Contagious Yawning Spreads to Birds

Apparently, parrots aren’t just smart, they’re competitive too. A couple of months ago, we covered recent research findings on contagious yawning in animals, reporting on the rarity of the phenomenon and its potential role as a form of social mimicry or even an indication of empathy. While certain primates clearly do yawn contagiously and dogs may yawn contagiously, the behavior hadn’t been reported in other animals and had been expressly ruled out in red-footed tortoises (although the tortoises may have had the last laugh, as they won the celebrated Ig Nobel Prize for their non-yawns).

Word of our mammal-centric coverage seems to have reached the small, oval ears of the always-influential parrot lobby, though, as just last week the journal Behavioural Processes published a study describing social yawning in budgerigars (Melopsittacus undulatus), the small Australian parrot often referred to as the parakeet. This study provides the first support for contagious yawning in a non-mammal, and even ups the ante by documenting what may be the first instance of contagious stretching, another stereotyped behavior that may play a social role for certain animals. Some may say that the paper’s timing is an utter coincidence and that only someone with delusions of grandeur would believe that it was even remotely linked to the AnimalWise post. We, speaking in our usual royal manner, prefer to think otherwise.

Fascinating, simply fascinating...

Michael Miller, Andrew Gallup and other researchers from the University of the Binghamton conducted an observational study of yawning and stretching in a group of approximately 20 adult male and female budgerigars living together in an aviary as an established flock. Over a period of about a year and a half, the research team video recorded the flock on 23 separate occasions. The recording sessions, each of which lasted 90 minutes, were conducted at varying times of the day, and the researchers took a number of precautions (such as ignoring the first 15 minutes of each tape) to ensure that the flock’s behavior was as natural and undisturbed as possible. Trained reviewers then systematically reviewed all of the tapes, recording the time and occurrence of each yawn and stretch, and categorizing each stretch by whether the bird extended one or both legs.

The researchers’ hypothesis was that, if yawning and stretching were spreading contagiously among the birds, the behaviors would occur in nonrandom “clumps” – that is, rather than being evenly dispersed throughout the recording sessions, multiple yawns (or multiple stretches of the same type) would take place in closely-spaced bouts and then be followed by a long interval until a new priming behavior triggered another bout. Further, they predicted that, although there might be might be overall tendencies tied to particular times of the day (for example, the budgerigars might, on average, yawn more frequently during evening sessions), if the yawning and stretching really were being triggered contagiously, then specific clumping patterns would not repeat themselves when multiple same-time-of-day sessions were compared.

To test their hypotheses, the researchers performed detailed, session-by-session analyses of each type of behavior. For example, they tallied how frequently each behavior occurred, measured the time between adjacent stretches and yawns, and sorted the adjacent pairs into different “bins” depending on the length of the interval. They also analyzed each session for clumping by breaking it down into a large number of short (20 to 30 second) intervals, which allowed them to identify “runs” of consecutive intervals that either did, or did not, contain the behavior in question. Finally, they statistically analyzed their data in a variety of ways to identify patterns and associations.

And the results?

Both yawning and stretching behaviors were indeed clustered within trials, and the period between adjacent yawns and stretches was “strongly biased toward very short (< 20 sec) and very long (> 300 sec) intervals,” especially for the yawns. Also, as hypothesized, despite the clustering for both behaviors, “neither behavior routinely occurred at specific times from the start of a session across multiple recordings at the same time of day. This suggests that the clumping of these behaviors was due to social influences, and not to underlying physiological effects as a result of similar circadian patterns.”

The research team summarized its findings and suggested directions for future investigation as follows:

The observational results presented here suggest that yawning and stretching are at least mildly contagious in budgerigars under semi-natural flock-living conditions. In line with each behavior’s presumed physiological function, contagious yawning and stretching may ultimately coordinate mental state and a group’s collective movements, but future research needs to test these predictions.

So, kudos to the budgerigars! Parrots everywhere can take pride in these findings, which point to previously-unknown areas of avian social signaling and coordination, and which may open up new avenues for studying collective behavior.


ResearchBlogging.orgMiller, M., Gallup, A., Vogel, A., Vicario, S., & Clark, A. (2011). Evidence for contagious behaviors in budgerigars (Melopsittacus undulatus): An observational study of yawning and stretching Behavioural Processes DOI: 10.1016/j.beproc.2011.12.012.

About these ads

Zeroing In On Parrot Math Abilities

It may seem surprising, but the concept of “zero” is actually a relatively recent mathematical innovation. Indeed, the first rudimentary use of a zero-like notation didn’t appear until around 300 BC, when the Babylonians began using a special placeholder symbol to designate the absence of another value in their base-sixty number system. While revolutionary in its own right, the Babylonian null placeholder was still rather limited (for example, it couldn’t be used alone and never appeared at the end of a number), and another millennium passed before gifted Indian mathematicians and astronomers introduced a fully functional “true zero” as part of a formalized system of arithmetic operations. Some 1,500 years later, with this important mathematical foundation finally in place, Apple launched the iPhone on the AT&T wireless network.

Are there any parallels in the animal world, any similarly gifted nonhuman mathematicians that have innovated with the concept of zero?

The answer seems to be yes: Alex, the male African Grey Parrot of book and movie fame (Alex & Me), may go down in history as the parrot equivalent of Albert Einstein, revolutionizing parrot mathematics with his insight into concepts of nothingness.

How many crackers do I see? None! (image: The Alex Foundation)

It was in late 2003, early 2004 that Alex appears to have had his great breakthrough regarding the mathematical usefulness of zero-like concepts. At that time, Irene Pepperberg and Jesse Gordon of Brandeis University, who had been working with Alex over an extended period on a variety of cognitive and communicative studies, decided to conduct some experiments to explore the extent of his numerical competence.

Alex already was adept at tests requiring him to identify numbers of objects – he knew the English words for one through six, and could provide accurate verbal responses to questions about, for instance, how many green blocks were included in a mixed array of blue, red and green blocks and balls. Pepperberg and Gordon now wanted to see whether Alex really understood the numbers he was providing and could grasp the interchangeability of numerical questions.

To do so, they flipped things around: rather than asking Alex to provide the number of objects in particular groupings as he had in prior experiments, they went in the other direction by asking him to indicate which object groups were associated with a particular number. That is, they presented Alex with a tray of objects of various materials, colors and shapes (for example, six green plastic spoons, four yellow tops and three orange wooden sticks), and asked him questions such as “What color six?” and “What toy four?” Alex’s task was to look at the objects on the tray and then respond correctly (in this case, with “green” based on the six green spoons and “top” based on the four yellow tops).

(I know, this all sounds a bit like Jeopardy: “Please be sure to phrase your answer in the form of a question…”)

Perhaps not surprisingly, Alex aced the test, responding correctly to this new battery of questions over 80% of the time. More significant, though, is how Alex – apparently bored with the questioning – spontaneously extended the scope of the experiment:

On the 10th trial within the first dozen, Alex was asked “What color 3?” to a set of two, three, and six objects. He replied “five”; the questioner asked him twice more and each time he replied “five.” The questioner, not attending to the tray, finally said “OK, Alex, tell me, what color 5?” Alex immediately responded “none.”

Now, Alex had previously been trained to use the word “none” in a different context – comparing objects for similarity or difference (for example, to respond to a question about which of two identically-sized objects was bigger) – but he had never been taught to use “none” to describe a quantity that was not present. Fascinated, Pepperberg and Gordon randomly interspersed six more “none trials” into the ongoing experiment. It turned out that Alex’s response was no fluke – he gave the correct “none” response in five out of six of these trials, an accuracy rate of 83.3%.

Here’s a brief video in which Pepperberg describes the experiment and Alex’s unexpected use of the “none” concept:

Thus, it appears that Alex spontaneously used “none” in a zero-like manner to label a null set and designate an absence of objects. As the researchers summarized it, “the notion of none, even if already associated with absence of similarity and difference (and lack of size difference), is abstract and relies on violation of an expectation of presence; that Alex transferred the notion from other domains to quantity, without training or prompting by humans, was unexpected.”

While Alex’s use of “none” may not be as full and robust as the true zero concept that we use today, it nonetheless (no pun intended) is quite impressive. Moreover, Alex’s insight may prove to be quite practical, with the parrot concept of “none” providing helpful guidance as we attempt to answer some of the more pressing questions of our time, including:


ResearchBlogging.orgPepperberg, I., & Gordon, J. (2005). Number Comprehension by a Grey Parrot (Psittacus erithacus), Including a Zero-Like Concept. Journal of Comparative Psychology, 119 (2), 197-209 DOI: 10.1037/0735-7036.119.2.197.

%d bloggers like this: