Computerized flash-fusion task
We modelled the flash-fusion task after the one used in Samaha & Postle and followed a similar protocol (Fig 1). Participants were seated ~60 cm away from a computer monitor with a 100 Hz refresh rate. The task consisted of white disk stimuli appearing against a black background. Each trial started with a white fixation cross in the middle of the screen, at 0.98 degrees visual angle. The fixation cross reduced in luminance before stimulus onset to prepare participants. The duration of this period was drawn from a uniform distribution between 1000 and 1500 ms. The fixation cross remained present for the duration of the trial. Disk stimuli, at 1.23 degrees visual angle, were presented either to the left or to the right side, at 2.45 degrees visual angle (0.20 cd/m2) of the fixation cross, with equal probability. Fifty percent of the trials were two-flash trials, in which a disk appeared for 40 ms, then disappeared for either 10, 20, 30, 40 or 50 ms, and then reappeared for 40 ms. The other fifty percent were one-flash trials, in which a disk appeared for durations that varied to match the stimulus onset to stimulus offset lengths for each stimulus type of the two-flash trials. 800 ms after stimulus offset the fixation cross increased in luminance to prompt participants to make a response. Participants held a mouse in both hands during the task and responded by clicking the left button with their left thumb to indicate when they perceived a single flash trial, or by clicking the right button with their right thumb to indicate when they perceived a two-flash trial. The next trial did not start until a response had been made and participants were instructed to focus on accuracy over speed. Participants completed 20 practice trials, followed by 600 test trials divided into 6 blocks. Participants took self-paced breaks in between each block.
Our task design deviated from Samaha & Postle’s design in three minor ways: due to the physical set-up of equipment in our lab, participants were seated at ~60 cm away from the monitor, whereas in the original design this was ~70 cm. Because the original task design did not specify any specific colour values for the stimulus disks, or the amount with which fixation luminance decreased before stimulus onset, we were unable to replicate these variables exactly. We therefore opted to use white stimulus disks to optimize contrast, and we decreased the luminance of the fixation cross by 40% before stimulus onset.
Figure 1: Task set-up and stimulus conditions . Reproduced from Samaha & Postle (2015). The image presented here has been altered from the original by removing a “confidence” step at the end of each trial, which was not analysed in the original study and therefore is not included here.
A: Depiction of a trial. The start of a trial was indicated with a white fixation cross, which reduced in luminance to prompt for an upcoming stimulus. Stimulus consisted of a single flash event in half of the trials and a double flash event in the other half. Flash events were presented either to the left or to the right of the fixation cross with equal probability.
B: Double flash events consisted of a white stimulus disk appearing for 40 ms, followed by a gap of 10 – 50 ms, after which the stimulus disk reappeared for another 40 ms. Lengths of single flash events were varied to match the duration of the double flash events.