Behavioral neuroscientists have long postulated the existence of ‘decision-making neurons’, which are defined as individual or small groups of interneurons that play a critical role in switching between alternative behaviours. These neurons act by integrating a convergence of sensory inputs to direct downstream signalling towards motor pattern generating centres (Fig.1). To experimentally define neural decision-making neurons, one must show that the neural activities of those cells are necessary and sufficient to trigger a complete behavioural outcome.
Many studies have tried to identify decision-making neurons in a variety of model organisms. The current models of decision-making neurons have several limitations to a better understanding of the underlying circuit principles. First, most model organisms (for example, snails) used to date have limited genetic toolkits to dissect functional neural circuits at a cellular resolution. Second, most of the behavioural paradigms that have been used to identify decision-making neurons are simple reflexes rather than complex behaviours, which required a high degree of cognitive function observed during human decision-making.
The fruit fly, Drosophila melanogaster has provided promising behavioural paradigms of decision-making processes. Cutting-edge genetic intersection methods have allowed the establishment of a functional map of the underlying neural circuitry. Although these studies have contributed to a greater understanding of decision-making neurons, the fundamental principles underlying switching between alternative behaviours are still poorly understood. Here we propose to utilize our recently established behavioural paradigms, and their neural circuits which we have identified, to determine the fundamental principles underlying the means by which decision-making neurons can switch between two alternative behaviours.
To identify decision-making neurons, I established two alternative behavioural paradigms of male’s mating decision. Longer-mating-duration (LMD) occurs when males prolong mating following persistent exposure to rivals, whereas shorter-mating-duration (SMD) occurs when males are sexually satiated (Fig. 2). This simple behavioural metric is reproducible, quantifiable, and is easily amenable to genetic manipulation, making this a powerful model system for studying decision-making neurons. While LMD and SMD each involve unique sensory modalities, they have shared requirements for specific interneurons and memory circuits, suggesting the same decision-making neurons regulate the interplay between the two behaviours. To identify neurons that control the LMD or SMD decision, we screened known neuropeptide signaling pathways in Drosophila and identified four neurons that express the particular neuropeptide, which is known to be associated with both behaviours. Our goal for the next 5 years is to identify and fully characterize the integrative circuits of the NP-expressing neurons that regulate LMD and SMD. These studies will reveal fundamental circuit principles of decision-making neurons in Drosophila. The proposed work is not only relevant to studies of fly behaviour but also to the genetics of decision-making behaviour in other organisms. Indeed, like many other neural mechanisms initially dissected in Drosophila – e.g. learning and memory – the cellular and molecular basis of decision-making behaviour is likely to be conserved throughout evolution in the form of a molecular “toolkit”. This feature will help us begin to understand the neural basis of behaviour at the cellular and molecular level in organisms with more complex nervous systems.
The unique behavioural paradigm coupled with study of identified neurons will provide trainees with unparalleled expertise in neuroscience, genetics, and molecular biology – all highly desirable skills for academic, government and private sector research. The students will be able to contribute to scientific knowledge and will have a solid platform to develop their interests in pursuing a career in research or other related fields (Fig. 4).