All signals are not equal: acoustic signalling of individuality, sex and breeding status in a cooperative breeder

Repeated interactions between individuals in socially living animals select for the evolution of signals that convey information identifying individuals or categories of individuals, which may enable the discrimination of familiar versus unfamiliar individuals. Such information may help animals maximize their inclusive fitness by adjusting their own behaviour, allowing them to avoid conflict, preferentially direct help and/or ignore unreliable individuals. Acoustic signals in birds provide the potential to encode individual-specific information. We examined the degree to which individual identity, sex, breeding status, group membership and genetic relatedness were related to variability in six different call types, which occurred across a variety of different behavioural contexts in the apostlebird, Struthidea cinerea, a socially living and cooperatively breeding Australian passerine. We demonstrated that not all calls reflected the same extent of information. Of the six call types, call variation was related to individual identity in three call types, breeding status in two call types and sex and group relatedness in one call type. Finally, variation in two call types was not related to any of the measured variables. Our results suggest that some, but not all, acoustic signals in apostlebirds may be selected for individual distinctiveness between individuals and categories of individuals (male versus female, breeder versus nonbreeder), and these signals may be important in determining levels of cooperation and interaction between individuals in this cooperatively breeding society. © 2014 The Association for the Study of Animal Behaviour.

Robust modular artmap for multi-class shape recognition

This paper presents a fuzzy ARTMAP (FAM) based modular architecture for multi-class pattern recognition known as modular adaptive resonance theory map (MARTMAP). The prediction of class membership is made collectively by combining outputs from multiple novelty detectors. Distance-based familiarity discrimination is introduced to improve the robustness of MARTMAP in the presence of noise. The effectiveness of the proposed architecture is analyzed and compared with ARTMAP-FD network, FAM network, and One-Against-One Support Vector Machine (OAO-SVM). Experimental results show that MARTMAP is able to retain effective familiarity discrimination in noisy environment, and yet less sensitive to class imbalance problem as compared to its counterparts.