Eavesdropping and language dynamics

Communication in nature is not restricted to the transmitter-receiver pair. Unintended listeners, or eavesdroppers, can intercept the signal and possibly utilize the received information to their benefit, which may confer a certain cost to the communicating pair. In this paper we explore (computationally and mathematically) such situations with the goal of uncovering their effect on language evolution. We find that in the presence of eavesdropping, languages exhibit a tendency to become more complex. On the other hand, if eavesdroppers belong to a different (competing) population, the languages used by the two populations tend to converge, if the cost of eavesdropping is sufficiently high; otherwise the languages synchronize. These findings are discussed in the context of animal communication and human language. In particular, the emergence of synonyms is predicted. We demonstrate that a small associated cost can suppress synonyms in the absence of eavesdropping, but that their likelihood increases strongly with the probability of eavesdropping.     

Complex Adaptive Systems and the Evolution of Reciprocation

Complex adaptive systems play a major role in the theory of reciprocal altruism. Starting with Axelrod's celebrated computer tournaments, a wide variety of computer simulations show that cooperation can evolve in populations of selfish agents, both with direct and indirect reciprocation. Received 14 April 1998; accepted 16 June 1998.     

Predator hunting modes and predator–prey space games

Predators and prey are often engaged in a game where their expected fitnesses are affected by their relative spatial distributions. Game models generally predict that when predators and prey move at similar temporal and spatial scales that predators should distribute themselves to match the distribution of the prey's resources and that prey should be relatively uniformly distributed. These predictions should better apply to sit-and-pursue and sit-and-wait predators, who must anticipate the spatial distributions of their prey, than active predators that search for their prey. We test this with an experiment observing the spatial distributions and estimating the causes of movements between patches for Pacific tree frog tadpoles (Pseudacris regilla), a sit-and-pursue dragonfly larvae predator (Rhionaeschna multicolor), and an active salamander larval predator (Ambystoma tigrinum mavortium) when a single species was in the arena and when the prey was with one of the predators. We find that the sit-and-pursue predator favors patches with more of the prey's algae resources when the prey is not in the experimental arena and that the prey, when in the arena with this predator, do not favor patches with more resources. We also find that the active predator does not favor patches with more algae and that prey, when with an active predator, continue to favor these higher resource patches. These results suggest that the hunting modes of predators impact their spatial distributions and the spatial distributions of their prey, which has potential to have cascading effects on lower trophic levels.     

An Active Self-Optimizing Multiplayer Gaming Architecture

Multiplayer games are representative of a large class of distributed applications that suffer from redundant communication, bottlenecks, single points of failure and poor reactivity to changing network conditions. Many of these problems can be alleviated through simple network adaptations at the infrastructure level. In this paper, we describe a model in which game packets are directed along the edges of a rooted tree connecting the players, aggregated during the upstream flight and multicast from the root to the leaves. This tree is constructed based on a heuristic, and can dynamically adjust itself in response to changes in network conditions. This gaming infrastructure is built and maintained using active networks, which is currently the only open architecture suitable for these types of applications. We have designed and implemented a prototype using ANTS that performs these adaptations for unmodified DOOM clients. We present analytical and simulation results that illustrate the reduction in communication overhead, and show that the multicast tree can quickly adjust to changing network conditions. The overhead of the active network-based middleware is acceptable, especially in wide-area networks.