Perceptual attention in healthy participants is characterized by two biases, one operating in the horizontal plane, which draws attention leftward, and the other operating in the vertical plane, which draws attention upward. Given that these biases are reliably found in the same individual, and appear similar at a surface level, a number of researchers have investigated the relationship between horizontal and vertical attentional biases. To date, these investigations have failed to find an association, and this may be due to the fact that one-dimensional vertical and horizontal stimuli were presented separately rather than being measured from a single, two-dimensional stimulus. Across three experiments, two dimensional stimuli were presented, and participants marked the centre of the stimuli. In addition, the shapes of the stimuli were manipulated to determine whether this produced the same modulation of the two biases. Across 13 stimuli and three experiments there were no correlations between the vertical and horizontal biases. In addition, manipulations of stimulus shape, which affected biases in one dimension, did not affect biases in the other dimension. There were, however, consistent correlations between the degree of bias within each dimension across the different stimuli. This study has produced converging evidence that horizontal and vertical biases in spatial judgments rely on separate cognitive mechanisms. To account for these results we discuss a model whereby horizontal asymmetries rely more on space-based mechanisms whereas vertical asymmetries rely more on object-based mechanisms.