Eighty blindfolded healthy female subjects participated in an active and a passive straight-ahead pointing task to study the estimation of the subjective sagittal middle in the presence or absence of an active haptic exploration. Subjects were to point straight-ahead with their left or right index finger starting from different right- or left-sided locations, while performing either an active movement or with the hand being passively moved by the examiner. Results showed: (i) a significant effect of the hand used only on the active straight-ahead pointing task performance; (ii) a significant effect of the starting position both on the active and passive pointing performance; (iii) a significant correlation between the two protocols only in terms of spatial bias (algebraic error) but not in terms of precision (absolute errors). These results are discussed with regards to normal and neuropsychological studies of the egocentric frame of reference.