Focal Issues

The list of questions and issues was discussed by subgroups. Each subgroup consisted of participants representing all levels of behavioral organization from genes to ecology. Each subgroup condensed the list of questions and issues and a consensus emerged for the following focal issues:

1. The origins of social organization

2. Division of labor in insect colonies

3. Communication and signaling

Groups then were assigned one of these three issues which they then discussed and reported back to the group.

Group Reports

The following reports were prepared by each of the three subgroups.


We discussed a set of features that marked transitions in the evolution of social insects. We attempted to characterize these in both general ways that would illuminate comparisons with other complex systems and in the concrete ways that are displayed in social insect species. Our initial characterization was:


incohesive group — unstructured

Structured groups


Facultative group living

Obligatory group living



Reproductive differentiation

Somatic task differentiation



Brood tending



Irreversible differentiation

Morphological differentiation


Somatic task differentiation

With this framework in hand we explored a series of possible analogies with other systems. Investigation of these would perhaps illuminate both shared and diverse features of the different systems. Comparisons were discussed with

We concluded our discussion by attempting to identify minimal conditions that would permit social cooperation to occur. While tolerance was suggested, this was elaborated to point to ignorance (lack of recognition) of others, or shared expectation (or behavioral algorithms) so that negative consequences of interaction would be minimized to permit positive consequences of interaction to take hold. There is a tension between cohesiveness and differentiation.

Questions for further research:

Are solitary females tolerant of each other or do they usually fight? Both types are known, but what are the factors that influence the different responses?

How does increasing size affect colony cohesiveness and communication?


We subdivided division of labor into two major questions: 1) How do we integrate between levels of organization? 2) What genetic, anatomical, neurophysiological, developmental, behavioral, and environmental changes occur with the transition to a higher level of social organization?

We dealt with question 1 by recognizing the different levels of organization in societies (including human society)

Levels of organizations


Cells (neurons)



Task groups



We realized that to fully understand the emergent property of any level, we must study the level below.

To address question 2, we focused on the transition from simple to complex societies, or in the terminology we used, hierarchical to networked societies. We discussed how best to define these two extremes of the societal complexity spectrum and how to empirically determine them. We then asked whether the transition from hierarchical to networked societies is continuous or discrete and what are the causes/constraints of the transition. The causes can be environmental, genetic, or group size, while one obvious constraint is communication among members.

Next we discussed the mechanisms for the transition. How much genetic change is necessary for the development of a complex society from a simple one? Can higher levels of social behavior evolve from simpler ones without changes in anatomy and physiology? Are fundamental differences in learning and sensory capacity required, and can we identify specific physiological, anatomical, and environmental correlates? Are there environmental or phylogenetic constraints on the evolution of anatomic castes?

A related question is in a highly social organization, what are the biochemical, neuronal, correlates of behavioral transitions in workers.


The subgroup "Communication" identified eight major topics which all relate to societies of different group sizes and social structures. For each of these topics, the group considered changes in design properties that might be associated with group size. In particular we focused on mutualistic egalitarian societies, primitively hierarchical structures, and network organized eusocial systems.

Honest versus dishonest signaling

A careful examination of existing literature should be done to test the application of these concepts to social insects. Our working hypothesis is that dishonest signaling would be less likely in large, network-based societies where as in mutualistic of primitively eusocial societies may have dishonest or manipulative communication.

Signals versus cues

The definitions in the literature are confusing. We aim for a non-ambiguous definition of these concepts. The waxy layers on the outer cuticle of insects mainly consisting of fatty acids and hydrocarbons are required as a protection from microorganisms and desiccation. In many social insects these compounds also serve as cues and in some situations may have evolved to become signals for recognition or other functions.


We consider that direct, inter-individual communication (primarily observed in communal and primitively eusocial societies) becomes augmented or perhaps replaced by mass communication in networked eusocial systems of larger size. In these larger societies, for example, mass communication is involved in feedback loops in maintaining social homeostasis.


The signal repertoire increases with increasing size and social complexity. Examination of existing literature would be a first step in testing these ideas.

Signal design, economy and efficacy

We addressed the question of multimodality and signal to noise ratios. The importance of constraints in signal design was also considered along with the significance of signal redundancy and amplification. The possibility of a primitive syntax by simple signal combinations was discussed along with the possibility that social insects construct their own medium for signal transmission (nest structures).

Context dependant response

We discussed the concepts of message versus meaning in a signal and the ways in which environmental context influences responses. The "motivational state" affecting the response threshold to stimuli was also discussed.


We discussed the anonymity and specificity of recognition systems and their relationship to directly interacting groups versus large mass communicating societies. The hypothesis is that inter-individual recognition should be more specific in directly interacting societies and become more group based to serve as group recognition signals in network based, large societies.

Signal processing

We considered the neurobiological basis of signal processing including sensory physiology, and central processing. The possible role of learning in communication process was considered along with the neuroendocrine aspects of communication.

SFI-Social Insect Page