A nonhuman animal shows spontaneous rhythmic entrainment to a musical beat

1
A nonhuman animal shows spontaneous rhythmic
entrainment to a musical beat

• A.D. Patel1, J.R. Iversen1, M.R. Bregman1,2,
  I. Schulz3, C. Schulz3
  

1The Neurosciences Institute, 2UCSD,
3Birdlovers Only Rescue Service,
Inc.

The Neurosciences and Music III, June 25-28,
2008, Montreal
2
Introduction

• Beat perception and synchronization (BPS) is a
  fundamental aspect of music perception
  
• A true musical universal
• An important topic in research on music cognition
  (e.g., Desain Honing, 1999)
  
• Not commonly observed in other species
• Is synchronization to a musical beat uniquely
  human?
  
• Relevant to evolutionary theories of music
• If other species are capable of BPS, then it is
  unlikely to be an adaptation for music-making

3
What kinds of brains can sync to a beat?

• BPS involves tight integration between the
  auditory and motor system
  
• So does vocal learning (rare in animal kingdom).
• Humans are unique among primates in having
  complex vocal learning (Egnor Hauser, 2004)
  
• Possible overlap in neural circuitry for vocal
  learning and BPS (see Patel, 2006,
  2008 for details)
  
• Vocal learning and rhythmic synchronization
  hypothesis
  
• Only vocal-learning species are capable of BPS
  (Patel, 2006)
  
• Predicts that our closest genetic relatives cant
  acquire BPS

4
Differences between BPS and synchronized animal
displays

• Many species produce periodic signals in
  synchrony with conspecifics (Greenfield, 2005)
  
• e.g., Frogs, crickets, fireflies
• BPS differs from this in several ways
• Complex rhythmic stimuli
• Range of tempi
• Cross-modal

5
Apparent BPS in a nonhuman animal

• 2007 YouTube video a cockatoo (Cacatua galerita
  eleanora), dancing to rock music
  
• Often seemed synchronized to the beat
• Getting timing cues from humans? Can adjust to
  different tempi?
  

Snowball a 12 year-old male Sulphur-crested cock
atoo
6
Experimental study - overview

• Present Snowballs favorite song at a range of
  tempi and quantify his dancing from video
  
• Five video sessions between January and May 2008,
  at Snowballs home (Indiana, USA)
  
• No humans dancing off-camera. In sessions 4 and
  5, all rhythmic movements to music by humans were
  supressed (e.g., head bobbing)
  

7
Methods 1 stimuli, procedure

• 1 minute, 18 second excerpt of Everybody (Back
  Street Boys) rock
  
• 4/4 time, 108.7 beats per minute (BPM)
• Tempo manipulated using Audacity software
• Original, /- 2.5, 5, 10, 15, and 20
• Snowball placed on back of an armchair (preferred
  dancing spot)
  
• Tempi generally presented in slow-to-fast order
• Verbal encouragement, no food reward

8
Methods 2 video, audio analysis

• Video taken at 60 frames per second
• 30 fps in session 2
• Rhythmic movements coded from video
• Sound off, with coder blind to condition
• Focused on up-down head bobs
• Musical beat times obtained using a beat tracking
  algorithm (Ellis, 2007)
  
• Checked by JRI, an experienced drummer

9
Methods 3 measuring movements

• Each head bob assigned a relative phase with
  respect to the nearest beat, e.g.
  
• On the beat 0 degrees
• 25 of one beat period ahead of the beat -90
  deg.
  
• 25 of one beat period after the beat 90 deg.
• Produced a sequence of phase vectors for each
  trial
  

Example time series Red head bob times Blac
k musical beat times (105 BPM)
10
Methods 4 testing for sync

• At the trial level Rayleigh test on phase
  vectors
  
• Test is sensitive to consistency of phase angle
  and distance of angle from 0 (perfect synchrony)
  
• Hence concerned with both period and phase
  (Fisher, 1983)
  
• Since Snowball exhibited periods of synchrony
  surrounded by much non-synchronized dancing, we
  also searched for synchronized bouts in each
  trial
• At least 12 successive head bobs synchronized to
  the beat, using the Rayleigh test on windows of 8
  beats
  

11
Results 1

• Overall Rayleigh test 9/40 (20) trials showed
  significant sync (all psynchronized bouts in these trials are shown
  below

Sessions 4,5
Sessions 1-3
Tables from Patel, Iversen, Bregman, Schulz
Schulz (2008, ICMPC10)
12
Results 2 Tempo sensitivity

• During synchronized bouts, Snowballs dance tempo
  closely matches the music tempo
  
• But Snowball danced at a number of different
  tempi on each trial

R2 .95, p
13
Results 3a A closer look within trials
Snowball showed substantial tempo drift
Snowball
Music tempo (105 BPM)
14
Results 3b A closer look within trials
Snowball showed substantial tempo drift
Snowball
Music tempo (130 BPM)
15
Results 4 Phase histograms
During sync bouts, Snowball was well entrained
105 BPM, entire trial
130 BPM, entire trial
-
-
105 BPM, sync bouts
130 BPM, sync bouts
-
-
16
Could it have happened by chance?

• Is Snowball simply dancing at his own preferred
  tempo in response to the music, plus some tempo
  drift?
  
• Perhaps music makes him dance, but no true sync
• Are we simply picking out times that (by chance)
  his own movements happen to be in sync with the
  beat?
  

17
Checking on chance - 1

• We used a phase-agnostic Rayleigh test to
  identify the different tempi at which Snowball
  danced on each trial
  
• These are the peaks above dotted lines in plots
  below. Dotted lines p.01

105 BPM, entire trial
130 BPM, entire trial
-
-
Music tempo
Music tempo
18
Checking on chance - 2

• We determined the slowest and fastest tempi on
  each trial at which Snowball showed rhythmic
  dancing, from plots above (p
• This tempo range showed a relationship to musical
  tempo
  
• Not expected if Snowball is simply dancing at his
  own tempo ( drift) in response to the music.

19
Checking on chance - 3
Green high end of range, blue low end of range
Snowball
Musical tempo (40 trials)
20
Checking on chance 4 - Monte
Carlo test

• Randomly scramble the values representing the
  bottom end of Snowballs tempo range (x 10,000).
  
• Compare slope of resulting line to slope we
  actually observe
  
• Our observed data occurs with p
• Evidence for true tempo sensitivity

1
21
Checking on chance - summary

• Snowball shows true tempo sensitivity in his
  dancing, though he dances at a range of tempi on
  any given trial in response to music
  
• During 20 of trials, he shows periods of
  sustained entrainment to a musical beat
  
• We believe this demonstrates BPS in a nonhuman
  animal

22
Discussion - 1

• Experimental evidence that a nonhuman animal can
  synchronize to a musical beat
  
• Snowball is not unique
• See Schachner et al., (Neurosciences and Music
  III, 2008) for data from other vocal-learning
  species
  
• Supports the vocal learning and rhythmic
  synchronization hypothesis
  
• Degree of sync is reminiscent of young children,
  not adults (cf. Eerola et al., 2006)
  

23
Discussion - 2

• Issues for future work
• What range of music (styles and tempi) can
  Snowball sync to?
  
• Do his different rhythmic movements (e.g. head
  bobs, side-to-side head movements, foot lifting)
  mark out different levels in the metrical
  hierarchy?
• What is the relationship of his dance movements
  to natural display movements of cockatoos?
  
• What influence do social cues have on his
  dancing?
  
• Possible practical implications
• An animal model for sync to a beat could be
  relevant for understanding how rhythmic music
  helps people with Parkinsons disease (Sacks,
  2007)

24
References
Desain, P., Honing, H. (1999). Computational
models of beat induction The rule-based
approach. Journal of New Music Research,
282942. Eerola, T., Luck, G., Toiviainen, P
. (2006). An investigation of pre-schoolers
corporeal synchronization with music. In M.
Baroni, A. R. Addessi, R. Caterina, M. Costa,
Proceedings of the 9th International Conference
on Music Perception and Cognition (ICMPC9),
Bologna/Italy, pp. 472476. Ellis, D. (2007). B
eat tracking by dynamic programming. J. New Music
Research, 36, 51-60. Egnor, S. E. R., Hauser,
M. D. (2004). A paradox in the evolution of
primate vocal learning. Trends in Neurosciences,
27, 649654. Fisher, N. I. (1983). Statistical
analysis of circular data. Cambridge Cambridge
University Press. Greenfield, M. D. (2005). Mec
hanisms and evolution of communal sexual displays
in arthropods and anurans. Advances in the Study
of Behavior, 35 162. Patel, A. D. (2006). Mus
ical rhythm, linguistic rhythm, and human
evolution. Music Perception, 2499104.
Patel, A.D. (2008). Music, Language, and the Bra
in. NY Oxford Univ. Press Patel, A.D., Iverse
n, J.R., Bregman, M.B., Schulz, I., Schulz, C.
(in press). Investigating the human-specificity
of synchronization to music. In Proceedings of
the 10th International Conference on Music
Perception Cognition (ICMPC10), August 2008,
Sapporo, Japan. M. Adachi et al. (Eds.),
Adelaide Causal Productions.
Sacks, O. (2007). Musicophilia Tales of Music a
nd the Brain. New York Knopf.
Schachner, A., Brady, T. F., Pepperberg, I., H
auser, M. (2008). Spontaneous entrainment to
auditory rhythms in vocal-learning bird species.
Poster presented at The Neurosciences and Music
III, Montreal, Canada.