Interhemispheric transfer (IT) of visuomotor information can be assessed by measuring the crossed-uncrossed difference (CUD) in the Poffenberger paradigm [14]. We report a series of experiments aimed at casting light on two aspects of IT: the importance of motor response variables and the nature of the left-right asymmetries found in a recent meta-analysis of studies using the Poffenberger paradigm [10]. In Experiments 1 and 2 we found no effect of varying the finger used for response while there was a larger CUD when using the left hand in comparison to the right hand. In Experiment 3, which employed visual stimuli different from the two preceding experiments, CUD asymmetries were related to hemifields rather than to hands with a CUD larger for the right than for the left visual hemifield. These findings suggests that both motor and visual variables are important for CUD asymmetries to occur and are in keeping with a horse race model according to which overall speed of IT is determined by whichever signal (visual or pre-motor) is more rapid in crossing from one side to the other. As to the issue of lateral asymmetries in the CUD, the results of Expt. 3 have also shown that asymmetries arise only in tasks requiring an IT and therefore cannot depend upon additive hemispheric effects.