Abstract
Background Compression of space is found for targets flashed just before saccades, but not when they are flashed in the dark. Recently it has been suggested that compression depends on the availability of visual references after saccades (Lappe, Atwater and Krekelberg, 2000). Compression also depends on the method of reporting position, under some conditions being found for verbal report but not for blind pointing (Burr, Ross and Morrone, 2001). Methods and Results Subjects made a 15° horizontal saccade. A blue bar was briefly presented before the saccade. Stimuli were displayed on a touch screen, viewed through a liquid crystal shutter that closed at a variable delay (from 20–320ms) after the start of saccades. Subjects were tested with or without a bright flash after the shutter closed. Closing the shutter removed visual references and, without the flash, rendered the scene dark. The flash was perceived as bright glare, providing light but not restoring visual references. Subjects indicated the perceived position of the bar by verbal report or by touching the screen with a hidden hand. In the flash conditions, strong compression was found both for verbal report and for blind pointing, even at the shortest shutter latencies. In the non-flash conditions compression was found for verbal report at most latencies and for pointing only at longer latencies. Conclusions Results from the flash experiment show that compression does not depend entirely on the availability of visual references. Results at different shutter delays without the flash show that the rapid onset of darkness is necessary for veridical pointing.
Acknowledgements: This work was supported by the Human Frontiers Science Foundation, the Australian Research Council and the Italian Ministry of Universities and Research