This week a brief update consisting of a short interview with Ani Patel (Senior Fellow at the Neuroscience Institute in San Diego, US) at a conference workshop at Indiana University-Purdue University Indianapolis (IUPUI) talking about Snowball: the dancing cockatoo that so gracefully helped boosting the visibility of research in the neuroscience and cognition of music. The other video shows Snowball (and his owner Irene Schulz) at the World Science Festival. Is Snowball listening or imitating?
However, see earlier entry on beat induction for a critique.
Honing, H., Ladinig, O., Háden, G., & Winkler, I. (2009). Is Beat Induction Innate or Learned? Annals of the New York Academy of Sciences, 1169 (1), 93-96 DOI: 10.1111/j.1749-6632.2009.04761.x
Patel, A., Iversen, J., Bregman, M., & Schulz, I. (2009). Experimental Evidence for Synchronization to a Musical Beat in a Nonhuman Animal Current Biology DOI: 10.1016/j.cub.2009.03.038
Snowball, the dancing cockatoo: Does he listen or imitate?
Summary from the responses on Facebook, responding on blog:
Roelof Reinders: I was right :)
HH: I also think he is imitating, or, to be more precise, Snowball needs his owner around to show the dancing behavior. However, Current Biology (and its reviewers) thinks of it as evidence of entrainment/beat induction in birds (see blog above).
John Bispham: wow: Am I seeing things or did a single parrot just manage to get an entire panel of scientists to dance? ;-)
Olivia Ladinig: To me it even seems that he is out of beat when she is out of beat (happens two times at the end of the performance in my perception)!
HH: N=1 !
Roelof Reinders: This reminded me of something I heard about years ago. Maybe the bird is the next Clever Hans?
HH: indeed.
John Bispham: apologies for the british sarcasm ;-). More seriously though Patel and co do claim in CB that the studies were done without human movements. Of course there has been a considerable learning phase where i imagine imitation was the primary tool and probably resulted in some learnt assocations. Also until we understand the mechanisms involved in thesynchronised sections I would agree that to call it entrainment with all that implies in relation to the human entrainment/beat induction literature is probably premature. Nevertheless this does appear to me to be good evidence of at least some form of auditory-motor loop.
Olivia Ladinig: Means only *one* cockatoo? Or only *one* person (me) thinks they are sometimes out of beat? Or only out of beat *one* time? (Did you see "Schachner et al (2009)? They report they have observed this in multiple parrots (and other species) - on Youtube vids.) I am confused, i thought science-folks where there and ensured it works even if the owner was out of the room?
HH: I reviewed Schachner in the blog link mentioned :-) Although they claim support for the'vocal mimicry hypothesis', scrutiny of the published data reveals that this ability so far is shown by parrots, but not by any species of the other animal groups that show vocal learning (songbirds; hummingbirds; cetaceans; pinnipeds; bats). So, a linkage between vocal learning and beat perception is by no means clear.
HH: Furthermore, Patel et al. (Current Biology) selected the moments from their videos - they refer to them as 'bouts' - when Snowball was 'in synchrony with the music' ! No wonder you find synchrony if you select for it :-)
John Bispham: I definitely agree with this last point. It's perfectly possible that it has learnt to associate certain movements (which of course have their natural "rhythm") with a sound source and assuming these "rhythms" vaguely match those of the source you're bound to find periods of "synchronisation" - bouts. So I guess one question to test this could be whether Snowball's movements more rhythmicaly stable during the synchronised phases than elsewhere.
HH: The latter was indeed shown by Patel et al.; See for all the references teh blog above.
Tommi Himberg: I posted this on my blog (http://bit.ly/nnc53) a while ago, expressing doubts about the vocal learning hypothesis and suggested that maybe movement imitation is the key. I got a comment from Snowball's owner (Irena Schulz, there in the video), who insisted Snowball does this on his own as well and did so already when she got him. She also invited me to come and see for myself...
I agree with Henkjan that the vocal learning hypothesis is problematic, and as we don't know how SB learned this in the first place, it is difficult to disentangle these issues. We need more snowballs.
John Bispham: many thanks. I'll take a look. warm wishes from UK.
Erin E. Hannon: Henkjan, I hope you signed your review ; ) Seems to me that even if he is imitating, he's still nearly synchronous with Irena. If it's simply imitation then wouldn't we expect a longer lag between her movements and his? If he is synchronous with her, I would argue that's still evidence of BPS. After all, a person moving around in a rhythmic fashion may be far more musical than a recording of backstreet boys...
HH: One of the problems is indeed how to decouple Perception and Action; The notion of BPS treats them as a whole. However, the newborn research suggests perception plays at least some role (in humans). In recent experiments Ani (Patel) is trying to figure out the need for Irene Schultz to be in the room during the experiment for Snowball to show his dancing behavior…
Justin London: Hey gang--I have doubts about the bouts as well. What hasn't been done is a perturbation study of snowball, just statistical analysis of simultaneous rhythmic behaviors. Absent evidence of phase or period correction, I don't see entrainment.
There are many skeptics out there that believe that Snowball is merely mimicing me. So, I invite you to the SNOWBALL CHALLENGE. If you are interested in attending, please read my blog post.
Honing, H., Ladinig, O., Háden, G., & Winkler, I. (2009). Is Beat Induction Innate or Learned? Annals of the New York Academy of Sciences, 1169 (1), 93-96 DOI: 10.1111/j.1749-6632.2009.04761.x
Very cool interview!!
ReplyDeleteAndrew
Visit my Bipolar News Site...
Snowball, the dancing cockatoo: Does he listen or imitate?
ReplyDeleteSummary from the responses on Facebook, responding on blog:
Roelof Reinders: I was right :)
HH: I also think he is imitating, or, to be more precise, Snowball needs his owner around to show the dancing behavior. However, Current Biology (and its reviewers) thinks of it as evidence of entrainment/beat induction in birds (see blog above).
John Bispham: wow: Am I seeing things or did a single parrot just manage to get an entire panel of scientists to dance? ;-)
Olivia Ladinig: To me it even seems that he is out of beat when she is out of beat (happens two times at the end of the performance in my perception)!
HH: N=1 !
Roelof Reinders: This reminded me of something I heard about years ago. Maybe the bird is the next Clever Hans?
HH: indeed.
John Bispham: apologies for the british sarcasm ;-). More seriously though Patel and co do claim in CB that the studies were done without human movements. Of course there has been a considerable learning phase where i imagine imitation was the primary tool and probably resulted in some learnt assocations. Also until we understand the mechanisms involved in thesynchronised sections I would agree that to call it entrainment with all that implies in relation to the human entrainment/beat induction literature is probably premature. Nevertheless this does appear to me to be good evidence of at least some form of auditory-motor loop.
Olivia Ladinig: Means only *one* cockatoo? Or only *one* person (me) thinks they are sometimes out of beat? Or only out of beat *one* time? (Did you see "Schachner et al (2009)? They report they have observed this in multiple parrots (and other species) - on Youtube vids.)
I am confused, i thought science-folks where there and ensured it works even if the owner was out of the room?
HH: I reviewed Schachner in the blog link mentioned :-) Although they claim support for the'vocal mimicry hypothesis', scrutiny of the published data reveals that this ability so far is shown by parrots, but not by any species of the other animal groups that show vocal learning (songbirds; hummingbirds; cetaceans; pinnipeds; bats). So, a linkage between vocal learning and beat perception is by no means clear.
HH: Furthermore, Patel et al. (Current Biology) selected the moments from their videos - they refer to them as 'bouts' - when Snowball was 'in synchrony with the music' ! No wonder you find synchrony if you select for it :-)
John Bispham: I definitely agree with this last point. It's perfectly possible that it has learnt to associate certain movements (which of course have their natural "rhythm") with a sound source and assuming these "rhythms" vaguely match those of the source you're bound to find periods of "synchronisation" - bouts. So I guess one question to test this could be whether Snowball's movements more rhythmicaly stable during the synchronised phases than elsewhere.
HH: The latter was indeed shown by Patel et al.; See for all the references teh blog above.
Tommi Himberg: I posted this on my blog (http://bit.ly/nnc53) a while ago, expressing doubts about the vocal learning hypothesis and suggested that maybe movement imitation is the key. I got a comment from Snowball's owner (Irena Schulz, there in the video), who insisted Snowball does this on his own as well and did so already when she got him. She also invited me to come and see for myself...
I agree with Henkjan that the vocal learning hypothesis is problematic, and as we don't know how SB learned this in the first place, it is difficult to disentangle these issues. We need more snowballs.
John Bispham: many thanks. I'll take a look. warm wishes from UK.
Snowball, the dancing cockatoo: Does he listen or imitate?
ReplyDeleteMore responses from Facebook:
Erin E. Hannon: Henkjan, I hope you signed your review ; ) Seems to me that even if he is imitating, he's still nearly synchronous with Irena. If it's simply imitation then wouldn't we expect a longer lag between her movements and his? If he is synchronous with her, I would argue that's still evidence of BPS. After all, a person moving around in a rhythmic fashion may be far more musical than a recording of backstreet boys...
HH: One of the problems is indeed how to decouple Perception and Action; The notion of BPS treats them as a whole. However, the newborn research suggests perception plays at least some role (in humans). In recent experiments Ani (Patel) is trying to figure out the need for Irene Schultz to be in the room during the experiment for Snowball to show his dancing behavior…
Justin London: Hey gang--I have doubts about the bouts as well. What hasn't been done is a perturbation study of snowball, just statistical analysis of simultaneous rhythmic behaviors. Absent evidence of phase or period correction, I don't see entrainment.
There are many skeptics out there that believe that Snowball is merely mimicing me. So, I invite you to the SNOWBALL CHALLENGE. If you are interested in attending, please read my blog post.
ReplyDeletehttp://birdloversonly.blogspot.com/2009/10/cockatoo-challenge.html