![OSB archive](https://www.ox.ac.uk/sites/files/oxford/styles/ow_medium_feature/s3/field/field_image_main/Archive%20OxSciBlog%20bann%20blue4a.jpg?itok=IZIIc-kU)
Passenger pigeons help to navigate
Many animals travel long distances in groups but little is known about how this may influence the navigational skills of individuals.
To test if travelling with others who know the way affects a bird's path-finding abilities a team from Oxford University, UCL, and Microsoft Research Cambridge, studied homing pigeons. They paired up experienced and less experienced - 'passenger' - pigeons on repeated flights and then recorded how well the birds navigated on their own.
I asked Benjamin Pettit of Oxford University's Department of Zoology, one of the authors of a report of the research in Proceedings of the Royal Society B, all about passengers, pigeons, and learning in a flock…
OxSciBlog: How might some animals be 'passengers' & others 'drivers'?
Benjamin Pettit: Animals that live in groups will often use each others' behaviour as a source of information - about food or predators, for example. This also applies to navigation.
In a travelling group, such as a migrating flock of birds, there will often be differences in experience, especially if animals of different ages have travelled the route a different number of times. Simulations of flocks suggest that only a minority need to know the way to guide the flock to its destination, so it is possible that only some of a flock is navigating, and the rest follow.
This raises the question of whether the 'passengers' in the group learn to navigate for themselves. Simply travelling with others who know the way already, like passengers in a car, could inhibit an individual’s route learning, making it harder to travel alone in the future. This has been called a 'passenger/driver' effect.
OSB: How did you examine the passenger/driver effect in pigeons?
BP: We track homing pigeons with mini GPS loggers. Over the past decade scientists have discovered a great deal about how a pigeon learns a route as it becomes more familiar with an area of the landscape.
In this experiment we compared route learning in two conditions - some pigeons flew alone, whereas others flew together with a trained 'demonstrator' pigeon, which had already learned a route home from that release site.
After a sequence of 12 flights, we tested the birds on their own. If being a 'passenger' interferes with learning, we would expect the birds trained in pairs to have more erratic routes when they then fly without the 'demonstrator'.
OSB: What does your study tells us about how pigeons share/gain information about a route?
BP: In fact, the birds trained in pairs learned homing routes just as well as if they had flown alone. This shows that a pigeon continues to pay attention to the landscape even if it has another pigeon to follow. The 'demonstrator' pigeons improved their homing routes as well, which was surprising.
In previous experiments, pigeons with this amount of experience settled into a particular homing route and rarely changed it. So rather than a homing route being strictly transferred from one bird to another, the pairs' routes ended up including new shortcuts that may have been discovered by the less experienced bird.
OSB: How might less experienced birds navigating benefit a flock?
BP: Birds can use a number of different cues to navigate over unfamiliar terrain, including geomagnetism, smells carried by the wind, and the position of the sun (or stars). So a bird over unfamiliar terrain is still likely to have some information to add to the flock's route choice. Theoretically, a flock can improve its navigational accuracy by combining information from as many birds as possible.
OSB: What further work is needed to examine route learning behaviour?
BP: Similar learning processes could be at work in flocks of migratory birds that travel in mixed-experience groups. This can be investigated through long-term tracking studies of wild birds, both species that migrate alone and those that form flocks. As for the pigeons, we already have plans to test how learning plays out in larger flocks. In particular, we will test whether a follower learns as quickly as leader, within a large flock.