Mechanisms of sound sequence discrimination learning motivated by reward in the rat

我们以前曾报道过，声音序列辨别学习需要胆碱能输入到大鼠的听觉皮层（AC）。在这项研究中，奖励被用来激发大鼠的辨别行为。因此，多巴胺能输入介导的奖赏信号可能在学习中发挥重要作用。我们在本研究中测试了这种可能性。老鼠被训练来区分两个声音成分的序列，并且对两个序列中的一个的舔行为以水作为奖励。为了确定负责学习的多巴胺能输入，用6-羟基多巴胺（6-OHDA）局部注射损伤AC的多巴胺能传入。注射减弱了声音序列辨别学习，但对序列刺激的声音成分之间的辨别没有影响。延髓外侧核局部注射6-OHDA可减弱声音辨别学习。然而，不仅声音序列的辨别学习受损，而且声音成分的辨别学习也受损。SCH 23390（0.2mg/kg，i.p.），一种D1受体拮抗剂，对声音序列辨别学习没有影响，但它减弱了对陌生刺激的舔行为。氟哌啶醇（0.5 mg/kg，腹腔注射），D_2家族拮抗剂能减弱声音序列辨别学习，但对两种不同声音成分的辨别和舔音没有明显的抑制作用。这些结果表明，D2家族受体激活的多巴胺能输入到AC所需的声音序列辨别学习。

We have previously reported that sound sequence discrimination learning requires cholinergic inputs to the auditory cortex (AC) in rats. In that study, reward was used for motivating discrimination behavior in rats. Therefore, dopaminergic inputs mediating reward signals may have an important role in the learning. We tested the possibility in the present study. Rats were trained to discriminate sequences of two sound components, and licking behavior in response to one of the two sequences was rewarded with water. To identify the dopaminergic inputs responsible for the learning, dopaminergic afferents to the AC were lesioned with local injection of 6-hydroxydopamine (6-OHDA). The injection attenuated sound sequence discrimination learning, while it had no effect on discrimination between the sound components of the sequence stimuli. Local injection of 6-OHDA into the nucleus accumbens attenuated sound discrimination learning. However, not only discrimination learning of sound sequence but also that of the sound components were impaired. SCH23390 (0.2mg/kg, i.p.), a D1 receptor antagonist, had no effect on sound sequence discrimination learning, while it attenuated the licking behavior to unfamiliar stimuli. Haloperidol (0.5 mg/kg, i.p.), a D2 family antagonist, attenuated sound sequence discrimination learning, while it had no clear suppressive effect on discrimination of two different sound components and licking. These results suggest that D2 family receptors activated by dopaminergic inputs to the AC are required for sound sequence discrimination learning.