Social interactions in natural populations of weakly electric fish

Abstract

Animals and their sensory systems evolved in specific environments and in the context of their particular ethological niches. It is often found that sensory neurons are tuned to the statistics of the natural scenes that they likely to experience. Accordingly, the importance of knowledge of natural stimuli and the problems faced and solved by sensory systems in their natural environments for the understanding of neural processing is increasingly recognized.

Weakly electric fish are successful model organisms for studying the neural mechanisms underlying sensory processing in vertebrates. These mostly nocturnal animals evolved an active electric sense employed in navigation, foraging and communication. Nocturnal conditions, murky water, and the tropical environment make their natural habitats challenging study sites. Therefore, most data on their natural behavior and their communication signals have been acquired under restricted lab conditions or remain anecdotal. However, their permanently active electric organ discharges provide an excellent opportunity to monitor the movements and communication of individual unrestrained fish. The central goal of the present thesis has been to establish and to apply a method for the non-invasive quantification of electrocommunication stimuli while animals roam and interact in their natural environment.

In Chapter 2, we present an automated tracking system allowing for the reliable and continuous tracking of wave-type electric fish based on the individual-specific frequency of the electric organ discharge. The system extracts frequency modulations of the EOD on short and long time scales, and estimates location and orientation of the tracked fish.

We acquired data on natural communication of the ghost knifefish, Apteronotus rostratus, during its reproductive period, by deploying our tracking system in the Panamanian rain forest (Chapter 3). We tracked individuals and characterized dyadic interactions and the corresponding electro-communication scenes. We showed that a specific communication signal, independent of context, was almost exclusively emitted in close proximity to a conspecific. During courtship, the communication of males was precisely locked to that of females. Our data also showed that competing male intruders can be detected and responded to over larger distances of up to 170 cm, even in the presence of a much stronger EOD of a nearby female conspecific. For the observed interactions we extracted frequency differences and estimated effective signal intensities, and related those to the response properties of the P-unit electro-receptors. Surprisingly, we found that in many relevant communication situations the electro-receptors will be driven only weakly by electric communication signals, either because of a frequency mismatch in courtship or because of large interaction distances in agonistic contexts. This study is the first account for the detailed monitoring and characterization of electric fish movement and communication in their natural habitat.

To determine the behavioral context of the male-female interaction observed in Panamá, we conducted a long-term breeding experiment in the laboratory with the closely related species A. leptorhynchus (Chapter 4). We used our tracking software to identify male-female communication scenes similar to those observed in the field and demonstrated its relationship to courtship and spawning. Sequence and dynamics of the chirping during courtship closely matched that observed in the field. We found that both the female long chirp signaling spawning and the quick and precisely timed male echo response to female chirps are conserved across species.

Applying our tracking system we revealed the properties of natural communication situations. We then demonstrated how our system can be used to further characterize the behaviors observed in the field in a tailor-made long-term laboratory study.