To evaluate the sound localization ability of patients with different degrees of unilateral conductive hearing loss (UCHL) in quiet and noisy environments, and to explore the changes and characteristics of sound localization. This was a cross-sectional study. 41 patients with UCHL were hospitalized in Shandong Provincial ENT Hospital from January to April 2024, including 22 males and 19 females, aged 18-55 years old, with an average age of 36.9 years. According to the pure-tone average (PTA) of 500, 1000 and 2000 Hz in the suffered ear, subjects were divided into slight-mild UCHL group (20 numbers) and moderate-moderately severe UCHL group (21 numbers). 21 patients with normal hearing (NH) were enrolled as controls. All subjects were assessed through pure-tone audiometry, horizontal sound localization test (including azimuth identification test in quiet and noisy environments), Chinese edition short form of Spatial Hearing Questionnaire (C-SHQ12) and twelve-item version of Speech, Spatial, and Qualities of Hearing Scale (SSQ12). SPSS, version 26.0, was used for statistical analysis. There were significant differences in the root-mean-square errors (RMSE) of the sound localization azimuth identification test in quiet and noisy environments among the NH group, slight-mild UCHL group, and moderate-moderately severe UCHL group (Quiet: =29.109, <0.001; Noisy: =24.351, <0.001). This presented statistically marked difference in the RMSEs between the two listening environments in the slight-mild UCHL group (=-4.911, <0.001). There was a statistical difference in the RMSEs between the normal and affected sides of the subjects in the slight-mild UCHL group in the quiet environment (=-2.055,<0.05), but not in the noisy environment. For moderate-moderately severe UCHL subjects, there were no differences in the RMSEs between the quiet and noisy environments (>0.05). What's more，no significant differences were found between normal side and affected side in both environments (>0.05). The RMSEs of UCHL patients in quiet and noisy environments were positively correlated with PTA of air-conduction in the suffered ears (Quiet: =0.681, <0.001; Noisy: =0.346, <0.05). RMSEs in quiet and noisy environments were negatively correlated with the average localization scores in C-SHQ12 (Quiet: =-0.576, <0.001, Noisy: =-0.613, <0.001) and in SSQ12 (Quiet: =-0.634, <0.001, Noisy: =-0.663, <0.001). The sound localization ability of UCHL subjects decreased compared with those with normal hearing, and the RMSE gradually increased with the worse of air conduction hearing threshold. The localization ability of UCHL subjects was further reduced in the noisy environment compared with that in the quiet environment. The slight-mild UCHL subjects had better localization performance in the normal ears while worse in the suffered ears, however, when they were in noisy environment or their hearing loss deteriorated, the localization advantage of the normal ears was no longer obvious, and both sides of the subjects presented poor localization performance.