Resting state connectivity differences in eyes open versus eyes closed conditions

Abstract

Functional magnetic resonance imaging data are commonly collected during the resting state. Resting state functional magnetic resonance imaging (rs-fMRI) is very practical and applicable for a wide range of study populations. Rs-fMRI is usually collected in at least one of three different conditions/tasks, eyes closed (EC), eyes open (EO), or eyes fixated on an object (EO-F). Several studies have shown that there are significant condition-related differences in the acquired data. In this study, we compared the functional network connectivity (FNC) differences assessed via group independent component analysis on a large rs-fMRI dataset collected in both EC and EO-F conditions, and also investigated the effect of covariates (e.g., age, gender, and social status score). Our results indicated that task condition significantly affected a wide range of networks; connectivity of visual networks to themselves and other networks was increased during EO-F, while EC was associated with increased connectivity of auditory and sensorimotor networks to other networks. In addition, the association of FNC with age, gender, and social status was observed to be significant only in the EO-F condition (though limited as well). However, statistical analysis did not reveal any significant effect of interaction between eyes status and covariates. These results indicate that resting-state condition is an important variable that may limit the generalizability of clinical findings using rs-fMRI.

Keywords: eyes closed; eyes open; functional network connectivity; independent component analysis; resting state fMRI.

Figure 1

51 resting state components are presented as subgroups based on their anatomical and functional properties, including 4 sub‐cortical (SC), 3 auditory (Aud), 8 sensorimotor (SM), 18 visual (Vis), 4 default‐mode networks (DMN), 12 cognitive control (CC), and 2 cerebellum (Cb) [Color figure can be viewed at http://wileyonlinelibrary.com]

Figure 2

The FNC matrix for all participants, displayed as one sample t‐statistics, thresholded with 0.01 levels FDR. The eyes open––fixated condition is shown on the left and the eyes closed condition is on the right [Color figure can be viewed at http://wileyonlinelibrary.com]

Figure 3

Paired t‐test results for EO‐F and EC FNCs, displayed as −sign(t‐statistics)*log10(p‐value) and on the right 0.01 levels FDR survivors. Blue color shows the regions that have higher correlation in EO‐F case and red color shows the regions that have higher correlation in EC case [Color figure can be viewed at http://wileyonlinelibrary.com]

Figure 4

Regression results of the age effect, top figures are for the EO‐F and bottom figures for the EC condition. The results are displayed as −sign(beta)*log10(p‐value). The FDR‐significant results (0.05) are shown to the right. Blue colors show the regions that have lower correlation with increasing age and red colors show the regions that have higher correlation with increasing age [Color figure can be viewed at http://wileyonlinelibrary.com]

Figure 5

Regression results of the gender effect, top figures are for the EO‐F and bottom figures for the EC condition. The results are displayed as −sign(beta)*log10(p‐value). The FDR‐significant results (0.05) are shown to the right. Blue colors show the regions that have higher correlation in females and red colors show the regions that have higher correlation in males [Color figure can be viewed at http://wileyonlinelibrary.com]

Figure 6

Regression results of the social status score effect, top figures are for the EO‐F and bottom figures for the EC case. The results are displayed as −sign(beta)*log10(p‐value). FDR‐significant results (0.05) are shown to the right. Blue colors show the regions that have lower correlation with increasing social status score and red colors show the regions that have higher correlation with increasing social status score [Color figure can be viewed at http://wileyonlinelibrary.com]