Pages

Apr 18, 2017

Do you see what I see?


As some of you know, Cat Hobaiter and I are putting together an online experiment to test human understanding of great ape gestures. This means pulling together video examples of all of our gestures, but we need to make sure that the gestures will be visible for everyone. Cat and I (and many of our colleagues) are used to looking past branches, with shaky cameras and digitally brightened videos. We are going to pick the clearest videos possible for the experiment, but I'm curious - how well can you spot a gesture?

Does it matter if the camera is shaky?



Does it matter if the angle is from underneath?


Does it matter if branches are in the way?
(this clip actually has two young bonobos gesturing)


Do the branches matter less if the movement is bigger?


Please give me feedback in the blog comments or on Twitter:
Were any of those gestures impossible to spot? Which condition was the hardest? After I hyped up the viewing difficulty at the top of this post, was it easier or harder than you expected to spot the gesture?

We are aiming to have the full experiment online by July, so WATCH THIS SPACE. We are also creating a website to host video examples of each gesture type for both bonobos and chimpanzees, so that you can finally actually see what the different gestures look like. 

5 comments:

  1. I feel like I need some training in what you're looking for. For example, is video (2) a directed scratch? If so, I can see that fine, but not the supposed response. And which part of (4) do you think contains the candidate gesture?

    For (1) I wasn't sure that the leftmost bonobo wasn't just taking food from the central one's mouth.

    The infant's begging gesture in (3) was nice but it was the only one where I was sure I knew what I was looking at.

    ReplyDelete
  2. I see gestures in all of these, but then I am no expert :) The camera work on clip one is very avant-garde ;)

    ReplyDelete
  3. 1. Shaky videos could be tricky; although the gesture was visible in this particular one, there could be situations where gesturing is more and the camera shake will be distracting.
    2. The angled video is problemmatic for me. Again, in this video I could see one gesture. But I will always have insecurity of missing out. It will also depend on the extent of being 'angled' though.

    ReplyDelete
  4. Kristin HavercampApr 21, 2017, 3:07:00 PM

    Hi Kirsty (and Cat)! I'm really looking forward to the online experiment, and also the video database of gesture types. It will be especially helpful for young (but also "seasoned") researchers, especially those studying chimpanzee and/or bonobo gestural communication, and will be a nice supplement to Cat and Richard's published lexicon. I have been analysing chimpanzee behaviour recorded by camera traps (living in non-habituated communities and thus, not yet or well studied/observed) for over a year, and quite often see gestures where I wonder if it might be a widespread behaviour or instead possibly culturally unique to that particular community or population. Having a video database to refer to will be really helpful in exploring this, however depending on the size/scope of the project of course.

    I agree with Richard in that the shaky video example (1) looked more like the bonobo on the left was just trying to take food from the other's mouth, or rub the mouth area (it looks like a food item falls to the ground at the start of the clip, does the other individual drop it?). This clip was probably the most difficult for me to focus on due to the big shakes of the camera.

    In my opinion, the view from underneath the individuals is fine, at least better than the shaky camera, and looking past branches is even better (however, I might just have a lot of experience doing this and so be more comfortable with it?).

    For the experiment, I would suggest beginning with a few example/practice clips (where the gesture is not as easily visible, especially one through branches which is likely to be frequent) which could then lead to the active videos/questions. This way the participants have at least a small amount of experience exploring the clips where individuals and/or their movements are more challenging to see, and it also might provide a little confidence boost, especially for those who have never seen a primate gesture before!

    ReplyDelete
    Replies
    1. Thanks so much for all the feedback! This is all going to be so helpful for selecting videos and setting up the experiment :-)

      Delete