Faites connaître cette offre !
Reference : UMR5169-ANTWYS-001
Workplace : TOULOUSE
Date of publication : Wednesday, September 04, 2019
Scientific Responsible name : Antoine Wystrach
Type of Contract : PhD Student contract / Thesis offer
Contract Period : 36 months
Start date of the thesis : 1 November 2019
Proportion of work : Full time
Remuneration : 2 135,00 € gross monthly
Description of the thesis topic
Navigation is one of the most crucial and challenging problems faced by animals, and increasingly, machines. When it comes to navigation in complex, realistic environments, animals are far better than machines, but how animals solve this problem is largely unknown. Solitary foraging ants provide a powerful system to investigate this question because on the one hand, they display exquisitely sophisticated navigational behaviours to find food in their environment and bring it back to their nest 1, and on the other hand, they do so with a nervous system numerically much simpler than that of vertebrates 2. We have good descriptions of their navigational behaviours, but a huge gap remains in our understanding of the underlying mechanisms. “…nothing is known about neural implementation of navigational space in insects”3. This lack of understanding results from the complete segregation between fields of neurobiology and cognitive ecology. Neurobiological studies are primarily lab-based where sensory-motor information and neural activity can be easily manipulated, but where insects are constrained in small artificial set-ups and forced to do specific, simple behaviours. This contrasts with cognitive ecology, where animals spontaneously express complex ecologically relevant behaviours as a result of how they interact with their natural environment, but where the animal's sensory information and motor output are much harder to obtain, and the manipulation of such information is practically impossible.
The aim of this PhD project is to characterise the sensory-motor mechanisms underlying ants' navigational behaviour in complex environments – and more precisely, to characterise the neural rules underlying ants' sensori-motor control during navigation. Overall, this project promises unprecedented insights into what Von Uexkull 4 called the 'Umwelt' (self-world) of an insect, that is, the way it sees the world, as a results of the filters developed by brain processing during both its evolutionary and individual histories.
This project will be highly interdisciplinary as it combines two methods:
1) An experimental approach with ants of the genus Myrmecia and Cataglyphis, both specialists in visual navigation. This approach will benefit from state-of-the-art virtual reality set-ups (already available at the lab) enabling online manipulation of the animal's perceived environment as well as recording its detailed motor response, while it is actually displaying its navigational task
2) A reverse-engineering approach, based on the simulation of ant-agent models, equipped with neural network models designed according to our knowledge of insect brain circuits, which predict how the visual information can be processed and how motor action can be driven accordingly. These 'virtual ants' will 'navigate' in the same virtual reconstructions of the ants' natural environment as the ones used in the experimental approach, so as to enable a direct comparison between reals ants and simulated agents.
Most of the project (the in silic and behavioural approaches) will be carried out within the ExPlaIn research team at the CRCA (University Paul Sabatier, Toulouse), under the supervision of Jean Marc Devaud et Antoine Wystrach, and with the help of other team members. Our laboratory possesses a new virtual reality set up for navigating ants. Part of the behavioural approach (2 months/year) will take place at ANU (Australian National University, Canberra) as a collaboration with Prof. Jochen Zeil (expert in neuroethology and insect navigation). Jochen Zeil, a current collaborator of Antoine Wystrach, possesses a virtual reality set up that has the major advantage of allowing to test ants trained in their natural environment in direct virtual reconstructions of these very environments.
The applicant should ideally have a multidisciplinary background, including behavioural experimentation with ants and strong modelling experience.
We talk about it on Twitter!