Learning and memory in moths - plasticity in behaviour and neurophysiology

Abstract: The moth, Spodoptera littoralis, was demonstrated to be a new model system for studying mechanisms underlying odour conditioning. In the experiments a restrained moth was trained to learn an association between a conditioned stimulus (CS) and an unconditioned stimulus (US). A proboscis extension reflex (PER) elicited by the conditioned odour was observed after learning. In paper I, an odour-sucrose association was demonstrated when the CS was presented 1-3 s before the US during training. A wide range of inter-trial intervals was able to support conditioning. In addition, both male and female moths were found to learn equally well. In paper II, moths were found to remember an odour-sucrose association for 5 days even after a single training trial. After one learning experience, proboscis extension performance to a conditioned odour was shown to consist of three distinct phases, providing the first behavioural evidence supporting a tripartite memory formation model in insects. In paper III the discrimination ability of the moths was studied, and already after a single experience, they were shown to discriminate between a rewarded and a non-rewarded odor. The moths were able to undergo reversal conditioning already after two learning trials. In the last paper, intracellular recordings from the suboesophageal ganglion (SOG) were made concurrent with behavioural conditioning. The intracellular recordings revealed that about half of the sucrose-responding neurons in the SOG changed their specificity after conditioning with an odour. Two different response patterns were observed. In the first, neurons responded to the conditioned odour; these neurons might be acting as "learning" neurons. In the second, the spontaneous firing rate increased after training; these neurons might be involved in sending reinforcement information to higher centres of the brain. As the in vivo preparation with intact mouthparts provides stable neurophysiological recording conditions, learning experiments can be performed while a neuron is monitored and a conditioned behavior is simultaneously observed. Using the well-studied olfactory system of the moth combined with the power of the PER preparation can now provide an essential experimental linkage between behavioural responses and cellular mechanisms underlying olfactory conditioning.