Shallow neural networks trained to detect collisions recover features of visual loom-selective neurons

Animals have evolved sophisticated visual circuits to solve a vital inference problem: detecting whether or not a visual signal corresponds to an object on a collision course. Such events are detected by specific circuits sensitive to visual looming, or objects increasing in size. Various computational models have been developed for these circuits, but how the collision-detection inference problem itself shapes the computational structures of these circuits remains unknown. Here, inspired by the distinctive structures of LPLC2 neurons in the visual system of Drosophila, we build anatomically-constrained shallow neural network models and train them to identify visual signals that correspond to impending collisions. Surprisingly, the optimization arrives at two distinct, opposing solutions, only one of which matches the actual dendritic weighting of LPLC2 neurons. Both solutions can solve the inference problem with high accuracy when the population size is large enough. The LPLC2-like solutions reproduces experimentally observed LPLC2 neuron responses for many stimuli, and reproduces canonical tuning of loom sensitive neurons, even though the models are never trained on neural data. Thus, LPLC2 neuron properties and tuning are predicted by optimizing an anatomically-constrained neural network to detect impending collisions. More generally, these results illustrate how optimizing inference tasks that are important for an animal's perceptual goals can reveal and explain computational properties of specific sensory neurons.

Critical periods shaping the social brain: A perspective from Drosophila

Many sensory processing regions of the central brain undergo critical periods of experience-dependent plasticity. During this time ethologically relevant information shapes circuit structure and function. The mechanisms that control critical period timing and duration are poorly understood, and this is of special importance for those later periods of development, which often give rise to complex cognitive functions such as social behavior. Here, we review recent findings in Drosophila, an organism that has some unique experimental advantages, and introduce novel views for manipulating plasticity in the post-embryonic brain. Critical periods in larval and young adult flies resemble classic vertebrate models with distinct onset and termination, display clear connections with complex behaviors, and provide opportunities to control the time course of plasticity. These findings may extend our knowledge about mechanisms underlying extension and reopening of critical periods, a concept that has great relevance to many human neurodevelopmental disorders.

Dopamine modulates the optomotor response to unreliable visual stimuli in Drosophila melanogaster

State-dependent modulation of sensory systems has been studied in many organisms and is possibly mediated through neuromodulators such as monoamine neurotransmitters. Among these, dopamine is involved in many aspects of animal behaviour, including movement control, attention, motivation and cognition. However, the precise neural mechanism underlying dopaminergic modulation of behaviour induced by sensory stimuli remains poorly understood. Here, we used Drosophila melanogaster to show that dopamine can modulate the optomotor response to moving visual stimuli including noise. The optomotor response is the head-turning response to moving objects, which is observed in most sight-reliant animals including mammals and insects. First, the effects of the dopamine system on the optomotor response were investigated in mutant flies deficient in dopamine receptors D1R1 or D1R2, which are involved in the modulation of sleep-arousal in flies. We examined the optomotor response in D1R1 knockout (D1R1 KO) and D1R2 knockout (D1R2 KO) flies and found that it was not affected in D1R1 KO flies; however, it was significantly reduced in D1R2 KO flies compared with the wild type. Using cell-type-specific expression of an RNA interference construct of D1R2, we identified the fan-shaped body, a part of the central complex, responsible for dopamine-mediated modulation of the optomotor response. In particular, pontine cells in the fan-shaped body seemed important in the modulation of the optomotor response, and their neural activity was required for the optomotor response. These results suggest a novel role of the central complex in the modulation of a behaviour based on the processing of sensory stimulations.

Noise-robust recognition of wide-field motion direction and the underlying neural mechanisms in Drosophila melanogaster

Appropriate and robust behavioral control in a noisy environment is important for the survival of most organisms. Understanding such robust behavioral control has been an attractive subject in neuroscience research. Here, we investigated the processing of wide-field motion with random dot noise at both the behavioral and neuronal level in Drosophila melanogaster. We measured the head yaw optomotor response (OMR) and the activity of motion-sensitive neurons, horizontal system (HS) cells, with in vivo whole-cell patch clamp recordings at various levels of noise intensity. We found that flies had a robust sensation of motion direction under noisy conditions, while membrane potential changes of HS cells were not correlated with behavioral responses. By applying signal classification theory to the distributions of HS cell responses, however, we found that motion direction under noise can be clearly discriminated by HS cells, and that this discrimination performance was quantitatively similar to that of OMR. Furthermore, we successfully reproduced HS cell activity in response to noisy motion stimuli with a local motion detector model including a spatial filter and threshold function. This study provides evidence for the physiological basis of noise-robust behavior in a tiny insect brain.

Sexual experience enhances Drosophila melanogaster male mating behavior and success

Competition for mates is a wide-spread phenomenon affecting individual reproductive success. The ability of animals to adjust their behaviors in response to changing social environment is important and well documented. Drosophila melanogaster males compete with one another for matings with females and modify their reproductive behaviors based on prior social interactions. However, it remains to be determined how male social experience that culminates in mating with a female impacts subsequent male reproductive behaviors and mating success. Here we show that sexual experience enhances future mating success. Previously mated D. melanogaster males adjust their courtship behaviors and out-compete sexually inexperienced males for copulations. Interestingly, courtship experience alone is not sufficient in providing this competitive advantage, indicating that copulation plays a role in reinforcing this social learning. We also show that females use their sense of hearing to preferentially mate with experienced males when given a choice. Our results demonstrate the ability of previously mated males to learn from their positive sexual experiences and adjust their behaviors to gain a mating advantage. These experienced-based changes in behavior reveal strategies that animals likely use to increase their fecundity in natural competitive environments.