*2.1. AG Improved TBI- and CMS-Induced Spatial Learning Deficit*

The effect of AG on spatial learning and memory was measured using the Morris water maze (Figure 1). During the five-day training, TBI impaired spatial learning. Supplementation of AG improved the TBI-induced deficit in spatial learning (Figure 1A). CMS impaired spatial learning (control vs. CMS), which improved following AG supplementation (CMS vs. CMS + AG) (Figure 1B). The sucrose preference of the CMS mice gradually decreased over six weeks. This declining preference meant that the mice developed an anhedonia-like tendency. AG did not affect this anhedonia-like behavior (Figure S1).

**Figure 1.** Spatial learning by *Angelica gigas* (AG) in traumatic brain injury (TBI) and chronic mild stress (CMS) mice as measured by the latency to the platform in the Morris water maze (MWM). (**A**) TBI model. There were repetitive training (within) effects [F(4,104) = 13.9, *p* < 0.001] and treatment group (between) effects [F(3,26) = 2.27, *p* = 0.044], but with non-significant within–between interaction differences [F(12,104) = 1.1, *p* = 0.36]. The post-hoc pairwise comparison showed a difference between control vs. TBI (*p* = 0.033), TBI vs. TBI + AG (*p* = 0.041). (**B**) CMS model. There were repetitive training effects [F(4,196) = 31.3, *p* < 0.001], treatment group effects [F(3,49) = 3.1, *p* = 0.034], and within–between interactions [F(12,196) = 2.2, *p* = 0.012]. Post-hoc pairwise comparison revealed a difference between the control and CMS groups (*p* = 0.019), AG vs. CMS (*p* = 0.01), CMS vs. CMS + AG (*p* = 0.042). All data were normally distributed and are represented as means ± S.E.M. Control: vehicle (DW) treated; AG: *Angelica gigas* 1 mg/kg; TBI: vehicle treated + traumatic brain injury; TBI + AG: *Angelica gigas* 1 mg/kg + traumatic brain injury; CMS: vehicle treated + chronic mild stress; CMS + AG: *Angelica gigas* 1 mg/kg + chronic mild stress. Repeated measure ANOVA, Tukey's HSD post-hoc test.
