Abstracts » Zoology and wildlife conservation

Artificial neural networks as models of stimulus control

Article Abstract:

Experiments were conducted to determine the feasibility of using artificial neural network models for predicting stimulus-response relationships in animal behavior research. Experiments involved networks that were trained on the same tasks as the animals. Empirical results considered include generalization, peak shift and stimulus intensity effects. Results obtained using the models were found to agree well with the empirical results.

Usage, Computer networks, Neural networks

---

Discrete conventional signalling of a continuous variable

Article Abstract:

Animals often use discrete signals in aggressive interactions. A model of fighting that allows for conventional signalling of subjective resource value has been investigated. In the model, perfect signalling and non signalling were not evolutionary stable strategies (ESSs). ESSs were found whereby partial information was communicated, with discrete signalling of a range of values, and not a precise one.

Observations, Signals and signaling, Signaling

---

A century of generalization

Article Abstract:

Data from both ethology and psychology about generalization is reviewed which shows how animals respond to sets of stimuli including familiar and novel stimuli. It is concluded that patterns of generalization are largely independent of systematic group, behavioral context, and sensory modality and of whether reaction to stimuli is learned or genetically inherited.

Science & research, Evaluation, Stimulus generalization, Animal behaviour

Some parts © 2013 Advameg, Inc.