Collective cognition in animal groups

Animal groups like schools of fish, swarms of insects and flocks of birds have always amazed us by superb group-management and collective effort in group migration, nest selection, prey-hunting, and foraging.

How do they do it?

Decision-making by individuals within such aggregates is so synchronized and intimately coordinated that it has previously been considered to require telepathic communication among group members or the synchronized response to commands given, somehow, by a leader. Recent studies have begun to elucidate how the repeated interactions among grouping animals scale to collective behavior, and have revealed, remarkably, that collective decision-making mechanisms across a wide range of animal group types are similar and share essential common features with mechanisms of decision-making within the brain.

Collective motion

An individual animal in a large group is sensitive only to its local neighbourhood and has no idea of the global orientation (or threat). It bases its movement on the basis of locally acquired cues like position, orientation or any sudden movement. So the group consists of individuals who have entirely different informational status!

From a bug's point of view:

The animal is moving in the x-direction and alpha is the field of vision so it cannot see the conical volume directly behind it. The area aound it is devided into three zones, namely:

1. Zone of repulsion (zor): The animal wants to maintain this much of distance between itself and the nearest neighbours. This repulsive tendency has very high priority for the animal.

2. Zone of orientation(zoo): The animal would orient according to the individuals present in this zone. This is necessary as it helps them to follow the group.

3. Zone of attraction(zoa): When the animal is not busy repelling, it moves(gets attracted) towards the group so as not to be on the periphery or be left behind.

 

 

Multiple stable collective states can exist for the exactly the same individual behavior, and be dependent on the previous history of group structure, such as whether behavioral parameters are increasing or decreasing (see the figure above), despite no individual having memory of that structure. This demonstrates that animal groups can exhibit a formof hysteresis, or ‘collective memory’. These multiple stable states are just like the multiple states in neurons. Multistability in neural systems has been suggested as an important mechanism for memory storage and temporal pattern recognition. So this is visible in the case of animal groups too.

 

Groups as a distributed system

An animal group has an opportunity to sense the environment at multiple locations at a time and with enirely new perspective. All these perceptions finally culminate in deciding the course of action of the group like avoiding a water-pool or fleeing a predator. This enomously increases the eficiency of the group much like our certain decentralized neural assemblies which eventually increase the computational capability of the neural system which is not possble in isolation.

The information hence, from each of the group-member has to be amplified so that any danger or predator can be avoided or a new food supply can be exploited. A positive feedback loop acts to amplif the information. But if everybody senses and judges, then group would very quickly become depolarized into a random swarm. So, there has to be a damping mechanism(negative feedback loop) which fades out some of the information to make the group a robust system. This balance of amplification and damping is maintained to keep the group safe and robust.

Speed-Accuracy trade-off

The decisions are based on uncertain information which has been accumulated over a certain period of time. This however, slows down the overall system reponse. Jumping to a decision based on an amplified information could lead to a sub-optimal decision proving a loss for the group although it is very quick. When there is no hurry, the group may place accuracy over speed but in a pressing situation like a predator attack, speed is the main consideration.

Our brain does the same.

 

Consensus decision making

Who makes the decisions? Why should everybody agree? How are the decisions made?

The consensus decisions are of three types:-

1. Unshared : taken by a single dominant animal with all other members abiding by its decision.

2.Partially shared : few individuals from a particular demogrphic group(eg., adult males) make the decision.

3. Equally shared : all group members contribute equally to the decision, independently of their individual identities or social status.

Note: The kind of decision taken doesnot imply whose interests are being served.

 

Foraging Ants: Equally shared decision. 30% move out as scouts. Come back and recruit new scout and lead it to the site and when a threash-hold(quorum) reaches, the recruiters carry the remaining ants to the new nest.

Swarming bees: Scouts move out. Come back and dance. Longer dance attracts more followers and these follow the scouts to respective sites. These come back and recruit more. When quorum is reached everybody is taken to that place. If a conflict arises mid-air, they settle down and repeat the process once again.

 

Conflict of Interests

The consensus decisions in which there is no or little conflict of interest are the decisions made by eusocial insects about choosing a new nest site, or by navigating birds about travel routes, because the goal (finding the best nest site, or taking the best route, respectively) is similar for all group members. But it is highly possible that many-a-time some are not happy with the decision. This occurs in the decisions that are mutually exclusive like resting or going out for foraging, travel destination- one offering more food while other more water.

In large groups (like bees), one group may voluntarily compromise their interests for the majority to maintain group solidarity. This compromise is termed as 'consensus cost' which the rebel bears. Given that decisions about activity timing and travel destination have to be made regularly during each day, day after day, related consensus costs can augment. In the case of red deer conflicts between the sexes about activity budgeting are so large that they lead to intersexual social segregation.

In smaller groups, the different interest groups may fight it out leading to coercion by the dominant.

Emergence of leaders

The one who needs something most, would "pace" for it and thereby leading the group. Eg., if in a group of two(pair of homing pigeons) one wants to rest while the other needs to forage (as its reserves have exhausted), then the latter one would lead the group to quicken the decision.

"Wisdom of crowds"

The larger the number of individuals participating in the decision-making exercise, the more is the accuracy of the decision made. The 'noisy' information from each individual is pooled and then the decision is made. This pooling of information is accordingly termed as "Information pooling".

Democratic decision making in social animals

table1

 

Selfishness

What happens if the scouting bee dances longer than the optimal duration for a given site so as to lie to pursuade others in moving to the site that it has chosen? This may happen out of selfishness to favour its interests.

In animal groups like huamans or gorillas and other primates, there are individual rewards in case the individual leads the group to a better state. The rewards are generally an increase in social status or monetary(in humans). What prevents the swarm insects from selfishness is still not clear and is a subject of research.

Collective cognition through environmental modification

In highly related grouping organisms, such as the social insects (e.g. ants, bees, wasps etc.), collective cognition can be particularly sophisticated because individual behavior and interactions have evolved to benefit the colony reproductive success (thus reducing inter-individual conflict), a functional integration so tight that they have been termed ‘super-organisms'.

Ants show this kind of environment modification by depositing pheromone trails on the path thereby facilitating quicker movement of the entire group. It helps in finding a shorter route as in a shorter route has high probability of being travelled quickly and hence, a thicker trail is deposited on it. This pheromone helps another ants to choose this path by laterally inhibitting the longer ones. These pheromones are of low volatility and hence remain for a longer perod of time. High volatility pheromones act as 'attention-catchers' and help in quick and sudden change of path maybe, to avoid danger or exploit a newly found reserve.

 

Conclusion

Through collective action, animals of many species can enhance their capacity to detect and respond to salient features of the environment. This collective decision making helps the group and hence each individual to make more accurate decisions with uncertain and incomplete information. Collective behavior allows access to important higher-order information-processing capabilities that are very d difficult, or impossible, to achieve in isolation. This ability is important and bears close comparison with distributed processing of information within and among neural assemblies.

 

References

  1. Couzin,I.D. (2009) Collective cognition in animal groups. Trends in Cognitive Sciences Vol.13,number-1, 36-43
  2. Couzin, I.D. et al. (2005) Effective leadership and decision-making in animal groups on the move. Nature 433, 513–516
  3. Couzin, I.D. and Krause, J. (2003) Self-organization and collective behavior in vertebrates. Adv. Stud. Behav. 32, 1–75
  4. Sumpter, D.J.T. (2006) The principles of collective animal behavior. Philos. Trans. R. Soc. Lond. B Biol. Sci. 361, 5–22
  5. Couzin, I.D. et al. (2002) Collective memory and spatial sorting in animal groups. J. Theor. Biol. 218, 1–11
  6. Conradt, L. and Roper, T.J. (2003) Group decision making in animals. Nature 421, 155–158
  7. Couzin, I.D. (2007) Collective minds. Nature 445, 715