Life is always transitioning from one state to another. Some state transitions can be physical, such as the small Arctic fox changing its fur from brown to white when a snowy winter approaches, but they can also be behavioural, such as when my fellow humans grumble and complain about it being too hot or cold. To find out how thermal stress may affect transitions between different personality states, Grant Navid Doering from the University of California, Santa Barbara, USA, and a group of international colleagues turned to a spider, Anelosimus studiosus.

The animals have two very distinct and heritable personalities, aggressive and docile, both of which typically occur in a single colony. However, the aggressive spiders become more irritable as the temperature rises, so Doering and his team first created a computer simulation to determine whether these differing personalities may influence the response of the colony to changes in temperature. They included three behavioural states in their calculations – one for the docile individuals and two for the aggressive (relaxed and agitated) – in addition to various physical parameters, including temperature, colony size, the rate at which the spiders speed up as the temperature rises and how often a relaxed spider is aggravated by an agitated spider. The simulation suggested that the composition of personalities in the colony and the temperature influenced the number of agitated spiders; more grumpy spiders and a hot house resulted in even more grumpy spiders. The simulation also revealed that fewer spiders became agitated in smaller colonies because they bumped into each other less often than in larger colonies.

To test their simulation, Doering and his crew formed colonies of live spiders with group sizes of either 6 or 20 individuals with varying personality mixtures: only aggressive, only docile, or half of each personality. The team then subjected each of these manufactured spider colonies to a temperature cycle ranging from 27 to 33°C, warming the colony and then cooling it again several times. During these heating and cooling cycles the researchers watched and documented all of the interactions, cordial or hostile, between the spiders.

The team observed that at a specific temperature, which they termed the social tipping point, the spider colonies transitioned from a relaxed to an agitated state as the temperature rose and then calmed down as the mercury dipped below that level. However, this temperature differed between the aggressive and docile spider colonies. During the heating phase, aggressive groups became agitated and displayed violent interactions more quickly, i.e. at lower temperatures, in comparison to the docile groups. It also took longer for these aggressive groups to calm down during cooling. Interestingly, the temperature at which the agitated spiders mellowed during cooling was lower than the transition temperature during heating for the aggressive groups, suggesting it takes more effort for overheated and worked up spiders to chill out. This trend was similar in the small and large aggressive groups, although there were more occasions when the spiders bumped into each other in the larger groups. In contrast, the social tipping point for the docile colonies was similar during heating and cooling and between the group sizes. Strangely, the shift from mild to aggressive behaviour in the small mixed colonies was more similar to that of the aggressive groups, whereas the shift in the large mixed colonies was more similar to that of the small and large docile groups even though the social tipping point was lower. Although Doering and colleagues are not entirely sure why, they suggest that larger mixed colonies may cope better with thermal stress.

This exciting research shows that the personality types of spider colonies do not shift gradually from one to another, but they can transition suddenly when the temperature is just right.

Doering
,
G. N.
,
Scharf
,
I.
,
Moeller
,
H. V.
and
Pruitt
,
J. N.
(
2018
).
Social tipping points in animal societies in response to heat stress
.
Nat. Ecol. Evol.
2
,
1298
-
1305
.