Timing-dependent valence reversal: A sufficiency approach at the level of individual dopaminergic mushroom body input neurons

Obtaining rewards and avoiding punishment are powerful goals of behavior, in animals and man. To maximize reward and to minimize punishment, it is beneficial to learn predictors of their occurrence. These learning processes have been studied in great detail in many species including man. The complete ‘other side of the coin’ of the memories established for rewards or punishment is not usually acknowledged, however. That is, it is equally important to predict reward loss and punishment-relief! Indeed, delivering versus withdrawing rewards induces affect and memory of opposing valence: it feels good to receive but it feels bad to lose a reward. These experiences will result in appetitive and aversive learning, respectively, of associated cues. In turn, receiving punishment versus being relieved from it induce negative and positive affect and support aversive and appetitive learning. These fundamental dichotomies of how reinforcement is processed are observed in animals and man alike, and are referred to as timing-dependent valence reversal. Of note, such valence reversal is but poorly understood in terms of its neurobiological underpinnings, in any experimental system.Given the deeply conserved function of dopamine neurons in processing reinforcement throughout the animal kingdom and certainly including humans, we propose to exploit the unique experimental advantages of the numerically simple brain of Drosophila to study the role of single, identified dopaminergic neurons in how one and the same kind of experience can induce two opposing kinds of memory for stimuli preceding or following it. Towards this end, we combine high-resolution behavioral analyses with optogenetics and the recently published synaptic connectome of the learning and memory center of the Drosophila brain, the so-called mushroom body.