The purpose of this study was to determine the optimal tube current setting and scanning mode for hominid fossil skull scanning, using multi-detector row computed tomography (CT). Four fossil skulls (La Ferrassie 1, Abri Pataud 1, CroMagnon 2 and Cro-Magnon 3) were examined by using the CT scanner LightSpeed 16 (General Electric Medical Systems) with varying dose per section (160, 250, and 300 mAs) and scanning mode (helical and conventional). Image quality of two-dimensional (2D) multiplanar reconstructions, three-dimensional (3D) reconstructions and native images was assessed by four reviewers using a four-point grading scale. An ANOVA (analysis of variance) model was used to compare the mean score for each sequence and the overall mean score according to the levels of the scanning parameters. Compared with helical CT (mean score=12.03), the conventional technique showed sustained poor image quality (mean score=4.17). With the helical mode, we observed a better image quality at 300 mAs than at 160 in the 3D sequences (P=0.03). Whereas in native images, a reduction in the effective tube current induced no degradation in image quality (P=0.05). Our study suggests a standardized protocol for fossil scanning with a 16 x 0.625 detector configuration, a 10 mm beam collimation, a 0.562:1 acquisition mode, a 0.625/0.4 mm slice thickness/reconstruction interval, a pitch of 5.62, 120 kV and 300 mAs especially when a 3D study is required.