**3. Results**

#### *3.1. Demographics and Neuropsychological Assessment*

Our analysis showed that there were no significant differences in age (t31 = −0.952, *<sup>p</sup>* > 0.05), education level (t31 = −1.297, *<sup>p</sup>* > 0.05) or sex (χ<sup>2</sup> <sup>1</sup> = 0.243, *p* > 0.05) between groups. The mTBI patients presented significantly worse insomnia severity and headache compared to HCs (all *p* < 0.01). In addition, patients exhibited impaired information processing speed (reflected by the DSC task) compared with HCs (*p* < 0.001, Table 1). All mTBI patients exhibited the same injury severity with a Glasgow Coma Score of 15, loss of consciousness < 30 min, and post-traumatic amnesia < 24 h. The most common cause of injuries was acceleration/deceleration caused by traffic accidents (6/16, 37.5%), followed by falls (5/16, 31.2%), assaults (3/16, 18.8%), and others (2/16, 12.5%).


**Table 1.** Demographic characteristics and neuropsychological measures in mTBI patients and HCs.

Values presented as Mean ± SD unless otherwise stated. mTBI, mild traumatic brain injury; HCs, healthy controls; GCS, Glasgow Coma Score; TMT-A, Trail-Making Test Part-A; FDS, Forward Digit Span; BDS, Backward Digit Span; DSC, Digit Symbol Coding; ISI, the Insomnia Severity Index; HIT, the short-form Headache Impact Test.

#### *3.2. fMRI Behavioral Performance*

Compared with HCs, patients with mTBI exhibited significantly longer RT in the TS condition [F(1,31) =4.247, *p* = 0.048], reflecting impaired cognitive flexibility on a behavioral level (Figure 2A). Furthermore, mTBI patients were significantly less accurate than HCs across all conditions (*p* < 0.05, Figure 2B). The detailed information about fMRI behavioral performance is described in the Supplementary Materials.
