Cueing a target location siginificantly improves identification of the target. The nature of this improvement is still under debate. In other studies, a masking has revealed the tuning of the filter selective for the target. In our study, we combined cueing with a masking paradigm to obtain a tuning function in the presence of a cue. Masking functions for orientation and for location were measured in the presence of location and orientation cues.

The test was a horizontal or vertical Gabor patch displayed in one of two locations on either side of fixation. The mask was a two-component ‘barcode’ patch (20% contrast) varying in orientation and location relative to the test. Attention was summoned through peripheral cues signaling the location/orientation of the test patch.

Interestingly, contrast thresholds in the absence of mask showed no cueing effect; thresholds were the same regardless of whether a cue was present. In the presence of a superimposed orientation mask, thresholds in the uncued condition were maximally elevated with a parallel mask. Thresholds decreased with increasing mask orientation from the test, following a Gaussian shape. The presence of a cue always reduced the masking effect but the shape of the function was cue-specific: the orientation cue improved thresholds within 5° of the cue orientation and showed a significant narrowing of the orientation tuning function whereas the location cue improved thresholds in a non-specific fashion (at all orientations).

When the mask location was varied with respect to the test, pilot data show a location cue narrowed the location tuning function. The increased selectivity obtained with an orientation or location cue is only evident if the mask varies in the cued dimension, suggesting that the cue allows the observer to use a better-matched template.