Brázdil Milan, Mikl Michal, Marecek Radek, Krupa Petr, Rektor Ivan
1st Department of Neurology, St. Anne's University Hospital, Brno, Czech Republic.
Neuroimage. 2007 Apr 1;35(2):827-35. doi: 10.1016/j.neuroimage.2006.12.020. Epub 2006 Dec 23.
To investigate the fundamental connectivity architecture of neural structures involved in the goal-directed processing of target events.
Twenty healthy volunteers underwent event-related functional magnetic resonance imaging (fMRI) while performing a standard oddball task. In the task, two types of visual stimuli - rare (target) and frequent - were randomly presented, and subjects were instructed to mentally count the target stimuli. Dynamic causal modeling (DCM), in combination with Bayes factors was used to compare competing neurophysiological models with different intrinsic connectivity structures and input regions within the network of brain regions underlying target stimulus processing.
Conventional analysis of fMRI data revealed significantly greater activation in response to the target stimuli (in comparison to the frequent stimuli) in several brain regions, including the intraparietal sulci and supramarginal gyri, the anterior and posterior cingulate gyri, the inferior and middle frontal gyri, the superior temporal sulcus, the precuneus/cuneus, and the subcortical grey matter (caudate and thalamus). The most extensive cortical activations were found in the right intraparietal sulcus (IPS), the anterior cingulate cortex (ACC), and the right lateral prefrontal cortex (PFC). These three regions were entered into the DCM. A comparison on a group level revealed that the dynamic causal models in which the ACC and alternatively the IPS served as input regions were superior to a model in which the PFC was assumed to receive external inputs. No significant difference was observed between the fully connected models with ACC and IPS as input regions. Subsequent analysis of the intrinsic connectivity within two investigated models (IPS and ACC) disclosed significant parallel forward connections from the IPS to the frontal areas and from the ACC to the PFC and the IPS.
Our findings indicate that during target stimulus processing there is a bidirectional frontoparietal information flow, very likely reflecting parallel activation of two distinct but partially overlapping attentional or attentional/event-encoding neural systems. Additionally, a simple hierarchy within the right frontal lobe is suggested with the ACC exerting influence over the PFC.
研究参与目标事件目标导向处理的神经结构的基本连接架构。
20名健康志愿者在执行标准oddball任务时接受了事件相关功能磁共振成像(fMRI)。在该任务中,随机呈现两种类型的视觉刺激——罕见(目标)刺激和频繁刺激,受试者被要求在脑海中计数目标刺激。动态因果模型(DCM)结合贝叶斯因子,用于比较具有不同内在连接结构和目标刺激处理基础脑区网络内输入区域的竞争神经生理模型。
fMRI数据的传统分析显示,与频繁刺激相比,包括顶内沟和缘上回、前扣带回和后扣带回、额下回和额中回、颞上沟、楔前叶/楔叶以及皮质下灰质(尾状核和丘脑)在内的几个脑区对目标刺激的激活明显更强。最广泛的皮质激活出现在右侧顶内沟(IPS)、前扣带回皮质(ACC)和右侧外侧前额叶皮质(PFC)。这三个区域被纳入DCM。在组水平上的比较显示,以ACC和交替以IPS作为输入区域的动态因果模型优于假设PFC接收外部输入的模型。以ACC和IPS作为输入区域的全连接模型之间未观察到显著差异。随后对两个研究模型(IPS和ACC)内的内在连接进行分析,发现从IPS到额叶区域以及从ACC到PFC和IPS存在显著的平行向前连接。
我们的研究结果表明,在目标刺激处理过程中存在双向额顶叶信息流,很可能反映了两个不同但部分重叠的注意力或注意力/事件编码神经系统的平行激活。此外,提示右侧额叶内存在一个简单的层级结构,ACC对PFC施加影响。