Changes in default mode network as automaticity develops in a categorization task

Farzin Shamloo, Purdue University


The default mode network (DMN) is a set of brain regions in which blood oxygen level dependent (BOLD) signal is suppressed during attentional focus on the external environment. Because automatic task processing requires less attention, development of automaticity in a rule–based categorization task may result in less deactivation of the DMN as compared to the learning stage. Furthermore, previous work suggests tha DMN changes its functional connectivity with other brain networks depending on task (Spreng et al., 2010). Hence we hypothesize that the functional connectivity of the DMN may change as automaticity develops. We tested these hypotheses by reanalyzing the functional magnetic resonance imaging (fMRI) data of 15 participants who were each trained in rule-based categorization for 20 sessions on consecutive workdays (Helie et al., 2010a). Each participant was scanned on his or her 1st, 4th, 10th, and 20th session. The results show deactivation of some DMN regions in sessions 1, 4 and 10, but not in session 20. In addition, analysis of variance shows a statistically significant decrease in the deactivation of four DMN regions between sessions 1 and 20, suggesting that automatic rule-based categorization does not inhibit DMN regions as much as rule-based category learning. These results provide preliminary evidence that DMN inhibition is reduced when the rule-based categorization task becomes more automatic. This is in line with the hypothesis that automatic task processing requires less attentional focus. We also investigated changes in functional connectivity pattern by running three seed-based coherence analyses with precuneus (important DMN region), medial prefrontal cortex (important DMN region) and premotor cortex (important in automatic categorization) as seed regions. The results from analyses with precuneus and MPFC as seed regions show that both DMN and non-DMN regions increase their functional connectivity with these two important DMN regions after automaticity developed. Some of the non-DMN regions are involved in taskprocessing. Likewise, functional connectivity with premotor cortex as the seed shows that the left inferior parietal lobule (a DMN region) increases its functional connectivity with premotor cortex. These results suggest that communication between DMN and task-related regions becomes more efficient with extensive practice. No region decreased its functional connectivity with precuneus or premotor cortex. However, as categorization became more automatic, coherence between MPFC and ventromedial prefrontal cortex decreased. These results bridge the cognitive and neuroscientific conceptions of automaticity in showing that the reduced need for cognitive resources in automatic processing is accompanied by a disinhibition of the DMN and stronger functional connectivity between DMN and task-related brain regions.




Helie, Purdue University.

Subject Area


Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server