Odour-background segregation and source localisation using fast olfactory processing

Abstract

The hungry insect relies on olfaction to find food patches. However the natural environment is full of different odours made up of a variety of odorants intermingling together. So how can the insect recognise which odorants belong to a patch of good food that is worth visiting? Segregating appetitive food odorants poses a difficult challenge for the insect, as it must separate target odorants from mixtures of odorants that come together from a variety of sources, a process called odour background segregation. During flight, the insect can use spatial and temporal information in turbulent odour plumes to determine whether odorants come from one source or multiple sources. The insect olfactory system can process odorants rapidly, matching the resolution of other senses such as vision and audition. When the insect arrives at an odour source, it must forage alongside other insects to localise the odour to its source and find its food reward. The insect could potentially gain information from other nearby conspecifics that would increase its success in foraging. Thus, during the search for food, the insect olfactory system must process information from odour plume structure and odour valence, from memories about good food sources and from social information transfer between conspecifics. This thesis concentrates on what information insects use during olfactory search to locate a food source and the neural mechanisms that convert an olfactory stimulus into a behavioural response. Two of the chapters focus on odour background segregation, primarily what temporal information insects (honey bees and fruit flies) can use to segregate an appetitive odorant from a background odour mixture. The last chapter focuses on social information transfer between insects (fruit flies) at an odorous food source, and whether such information can improve foraging. Firstly, I asked whether insects were able to distinguish a target odorant from an odour background using odorant onset asynchrony, when the insect had never experienced this target odorant alone. I addressed this question using Apis mellifera, the honey bee, by appetitively conditioning fixed honey bees to a target odorant while presenting a complex background mixture. In my first chapter, I demonstrated that honey bees could separate an unknown target odorant from a mixture using odorant onset asynchrony, however the onset asynchrony was in the range of seconds, two orders of magnitude larger than previously reported for segregation of known odorants. This implies that segregation of unknown odorants may depend on other neural mechanisms such as sensory adaptation. Secondly, I asked what the behaviourally relevant timescales of temporal stimulus cues were for odour source segregation. I addressed this by presenting Drosophila melanogaster with pulses of binary mixtures of attractive and aversive odorants in a wind tunnel, examining their responses to different onset asynchrony times and odorant combinations. In my second chapter, I demonstrated that fruit flies can distinguish between synchronous and asynchronous mixtures of odorants of opposing valence, therefore could use this information for determining the number of odour sources in their environment. Thirdly, I asked whether social interactions between insects could affect their proficiency of foraging and their memory expression. I addressed this using an automated conditioning assay for Drosophila melanogaster, where I conditioned flies in different group sizes to associate an odorant with food, and tested their short-term memory for the conditioned odorant. In my third chapter, I demonstrated that the associative memory of the conditioned odorant is extended for flies conditioned and tested in larger groups, compared to flies conditioned and tested as individuals or in pairs. This extended memory expression could be due to the increased number of social interactions between flies in the larger group, through which flies could transfer information about the location and quality of the food source. Altogether, these three chapters provide evidence that insects can use temporal information to segregate relevant odorant stimuli from background mixtures and can use social information to improve source localisation.