  |
 |
      |
The valid/invalid cue advantage in the Posner paradigm explained by ideal combination of information and without a change in quality of processing. We have been performing empirical experiments and model simulations of the ‘cueing task’, which has been one of the most influential paradigms in the study of visual attention (Posner, 1980). In this task, an observer must detect a signal at one of two locations, with a precue indicating the probable signal location. A ‘valid’ cue (signal at the cued location) typically leads to faster and more accurate responses, compared to an ‘invalid’ cue. Common explanations of the cueing effect hinge upon the concept of a limited attentional resource unevenly distributed across the visual field, causing an increase of perceptual quality at the attended (cued) location relative to the unattended (uncued) location. Early work by Kinchla looked to explain the valid cue advantage in terms of a weighted linear combination of information from the two locations (cued and uncued) rather than a change in processing at the cued location. We have extended such work by calculating performance of the optimal (ideal) observer. The ideal observer weights non-linearly the information at each location based upon the (a priori) probability of the signal appearing at that location. In the case of the cueing task, therefore, the information at the cued location is weighted more heavily. Our simulations have shown that this leads to a cueing effect in performance without proposing a change in perceptual quality at the cued and uncued locations. Thus, the ideal observer analysis presents a significant challenge to the general concept of attentional capacity limitations as the cause of the cueing effect in the cueing paradigm. In addition, we have used a new technique known as classification images to test whether attention changes the weighting of information or changes the tuning of the perceptual filters. Publications:
|
 |
| | viu home | ucsb psychology | contact us | |