Dixon Matthew L, Andrews-Hanna Jessica R, Spreng R Nathan, Irving Zachary C, Mills Caitlin, Girn Manesh, Christoff Kalina
Department of Psychology, University of British Columbia, Vancouver, British Columbia, Canada.
Institute of Cognitive Science, University of Colorado Boulder, Boulder, Colorado, USA.
Neuroimage. 2017 Feb 15;147:632-649. doi: 10.1016/j.neuroimage.2016.12.073. Epub 2016 Dec 28.
Anticorrelation between the default network (DN) and dorsal attention network (DAN) is thought to be an intrinsic aspect of functional brain organization reflecting competing functions. However, the effect size of functional connectivity (FC) between the DN and DAN has yet to be established. Furthermore, the stability of anticorrelations across distinct DN subsystems, different contexts, and time, remains unexplored. In study 1 we summarize effect sizes of DN-DAN FC from 20 studies, and in study 2 we probe the variability of DN-DAN interactions across six different cognitive states in a new data set. We show that: (i) the DN and DAN have an independent rather than anticorrelated relationship when global signal regression is not used (median effect size across studies: r=-.06; 95% CI: -.15 to .08); (ii) the DAN exhibits weak negative FC with the DN Core subsystem but is uncorrelated with the dorsomedial prefrontal and medial temporal lobe subsystems; (iii) DN-DAN interactions vary significantly across different cognitive states; (iv) DN-DAN FC fluctuates across time between periods of anticorrelation and periods of positive correlation; and (v) changes across time in the strength of DN-DAN coupling are coordinated with interactions involving the frontoparietal control network (FPCN). Overall, the observed weak effect sizes related to DN-DAN anticorrelation suggest the need to re-conceptualize the nature of interactions between these networks. Furthermore, our findings demonstrate that DN-DAN interactions are not stable, but rather, exhibit substantial variability across time and context, and are coordinated with broader network dynamics involving the FPCN.
默认网络(DN)与背侧注意网络(DAN)之间的反相关被认为是反映竞争功能的功能性脑组织结构的一个内在方面。然而,DN与DAN之间功能连接(FC)的效应大小尚未确定。此外,不同DN子系统、不同情境和时间下反相关的稳定性仍未得到探索。在研究1中,我们总结了20项研究中DN-DAN FC的效应大小,在研究2中,我们在一个新的数据集中探究了DN-DAN相互作用在六种不同认知状态下的变异性。我们发现:(i)在不使用全局信号回归时,DN和DAN具有独立而非反相关的关系(各研究的中位数效应大小:r = -0.06;95%置信区间:-0.15至0.08);(ii)DAN与DN核心子系统表现出弱的负FC,但与背内侧前额叶和内侧颞叶子系统不相关;(iii)DN-DAN相互作用在不同认知状态下有显著差异;(iv)DN-DAN FC在反相关期和正相关期之间随时间波动;(v)DN-DAN耦合强度随时间的变化与涉及额顶控制网络(FPCN)的相互作用相协调。总体而言,观察到的与DN-DAN反相关相关的弱效应大小表明需要重新概念化这些网络之间相互作用的性质。此外,我们的研究结果表明,DN-DAN相互作用并不稳定,而是在时间和情境上表现出很大的变异性,并且与涉及FPCN的更广泛网络动态相协调。