In their Postscript, Jans et al. emphasize the similarities between their criteria and ours for demonstrating split attention. However, they insist on starting with the unified window of attention as the null hypothesis. In evaluating the validity of this assumption, it is worth considering both the capacity and flexibility that would be needed for split attention. First, do recognition mechanisms have enough capacity to handle simultaneous input from multiple objects? If not, then attention can probably never be split. If, however, useful information can be extracted from two objects simultaneously, then it is relevant to ask whether attentional control is flexible enough to select two locations without selecting the region between them.