Suppr超能文献

一种将注意力缺陷多动障碍(ADHD)归因于强化学习过程中多巴胺相位/张力失衡的机制模型。

A mechanistic model of ADHD as resulting from dopamine phasic/tonic imbalance during reinforcement learning.

作者信息

Véronneau-Veilleux Florence, Robaey Philippe, Ursino Mauro, Nekka Fahima

机构信息

Faculté de Pharmacie, Université de Montréal, Montreal, QC, Canada.

Children's Hospital of Eastern Ontario, University of Ottawa, Ottawa, ON, Canada.

出版信息

Front Comput Neurosci. 2022 Jul 18;16:849323. doi: 10.3389/fncom.2022.849323. eCollection 2022.

Abstract

Attention deficit hyperactivity disorder (ADHD) is the most common neurodevelopmental disorder in children. Although the involvement of dopamine in this disorder seems to be established, the nature of dopaminergic dysfunction remains controversial. The purpose of this study was to test whether the key response characteristics of ADHD could be simulated by a mechanistic model that combines a decrease in tonic dopaminergic activity with an increase in phasic responses in cortical-striatal loops during learning reinforcement. To this end, we combined a dynamic model of dopamine with a neurocomputational model of the basal ganglia with multiple action channels. We also included a dynamic model of tonic and phasic dopamine release and control, and a learning procedure driven by tonic and phasic dopamine levels. In the model, the dopamine imbalance is the result of impaired presynaptic regulation of dopamine at the terminal level. Using this model, virtual individuals from a dopamine imbalance group and a control group were trained to associate four stimuli with four actions with fully informative reinforcement feedback. In a second phase, they were tested without feedback. Subjects in the dopamine imbalance group showed poorer performance with more variable reaction times due to the presence of fast and very slow responses, difficulty in choosing between stimuli even when they were of high intensity, and greater sensitivity to noise. Learning history was also significantly more variable in the dopamine imbalance group, explaining 75% of the variability in reaction time using quadratic regression. The response profile of the virtual subjects varied as a function of the learning history variability index to produce increasingly severe impairment, beginning with an increase in response variability alone, then accumulating a decrease in performance and finally a learning deficit. Although ADHD is certainly a heterogeneous disorder, these results suggest that typical features of ADHD can be explained by a phasic/tonic imbalance in dopaminergic activity alone.

摘要

注意缺陷多动障碍(ADHD)是儿童中最常见的神经发育障碍。尽管多巴胺与该障碍的关联似乎已得到证实,但多巴胺能功能障碍的本质仍存在争议。本研究的目的是测试ADHD的关键反应特征是否可以通过一个机制模型来模拟,该模型将紧张性多巴胺能活动的降低与学习强化过程中皮质-纹状体回路的相位反应增加相结合。为此,我们将多巴胺的动态模型与具有多个动作通道的基底神经节神经计算模型相结合。我们还纳入了紧张性和相位性多巴胺释放及控制的动态模型,以及由紧张性和相位性多巴胺水平驱动的学习过程。在该模型中,多巴胺失衡是多巴胺在终末水平的突触前调节受损的结果。使用这个模型,对来自多巴胺失衡组和对照组的虚拟个体进行训练,使其将四种刺激与四种动作进行关联,并给予完全信息性的强化反馈。在第二阶段,对他们进行无反馈测试。多巴胺失衡组的受试者表现较差,反应时间更具变异性,原因是存在快速和非常缓慢的反应、即使在高强度刺激时也难以在刺激之间做出选择以及对噪声更敏感。多巴胺失衡组的学习历史也明显更具变异性,使用二次回归解释了反应时间变异性的75%。虚拟受试者的反应特征随学习历史变异性指数而变化,从而产生越来越严重的损害,首先是仅反应变异性增加,然后累积表现下降,最后出现学习缺陷。尽管ADHD无疑是一种异质性疾病,但这些结果表明,ADHD的典型特征仅通过多巴胺能活动的相位性/紧张性失衡就可以得到解释。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b58e/9342605/64ecfe24bcdf/fncom-16-849323-g0001.jpg

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验