Abstract
In social species, group functions often benefit from variation among individual group members. Many highly integrated social insect colonies rely on division of labour among colony members and emergent properties of their collective behaviour and physiology. Response threshold models are a prominent proximate explanation of division of labour, but how variation in response thresholds arise is largely unexplored. We propose the Weak Worker Hypothesis, a novel conceptual framework suggesting that response thresholds are determined by an individual's susceptibility to the stressor that underlies the task. Thus, specific tasks are preferentially performed, or at least initiated, by the individuals that are most susceptible to the corresponding stressor. Consequently, 'weak' workers that are susceptible to a particular stressor play a disproportionate role in the group's defence against this stressor. The response threshold manifests as an internal evaluation of a task-specific stimulus that is influenced by the severity of the physiological perturbation of the individual, which simultaneously determines the susceptibility of this individual to succumb to the external disturbance. As long as individual stress susceptibilities vary among different stressors, this model generates division of labour and thus group stability. The Weak Worker Hypothesis provides a functional explanation for individual-level responses to environmental deviations from optimal conditions. Such a deviation could be directly perceived as stimulus and simultaneously lead to physiological stress, or the physiological stress caused by the deviation could be the stimulus itself. In support of the Weak Worker Hypothesis, we present experimental evidence of a link between individual heat susceptibility and fanning behaviour in honey bees (Apis mellifera L.). We also discuss other possible cases and how to test our idea empirically in other contexts, keeping in mind the important distinction between cause and consequence. Finally, we conclude that the Weak Worker Hypothesis could provide a useful extension of response threshold models for understanding the division of labour in social groups, which might have repercussions for applied social insect science, selective breeding and eradication efforts.