Coordination
Katarzyna Wojczulanis-Jakubas
14 December, 2021
Introduction
Why do we care about coordination?
Because coordination is a pattern, and a pattern means that things on the table are related to each other in not a random manner. If that is not random, perhaps there is something (a rule) behind. Discovering that rule we can understand the things (e.g. behaviour) better.
Before we jump into detailed examples of coordination in animal behavious, let’s first consider the issue (coordination) from a bit different perspective, so those who are more for maths could find the connection with animal behaviour.
If you take 100 random numbers (sampled from a standard normal distribution) for x and another 100 random numbers for y, and you plot x in respect to y you will see something like this:
Sampling from a standard normal distribution is sampling numbers randomly, within a set of numbers that are unimodally and symmetrically distributed with a bell-shaped curve, and with the mean and standard deviation fixed at 0 and 1 value, respectively.
x <- rnorm(100)
y <- rnorm(100)
plot.new()
plot.window(xlim = c(-1, 1), ylim = c(-1, 1), asp = 1)
lines(x, y, col = hsv(0.65, 1, 1))
lines(0.8 * x, 0.8 * y, col = hsv(0.8, 1, 1))
lines(0.6 * x, 0.6 * y, col = hsv(0.9, 1, 1))
lines(0.4 * x, 0.4 * y, col = hsv(0.95, 1, 1)) 
There is no apparent pattern. Even if you found it interesting, because you have an artistic soul, still there is nothing you can conclude/predict from such a plot, the relationship between x and y is just random.
Then, let’s make a small difference. If you take 100 random numbers again but now, for creating x and y, you apply sinusous and cosinous to each value, respectively, and you plot all that the same way as you have just done, you will see something like this:
random_numers <- rnorm(100)
x <- sin(random_numers)
y <- cos(random_numers)
plot.new()
plot.window(xlim = c(-1, 1), ylim = c(-1, 1), asp = 1)
lines(x, y, col = hsv(0.65, 1, 1))
lines(0.8 * x, 0.8 * y, col = hsv(0.8, 1, 1))
lines(0.6 * x, 0.6 * y, col = hsv(0.9, 1, 1))
lines(0.4 * x, 0.4 * y, col = hsv(0.95, 1, 1)) 
Now, we can clearly see there is something in it. Since, we have just created that, we know what it is. The rule here is simple (well, trigonometric functions are not that simple but just whole procedure is quite strightfoward) - we take some random numbers and calculate sine and cosine values of them, and we plot it.
Before we would go further, to make connection with animal behaviour (!), let’s do one more thing here - just to mesmerize you by the simplicity of the rule. If we take not random numbers but just 100 consecutive numbers, let’s say from 1-100, and perform sine, cosine and plotting, then the plot will look even more…
ordered_numers <- seq(1, 100,1)
x <- sin(ordered_numers)
y <- cos(ordered_numers)
plot.new()
plot.window(xlim = c(-1, 1), ylim = c(-1, 1), asp = 1)
lines(x, y, col = hsv(0.65, 1, 1))
lines(0.8 * x, 0.8 * y, col = hsv(0.8, 1, 1))
lines(0.6 * x, 0.6 * y, col = hsv(0.9, 1, 1))
lines(0.4 * x, 0.4 * y, col = hsv(0.95, 1, 1))
legend("bottomright", legend = "© Gaston Sanchez", bty = "n", text.col = "gray70")
Well…, things are simply coordinated here, right?!
So, when we see some kind of repetitivity in animals performance, we can expect there is an algorithm we could use to describe the behaviour, to model it, and maybe extrapolate its relevance to other systems (including human stuff).
Vacancy chain
A vacancy chain is a social structure through which resources are distributed to consumers. In a vacancy chain, a new resource unit that arrives into a population is taken by the first individual in line, who then leaves their old unit behind, and this old unit is taken by a second individual being in the line, leaving their old unit behind, and so forth.
Vacancy chain functions as a method of resource distribution in human domains such as housing, hospitaling, and labor markets. These are important issues in human societies. Interestingly, vacancy chains existing in natural world serve as a model systems. For example, hermit crabs and their shell exchange (Chase 1991).
Hermit crabs employ vacancy chains as a method of shell exchange (see the video). Most often this chain is performed in a coordinated manner. What is more, the coordination has its advantages (though some disadvantages also exist, see the table below).
CONCLUSION: Coordination of egg shells exchange may be an optimal behavioural strategy in given ecological context. Studying coordinated performance in animals we can model processes in human societies.
Coordinated parental care
Bi-parental care in birds has long been viewed as a tug-of-war between pair members and this is because the costs of parental care in terms of reduced parent survival or fecundity are assumed to be high (but see Wojczulanis-Jakubas 2021). The key point around this sexual conflict over parental care is that mates share the benefits but not the costs of their partner’s effort (Hinde and Kilner 2007; Johnstone et al. 2014). However, when both mates might be adversely affected by over-investment by the partner, the conflict (if still present) is not that apparent. This is the case in many socially monogamous species, where parents stay together for multiple breeding seasons, sharing both benefits and costs of their partner’s workload (Sánchez-Macouzet et al. 2014)(Griffith 2019). For such groups, the parents relation may be considered as a ‘family firm’, where both partners—the firm owners—work together for multiple seasons to maximize a common product (the off- spring) (Roughgarden and Akçay 2010; Roughgarden 2012). As such, parents may achieve a more efficient breeding outcome by coordinating their reproductive effort (Griffith 2019).
Little auk is an excellent model species for studying coordination of parental care (Wojczulanis et al. 2021). It is a small long-lived seabird, with long-term pair bonds, and long and extensive bi-parental care (so the costs of parental care are supposed to be high) and partners should benefit of avoiding over-exploitation of the partner. Importantly, the little auk exhibits a bimodal foraging strategy, during the chick rearing period, meaning the parents perform long foraging trips that primarily serve adult self-maintenance (some food is also brought to the chick), alternated with a serie of short trips that serve solely to provision the offspring. From the offspring’s perspective, the adults’ long trips represent extended periods of waiting for food whereas the short ones represent a great amount of food over a short period of time. Both situations may be disadvantageous for the offspring. During long trips, the risk of offspring starvation increases considerably, whereas during short trips the young birds may face a difficulty assimilating large amounts of food.
This may be particularly profound when both parents make their long and short trips simultaneously. This may then affect the chick’s body condition, leading to starvation (if the fasting period is too long) or obesity. In such a context, one may expect a coordinated performance of pair members by avoiding an overlap of long trips (Wojczulanis-Jakubas et al. 2018) and all the references there).
The question one can ask here is: do the little auk parents coordinate chick provisioning (i.e. avoid overlapping their long trips)?
To answer such a question one first needs to establish a parameter that is going to be a proxy of the parental coordination. In this case, it will be the range with which the long foraging trips of the parents overlap. Then, one needs to calculate the value of the parameter for each breeding pair (nest), and so for the whole population (i.e. group of studied pairs). Then, to know if the parameter is somehow significant, i.e. different from that what could be expected by chance, one needs a reference. To get the reference, the real data are shuffled (i.e. randomized), and the parameter in question is calculated based on these shuffled data. If the randomization procedure is repeated over 1000 times, one will get 1000 parameters of the randomized data, and of these 1000 represenation of the parameter makes up its distribution, when it is created simply by chance.
Based on data from Wojczulanis-Jakubas et al 2018, prepared by Marcelo Araya-Salas
With such a distribution of the parameter calculated based on randomized data, one can interpret the observed value of the parameter, and notice that the observed value is different from that what we could get simply by chance.
CONCLUSION: Given all above, we can conclude that little auk parents perform their foraging trips in apparently coordinated manner, actively avoiding to overlap in time their long foraging trips. What is more, this seems to be adaptive. The parental coordination minimizes the variation in duration of time intervals between the feedings, i.e. chick gets the food in regular time intervals.
Obviously the pattern of coordinated provisioning, to be adaptive in a species, it will be species-specific. For the little auk, apparently the most adventegaous is when parents avoid overlappling their long foraging trips, so they feed the young visiting the nest independently, and in an alternated manner. For some passerines species, simultanous visits of both parents will be the most profitable strategy, and this is because this way they minimize the risk of nest predation (what apparently is more important in their ecological/environmental context).
Importantly, regardless of the pattern (alternated/simultaonus) of the feedings, their non-random nature indicates on parental coordination, and that in turns is important for understanding the evolution of parental care.
Coordinated sining
To discuss another aspect of coordination, let me first introduce you very interesting system of singing hummingbirds.
When thinking about hummingbirds, the most frequent trait we refer to is their tiny size and beautiful colours, and these are very true.
We usually do not really refer to singing as most of the species do not really sing. There are some species (Phethornis sp.), however, which for singing is an important behaviour (notabene, during the courtship display).
What is more Phethornis sp are lekking species, meaning that males gather at given location, and there is where they sing. The leks are not very crowdy, there are 5-10 males but each individual at the lek has usually two neighbours, with which it does not “want” to interfere with an acoustic signal (even if all males together may attract female to the lek, at the very end of the female visit only one male will benefit, so when singing males actually compete with each other). How the problem is solved? Well, if that chapter was not about coordination, the answer could not be that straightforward.
Copyright: Marcelo Araya-Salas.
So, when you enter a lek you can clearly hear all the males singing at the same time, and if you listen to carefully, you will hear how birds are struggling during this performance.
Sometimes you may hear them coordinating their singing, i.e. synchronizing the signal or avoiding the overlap. So, the first question here that we could ask about such a behaviour is: could such a coordination be a coincidence? If not, how it is actually achieved? How to test whether the observed pattern is different from that what could be expected by chance, you could learn from the little auk story (see above). This kind of method (randomization) applied to long billed hermit Phaetornis longirostris show that both alternation and singing overlap are not random (Salas-Araya et al 2017).
To learn about mechanisms of the singing coordination, the duration of the time intervals between consecutive signals of singing individuals were measured, when they were singing solo, and with a company of their neighbours. The results are pretty obvious. To get the coordination sensu alternate the songs, the guys extend the duration of the time breaks between the syllables (see below, Salas-Araya et al 2017.
What is even more interesting is that the coordinted singing pattern varies depending on the distance between singers: the birds alternate their songs when they are in close proximity, and they overlap signals when being at farther distances in respect to each other.
CONCLUSION: When using signals to attract mates or defend resources, animals often overlap the voices of other individuals in close proximity. In such contexts signal masking is likely and animals would benefit by adopting behavioural strategies that modify the timing of signals to minimize the negative effects of masking or take advantage of its signalling value. The example of long-billed hermints demonstrates that vocal coordination is an active process and may be adaptive.
Literature cited
Araya-Salas M, Wojczulanis-Jakubas K, Phillips EM, et al (2017) To overlap or not to overlap: context-dependent coordinated singing in lekking long-billed hermits. Anim Behav 124:. doi: 10.1016/j.anbehav.2016.12.003
Chase I (1991) Vacancy Chains. Annu Rev Sociol 17:133–154. doi: 10.1146/annurev.soc.17.1.133
Griffith SC (2019) Cooperation and coordination in socially monogamous birds: moving away from a focus on sexual conflict. Front Ecol Evol 7:1–15. doi: 10.3389/fevo.2019.00455
Hinde CA, Johnstone RA, Kilner RM (2010) Parent-offspring conflict and coadaptation. Science (80- ) 327:1373–1376. doi: 10.1126/science.1186056
Hinde CA, Kilner RM (2007) Negotiations within the family over the supply of parental care. Proc R Soc B Biol Sci 274:53–60. doi: 10.1098/rspb.2006.3692
Johnstone RA, Manica A, Fayet AL, et al (2014) Reciprocity and conditional cooperation between great tit parents. Behav Ecol 25:216–222. doi: 10.1093/beheco/art109
Rotjan RD, Chabot JR, Lewis SM (2010) Social context of shell acquisition in Coenobita clypeatus hermit crabs. Behav Ecol 21:639–646. doi: 10.1093/beheco/arq027
Roughgarden J (2012) The social selection alternative to sexual selection. Philos Trans R Soc Lond B Biol Sci 367:2294–303. doi: 10.1098/rstb.2011.0282
Roughgarden J, Akçay E (2010) Do we need a Sexual Selection 2.0? Anim Behav 79:2–6. doi: 10.1016/j.anbehav.2009.06.006
Sánchez-Macouzet O, Rodríguez C, Drummond H (2014) Better stay together: Pair bond duration increases individual fitness independent of age-related variation. Proc R Soc B Biol Sci 281:. doi: 10.1098/rspb.2013.2843
Wojczulanis-Jakubas K (2021) Being the winner is being the loser when playing a parental tug-of-war – a new framework on stability of biparental care. Front Ecol Evol 9:. doi: 10.3389/fevo.2021.763075
Wojczulanis-Jakubas K, Araya-Salas M, Jakubas D (2018) Seabird parents provision their chick in a coordinated manner. PLoS One 13:e0189969
Wojczulanis K, Dariusz J, Lech J (2021) The Little Auk Alle alle: an ecological indicator of a changing Arctic and a model organism. Polar Biol. doi: 10.1007/s00300-021-02981-7