Nguyen, L & Spehar, B. Visual adaptation to natural scene statistics and visual preference. Vision Research - Recalibrating the visual system (Special issue)

Published: 12 December 2020| Version 1 | DOI: 10.17632/nzmh4kk5zd.1
, Branka Spehar


Stimuli and data from Nguyen & Spehar - Visual adaptation to natural scene statistics and visual preference. Visual preference was measured using a two alternative forced choice (2AFC) paired comparison procedure. On each trial, two images were presented side by side and the task was to indicate, via keypress, which of the two images was visually preferred. In our experiment visual preference was measured for synthetic noise images with spectral slope alpha of 0.25, 0.75, 1.25, 1.75 and 2.25. Each level of alpha was paired with every other level resulting in a total of 20 unique pairs for one experimental sequence. Within each experimental sequence every alpha level was shown eight times in total; four times as the left image in a pair and four times as the right image in a pair. There were six blocks in total: three baseline blocks and three adaptation blocks. These blocks alternated such that a baseline block always preceded an adaptation block. In other words, the blocks were split into three different baseline-adaptation pairs. The three baseline blocks consisted of trials in which the five alpha levels (0.25, 0.75, 1.25, 1.75 and 2.25) were paired with each other as described above. The two images in any given pair had the same Fourier phase but the Fourier phase varied randomly between different pairs. The three adaptation blocks were the same as the baseline blocks except that each adaptation block was preceded by an adaptation period of 150 seconds to one of the three adapting alpha levels:, alpha = 0.25, 1.25 and 2.25. The order of different adaptation conditions (0.25, 1.25 or 2.25) was randomised across participants. In total, each participant underwent six blocks. For the majority of participants (40/54), each of the three baseline and three adaptation blocks consisted of 40 trials in total (20 unique pairings repeated twice), resulting in 240 trials in total. For the remaining 18 participants, each of the baseline and adaptation blocks contained 20 trials, resulting in 120 experimental trials in total. The summary data file shows the proportion by which experimental stimuli with each of the five levels of alpha were chosen in each of the three baseline and three adaptation conditions. Data for each participant are shown in separate rows and different baseline and adaptation conditions have been labelled by the adapting level of alpha and also colour coded. The raw data files for each participants can be found in the Raw_Data folder.



University of New South Wales


Adaptation, Preference