There has been a limited number of pioneering eye-tracking studies on the way cognitive loading varies with different translation tasks; therefore, much still needs to be learned regarding (1) whether ‘eye tracking’ can be adequately applied to the study of the effect of directionality on cognitive loading at a textual level; and (2) whether the predictions suggested by the Revised Hierarchical Model (Kroll and Stewart 1994) regarding ‘translation asymmetry’ are valid at a textual level.