Today I provide you a brief summary of one of my favorite and most curious research projects I ever came across. It deals with the question 'How ants carry their stuff?'. May the amusement and enlightenment be with you!
Source: pixabay
First I've to state that collective interactions like those needed for carrying large loads require a high degree of coordination and can be observed primarily in humans and ants. You probably would not have thought that there is a skill where ants are the animals most similar to us.
It is well-known that group-living animals coordinate their actions, but the underlying mechanisms are debated. Basically, there are various suggestions, reaching from short-ranged interactions in which individual animals respond only directly to their nearest neighbor, all the way to far-reaching interaction models that involve large parts of the groups.
One reason why it is very difficult to validate the proposed mechanisms rises from a lack of possible quantitative evaluable criteria which can be taken into consideration. However, this is complete different for ants, because the collective goal of the group and thus the reason for their cooperation is well understood: Bring the food to the nest.
Ants as Research Animals
In the referenced work ants of the spezies Paratrechina longicornis were investigated. They are found in different areas around the world. Aviram Gelblum and his co-workers focused on the cooperative transport mechanism that ants use to deliver materials and food to their nest.
The Experiment:
They used ring-shaped food or rubber bands, which they placed in the vicinity of ant nests. After a discovery and the subsequent recruitment phase, the transport of the respective object took place. They filmed the ants, tracked their paths, their behavior, attachment and detachment as well as the the angle between individual ants and the object and examined these parameters in terms of the resulting movement of the transported goods.
Source: Axel Rouvin
The Findings:
They showed that the median speed increases linearly with the number of carrying ants, at least for up to 15 ants. Interestingly the speed hardly depends on the ant distribution around the cargo. The efficiency of the transport rises with decreasing number of ants, due to a reduced rotational movement, thus less waste of energy.
Another basic but essential finding is that a transport process is always facilitated by two types of ants: those that lift the object and those that pull. A significant contribution by ants exerting pushing forces is extremely rare.
For the steering they investigated a number of assumptions, including that a single attached ant directs the entire transport process. However, they did not find any correlations between the aquired data (orientations, etc.) of persistently attached ants that are statistically distinguishable from non-persistent ants. Their study also showed that the collective movement is not explained sufficiently well by a wisdom-of-the-crowds type model.
Random but cool ant pic.
Source: pixabay
The Key Results
In contrast to the carrying ants, freely moving individuals are well informed about the correct nest direction. Whenever such ants attach to the load, they steer it in a way to increase the accuracy of the transport. These ant-attaching-processes were even investigated in terms of their temporary, increase for informational input:
'Quantitatively, we found that within the several seconds that follow their attachment, ants inject about 0.5 bits of directional information into the system. [...] This causal effect implies that newly attached ants adopt an influential role. The peak in directional accuracy [...] implies that carrying ants are less informed than newly attached ants. The quick deterioration of accuracy after its initial rise shows that while newly attached ants have a transient positive influence, their directional knowledge rapidly decreases following the time of attachment.' - Ref. [1]
This observation may be explained as follows:
The ants do not strongly rely on vision, but rather on the perception via their antennas, which are severely affected by the load carried.
In summary, it can be said that the transport process relies on a squad of ants that raises and moves the good, then new ants join in and steer the whole groups motion. Of course the other attached ants react to reinforce the short-term positive impact of the new ant as best as possible. After a few rounds of detaching and attaching, the whole group of ants reaches its nest together with the carried good.
For more detailed information and very interesting graphs and illustrations I refer to the freely accessible paper of Aviram Gelblum et al.
They also provided some short clips, showing their experimental set up:
https://media.nature.com/original/nature-assets/ncomms/2015/150728/ncomms8729/extref/ncomms8729-s2.mov
'A short clip of tracked cooperative transport. Tracking includes the location (blue dot) and orientation (unmarked) of the load as well as the trajectory of each individual ant within the frame (marked by the different colors). Ants are classified as either carrying (yellow number) or free (white number).' - Ref. [1]
https://media.nature.com/original/nature-assets/ncomms/2015/150728/ncomms8729/extref/ncomms8729-s3.mov
'Relocation test. This video shows an example of relocation test along the transport trajectory. During transport towards the nest, the carried load and attached ants were lifted by tweezers, gently brought back and placed at the initial point of the transport, located less than 1 meter back along the trail. Following the relocation only 1 out of the 15 carriers left the object while the rest of the group immediately continued the transport to the original direction towards the nest. This example suggests that the random-walklike behavior presented in the clean board experiments was not a response to the gentle relocation of the ants-object conglomerate but, rather, due to lack of relevant information.' - Ref. [1]
For more details I warmly recommend to go through Ref. [1], but you can also ask any questions in the comments!
, who recently wrote a nice article, which reminded me on the just presented, awesome scientific research!mountain.phil28
References:
- A. Gelblum, et al. Ant groups optimally amplify the effect of transiently informed individuals, Nature Communications, 2015, 6 (7729)
- T. J. Czaczkes, F. L. W. Ratnieks Cooperative transport in ants
(Hymenoptera: Formicidae) and elsewhere. Myrmecol. News, 2013, 18, pp. 1–11 (2013). - S. Berman, Q. Lindsey, et al. Experimental Study and Modeling of Group Retrieval in Ants as an Approach to Collective Transport in Swarm Robotic Systems Proc. IEEE, 2011, 99, pp. 1470–1481
- I. D. Couzin, J. Krause, et al. Collective memory and spatial sorting in animal groups J. Theor. Biol., 2002, 218, pp. 1–11.
- M. Ballerini, et al. Interaction ruling animal collective behavior depends on topological rather than metric distance: evidence from a field study Proc. Natl Acad. Sci. USA, 2008, 105, pp. 1232–1237
- S. B. Rosenthal, et al. Revealing the hidden networks of interaction in mobile animal groups allows prediction of complex behavioral contagion Proc. Natl Acad. Sci USA, 2015, 112, 201420068