Radio frequency identification and tracking of individual ants engaged in colony scale division of labour

Lead Research Organisation: University of Bristol
Department Name: Biological Sciences

Abstract

Radio Frequency IDentification (RFID) is an automatic identification method, allowing remote retrieval of identification codes from devices called RFID tags or transponders. An RFID tag is a small object, requiring no internal power source that can be attached to or incorporated into a product, animal, or person. These tags contain antennae that respond to prompts from an RFID transceiver by emitting a radio signal that codes for a unique ID. Recent progress in RFID technology has led to the miniaturization of such tags to sub-millimetre dimensions. These microtransponders, once activated by a well localised beam of light, emit a radio frequency identification code. The miniaturization of these tags, as well as their ultra light weight and low price, now permit the use of RFID for insect identification. To make the most of these new possibilities, we will use the RFID system to investigate social insects.The behaviour of ant colonies is extremely complex and includes collective phenomena like nest building, decision making and partitioning of the work force (division of labour). This colony scale activity is not the consequence of some central control but rather emerges from the actions of individual ants as each of them reacts to her immediate environment. The mechanisms in which diverse colony scale behaviours emerge from single ant actions have been extensively studied. We will develop a novel experimental setup incorporating RFID technology that will enable us continuously to track a large number of identified ants as they collaborate. This novel experimental tool will provide simultaneous experimental access at both the colony level and the level of individual ants. This will facilitate new insights and a deeper understanding of the connection between individual and collective behaviour in social insects.We intend to focus on the problem of 'division of labour', namely how an ant colony divides its work force among the different required tasks in response to changing external and internal conditions. The system we propose would allow the tracking, over time, of the ants engaged in different tasks as well as the ants that switch between tasks to maintain colony plasticity. In a second stage of the experiments, we will employ automatic computer controlled doors that would enable us, for example, to prevent specific ants from switching tasks. We will use these novel tools to investigate the role of different individuals and their importance to colony performance. We will focus our study on the importance of phenotypic diversity and ant individuality for division of labour. Would a hypothetical colony of perfectly identical ants perform as well as a colony composed of unique individuals? The role of individuality in division of labour is likely to be of key importance in other social insect phenomena as well as having implications for emerging technologies such as collective robotics and sensor networks.

Publications

10 25 50
publication icon
Franklin E (2010) Blinkered teaching: tandem running by visually impaired ants in Behavioral Ecology and Sociobiology

publication icon
Franks N (2010) Ant search strategies after interrupted tandem runs in Journal of Experimental Biology

publication icon
Richardson TO (2010) Record dynamics in ants. in PloS one

publication icon
Robinson E (2012) Experience, corpulence and decision making in ant foraging in Journal of Experimental Biology

publication icon
Robinson EJ (2009) Flexible task allocation and the organization of work in ants. in Proceedings. Biological sciences

publication icon
Robinson EJ (2009) Do ants make direct comparisons? in Proceedings. Biological sciences

publication icon
Robinson EJ (2014) How collective comparisons emerge without individual comparisons of the options. in Proceedings. Biological sciences

 
Description The behaviour of ant colonies is extremely complex and includes, collective decision making and partitioning of the work force (division of labour) which emerge without centralised control from the actions of individual ants as each reacts to her immediate environment. We developed a novel approach to the study of this system by using Radio Frequency Identification (RFID) technology. This is an automatic identification method, allowing remote retrieval of identification codes from tiny RFID tags that require no internal power source, and which we attached to ants for the first time. This novel experimental tool provided us with simultaneous experimental access at both the colony level and the level of individual ants, allowing us to explore the connection between individual and collective behaviour in social insects.



We focussed on the organisation and division of labour, specifically how an ant colony divides its work force among the different required tasks in response to changing external and internal conditions, and what rules do individual ants follow to lead to collectively cohesive behaviours.



One of the most clear-cut divisions of labour among worker ants is between those working inside and outside the nest. The RFID technology allowed us to link an individual's experience of extra-nest tasks, the current level of activity in the colony and the individuals' fat stores to determine what factors best predicted which ants engaged in extra-nest tasks. Our results indicated that social factors (current activity) and physiology (fat stores) were more important than experience, and that in the absence of social factors, physiology was a good predictor of which ants left the nest to forage. To develop this further, we carried out a second study taking even more factors into account (fat stores, age, spatial location and previous intra-nest and extra-nest activity) and we manipulated the demand for different tasks in the colony (foragers and brood carers) to determine which individuals responded to the changing needs of the colony. This was the first study to simultaneously consider so many possible predictors of task allocation. Although experience is not a significant predictor, it does correlate with fat stores. We used automatic computer controlled doors to manipulate the amount of experience that specific ants could gain, to investigate the direction of causation. Taken together, all our data suggest a flexible task-allocation system allowing the colony to respond rapidly to changing needs, using a simple task-encounter system for generalized tasks, combined with physiologically based response thresholds for more specialized tasks. This could provide a social insect colony with a robust division of labour, flexibly allocating the workforce in response to current needs.





We also used the RFID technology to investigate collective decision-making in colony relocation. Previous work had suggested that ant decision-making was involved direct comparisons of alternatives at the individual level. Our results from individually identifying the ants involved in the scouting and decision-making process show that choice between new nests of differing qualities can emerge without requiring direct comparison by individual ants. Our results suggest ants use a simple acceptance threshold to decide whether to accept a nest or to continue searching. The publication of these results attracted international media attention (e.g. Science News, NPR, Reuters, BBC Today programme). We investigated, tested and developed this new model of collective-decision making with further experiments, simulations and analytical modelling. We were also able to use the automatic computer controlled doors to manipulate the information available to individual ants, by preventing any ants which had entered one nest from entering the other nest. This eliminated direct comparisons by individuals, while still allowing comparison at the colony level. This work has demonstrated that a very simple individual level rule can lead to a robust collective decision process.





The RFID technology in this project has allowed us to generate unprecedented detailed individual ant behavioural data, and has also made possible behavioural manipulation at the individual level. Task-switching and collective decision-making are essential processes in maintaining colony robustness. Our new insights into the simple rules governing collective decision-making and flexibility in division of labour will be valuable to those studying the relationship between individual and emergent group properties, both in animal behaviour, and in emerging technologies such as collective robotics and sensor networks.
Sectors Communities and Social Services/Policy