Dopaminergic control of preference for reward maximization and reward unpredictability during free choice.

为了提高它们的生存机会，动物必须最大限度地摄取食物(能量增加)。然而，在某些情况下，哺乳动物和鸟类似乎采取了一种次优的寻找策略，更喜欢提供较低报酬率的不可预测的食物选择。这种非理性选择的原因尚不清楚，特别是因为所使用的大多数实验程序没有尝试分别研究奖励最大化和奖励不可预测性的影响。例如，将固定进度表和可变进度表的表现与相等的平均值(如经常报道的那样)进行比较，几乎没有关于动物动机的信息。它优先选择可变时间表是因为对不可预测性感兴趣，还是因为食物可以更快获得，或者因为这两个组成部分的结合？事实上，在寻找食物时，最大化和不可预测性可能被结合在一起，就像在人类赌博中一样：个体试图最大化收益，但不可预测性是保持他们对这项任务的兴趣所必需的。在这个项目中，我建议了五个实验，它们的目的有两个：(1)了解食物最大化和食物不可预测性在多大程度上可以控制自由选择任务中的行为选择；(2)展示多巴胺如何决定对奖励最大化和奖励不可预测性的敏感性。事实上，多巴胺是一种神经递质，主要由奖赏和条件线索招募，但也对线索-奖赏关联的不可预测性非常敏感。我计划用鸽子作为动物模型，因为大量文献都致力于鸽子的选择过程，而与此同时，鸟类的相关神经生物学机制却知之甚少。因此，本申请的结果不仅将为本研究产生重要的结果，而且还可以提供关于脊椎动物类别中与奖励相关的行为的不变机制的见解。

In order to improve their chance of survival, animals must maximise food intake (energy gain). However, in some circumstances, mammals and birds seem to adopt a suboptimal seeking strategy, preferring an unpredictable food option which provides a lower reward rate. The reasons for this irrational choice are not clearly established, notably because most experimental procedures used have not attempted to study the effects of reward maximisation and of reward unpredictability separately. For example, comparing performance on a fixed schedule and a variable schedule with an equivalent mean (as often reported) brings little information about the motivation of the animal. Does it preferentially choose the variable schedule because intrigued by unpredictability or because food can be obtained sooner or because of a combination of these two components? In fact, maximisation and unpredictability might be combined in food seeking, as they are in human gambling: the individuals try to maximize gains, but unpredictability is necessary to hold their interest in the task. In this project, I suggest five experiments whose aim is twofold: (1) understanding the extent to which food maximization and food unpredictability can control behavioural choices in a free-choice task, and (2) showing how dopamine determines sensitivity to reward maximization and reward unpredictability. Indeed, dopamine is a neurotransmitter chiefly recruited by rewards and conditioned cues, but also very sensitive to the unpredictability of cue-reward associations. I plan to use pigeons as animal model since a large body of literature is devoted to processes of choice in pigeons while at the same time the relevant neurobiological mechanisms are less understood in birds. Thus, the results of the present application would not only generate important results for the present study but could also provide insights about invariant mechanisms of reward-associated behaviour across vertebrate classes.