The Time — Dependent Storing of Optical Information in the Honeybee

Abstract

In several series of experiments the learning behaviour of honeybees and its dependence on the parameters “quantity of reward” and “duration of reward” is analysed. The bees were trained to learn colour marks of artificial foodsources. The experiments show that over a wide range of different quantities of reward the learning behaviour is not influenced. In contrast to these findings, temporal parameters of the program of reward have a specific influence on the learning efficiency in honeybees.

In experiments with different programs of reward it can be shown that interruptions of sucking lead to an improvement of the certainty of reproduction of the signal learned. The temporal consolidation of the learned signals was analysed by registering the learning behaviour for a duration of 16 minutes after a single reward. The effect of repeated testing on the choice behaviour and the motivation is investigated and its significance discussed for very short rewards. Information-theoretical considerations show a convergence between the dynamic and stochastical description of the learning system.

In an analysis of the learning behaviour of honeybees MENZEL (1968) showed that a colour as a learning signal is stored in a short-term or long-term form. With rewards lasting no longer than 5 seconds, the learning signal is always stored in the short-term form; with rewards lasting longer than 10 seconds, it is always stored in the long-term form. These results are the basis for further investigations analysing the learning capacity of bees after short rewards. Parts of these experiments will be discussed in this paper.