Investigation of neural mechanisms of real-time functional MRI neurofeedback (rtfMRI-nf) training requires an efficient study control approach. A common rtfMRI-nf study design involves an experimental group, receiving active rtfMRI-nf, and a control group, provided with sham rtfMRI-nf. We report the first study in which rtfMRI-nf procedure is controlled through counterbalancing training runs with active and sham rtfMRI-nf for each participant. Healthy volunteers ( = 18) used rtfMRI-nf to upregulate fMRI activity of an individually defined target region in the left dorsolateral prefrontal cortex (DLPFC) while performing tasks that involved mental generation of a random numerical sequence and serial summation of numbers in the sequence. Sham rtfMRI-nf was provided based on fMRI activity of a different brain region, not involved in these tasks. The experimental procedure included two training runs with the active rtfMRI-nf and two runs with the sham rtfMRI-nf, in a randomized order. The participants achieved significantly higher fMRI activation of the left DLPFC target region during the active rtfMRI-nf conditions compared to the sham rtfMRI-nf conditions. fMRI functional connectivity of the left DLPFC target region with the nodes of the central executive network was significantly enhanced during the active rtfMRI-nf conditions relative to the sham conditions. fMRI connectivity of the target region with the nodes of the default mode network was similarly enhanced. fMRI connectivity changes between the active and sham conditions exhibited meaningful associations with individual performance measures on the Working Memory Multimodal Attention Task, the Approach-Avoidance Task, and the Trail Making Test. Our results demonstrate that the counterbalanced active-sham study design can be efficiently used to investigate mechanisms of active rtfMRI-nf in direct comparison to those of sham rtfMRI-nf. Further studies with larger group sizes are needed to confirm the reported findings and evaluate clinical utility of this study control approach.

中文翻译：

---

使用平衡主动假研究设计验证用于认知训练的实时功能磁共振成像神经反馈程序

实时功能 MRI 神经反馈 (rtfMRI-nf) 训练的神经机制研究需要有效的研究控制方法。常见的 rtfMRI-nf 研究设计包括接受主动 rtfMRI-nf 的实验组和接受假 rtfMRI-nf 的对照组。我们报告了第一项研究，其中通过对每个参与者进行主动和假 rtfMRI-nf 的平衡训练来控制 rtfMRI-nf 程序。健康志愿者 (= 18) 使用 rtfMRI-nf 上调左背外侧前额皮质 (DLPFC) 中单独定义的目标区域的 fMRI 活动，同时执行涉及随机数字序列的心理生成和序列中数字的连续求和的任务。 Sham rtfMRI-nf 是根据不同大脑区域的 fMRI 活动提供的，不参与这些任务。实验程序包括以随机顺序使用主动 rtfMRI-nf 进行两次训练运行和使用假 rtfMRI-nf 进行两次训练。与假 rtfMRI-nf 条件相比，参与者在活动 rtfMRI-nf 条件下左侧 DLPFC 目标区域的 fMRI 激活明显更高。相对于假条件，在活动 rtfMRI-nf 条件下，左侧 DLPFC 目标区域与中央执行网络节点的 fMRI 功能连接显着增强。目标区域与默认模式网络节点的 fMRI 连接性也得到了类似的增强。活动条件和假条件之间的功能磁共振成像连接变化与工作记忆多模式注意力任务、接近回避任务和轨迹制作测试的个人表现测量表现出有意义的关联。我们的结果表明，平衡主动假研究设计可以有效地用于研究主动 rtfMRI-nf 的机制，并与假 rtfMRI-nf 直接比较。需要进行更大组规模的进一步研究来证实报告的结果并评估该研究控制方法的临床效用。