with assay scoring
    • It is very challenging to attend to the response and perform the assay at the same time. At best, one can attend to a single feature on time scale of seconds. With video recording, one can score multiple features with millisecond precision
  • Ability to do blind scoring
    • With video, someone else who did not perform the assay can score it, decreasing bias
  • Precision and transparent behavior scoring
    • By having video, one can use behavior coding programs like BORIS or Observer XT, one can mark specific time points in the video with scoring judgments. There is a transparent record of what judgements were made regarding a behavior. This serves the purpose of having a record that others can verify but also can be great for training purposes too. Just give your coding file to a trainee to learn from and amend if necessary.
  • Possibility of applying automated scoring software (either existing or to-be-created) on the behavior. Some behaviors have automate scoring software, usually commercial. For example, CleverSys makes one for the formalin test, I think. Even if the software you need does not yet exist, having a video record allows you to apply new tools to past experiments when they come out.
  • Excellent training resources: The view from my iPhone camera is the best seat in the house. I’ve made a lot of training material for undergrads and rotation students using the videos of me performing assays. You kill two birds with one stone. This is much better than trying to having a second or third person in the room peeping over your should. Not good for the assay and not good for them either.

There is really no reason not to do this. We all walk around with extremely capable recording devices in our pockets (our smart phones). Why not use them to do better science? Want to check some results? Look at the videos. Want to see how people really perform their assays, including those key details that sometimes get overlooked. Look at the video. I think pain research, and behavioral neuroscience in general, would be well served if we recorded our assays and shared these recordings openly. You can put your videos in a repository like figshare or similar scientific data repository when you publish (or not… even unpublished stuff can go up if you think it’ll be useful).

What do you think? Is there any downside to this proposal?

@runDRG @ShanTan @tberta @thicunha @LegakisL @liz

In terms of reproducibility, bias, etc, I think this could be great.

But do you have issues with your animal ethics board? My institution’s IACUC is pretty strict about video recording or photographing our animals. We are only really allowed to if it is specifically written into our protocol, and we have to use a designated camera (so no iphones). I think they would probably be resistant to making these videos publicly available. I think this is in place primarily to protect both the researchers and institution from misunderstanding by animal rights groups, which is especially important when you do pain research!

Logistically, how do you record? Camera/phone in one hand and von frey filament in the other? My von frey cages are painted black on the sides (so the rats can’t see each other), but I think this would make it difficult to mount a camera to record all animals at once.

Good point about IACUC but for publication I see people submit videos of behavior like optogenetics stuff. So’s it’s likely an institution specific thing.

For example, ours says:

Publication of Animal Photographs: No specific permission is required to publish approved, researcher-made photographs of animals in a peer-reviewed scientific journal.  Submission of photographs for publication in other types of written or electronic media requires prior clearance from the Office of Communication.  Photographs of abnormalities/lesions should be framed and/or cropped to focus attention on the lesion rather than the animal.  Photographers should exercise care to prevent extraneous information (e.g. cage cards, location of animal facility or laboratory, identity of animal handlers, etc.) from appearing in the photograph.

But I agree that everything must be done within the guidelines of the institutional IACUC. I use a dedicated ‘dirty’, older generation iPhone that is not my actual day-to-day one. Even if one can’t make all the video public, having it for internal purposes is useful. In terms of animal welfare, it helps you get more value from the experiments and the animals.

Logistically, how do you record? Camera/phone in one hand and von frey filament in the other? My von frey cages are painted black on the sides (so the rats can’t see each other), but I think this would make it difficult to mount a camera to record all animals at once.

Practically, I got cheap tripods for iPhone and GoPro on Amazon. $10. It works remarkably well. I move the camera each time for each animal I’m testing. The tripod sits underneath the rack (this is the one with the GoPro).

I can put it close up to the paw if I need. Slow 240 fps videos also possible.

Love this idea a lot. You find the camera good enough quality to be able to decipher feet on animals? Also is your hand in the way at all when you’re doing VF?