Figure 5.

Ruling out possible confounding factors. A shows a sketch of “same destination” comparisons (red frames), demonstrating that such trials share the same component of movement (upward in the top panel and rightward in the bottom one). Higher similarity between such comparisons, as shown in the mIPS, may therefore be simply attributable to a broad tuning to the direction of movement rather that to the spatiotopic location of the target. Importantly, “same origin” comparisons (purple frames) also have the same components of movement. The fact that, in the mIPS, “same destination” comparisons yield significantly higher similarity values than “same origin” comparisons (see Fig. 4C) rules out this possible confound. B describes the matrix of all possible comparisons. Comparisons of saccades with opposite laterality (one to the left of fixation and the other to the right) are discarded (marked in X). C and D show the similarity indices in the FEF and mIPS using the new (stricter) inclusion criteria (conventions as in Fig. 4). The FEF shows a retinotopic coding pattern (i.e., vector encoding), whereas the mIPS shows sensitivity to both the vector of movement and its destination with respect to the head.