Dataset for "Colour crowding explained as adaptive spatial integration"

Description

Dataset for paper: Colour crowding explained as adaptive spatial integration Please refer to the readme file for a key to the variables included in Matlab files. Below an abstract of the work: Crowding is the inability to recognize an object in clutter, classically considered a fundamental low-level bottleneck to object recognition. Recently, however, it has been suggested that crowding, like predictive phenomena such as serial dependence, may result from optimizing strategies that exploit redundancies in natural scenes. This notion leads to several testable predictions, such as crowding being greater for non-salient targets and, counter-intuitively, that flanker interference should be associated with higher precision in judgements, leading to lower overall error rate. Here we measured colour discrimination for targets flanked by stimuli of variable colour. The results verified both predictions, showing that while crowding can affect object recognition, it may be better understood not as a processing bottleneck, but as a consequence of mechanisms evolved to efficiently exploit the spatial redundancies of the natural world. Analyses of reaction times of judgments shows that the integration occurs at sensory, rather than decisional levels.

Steps to reproduce

Seventeen participants were recruited for the experiments, eleven female and six male. Two were authors while all the others were naïve to the purposes of the experiment. All participants had normal colour vision (assessed by Ishihara Colour Blindness Test) and normal or corrected-to-normal vision. Stimuli were programmed in MATLAB using the Psychophysics Toolbox extensions. Target and flankers were cowhide circular patches of 1.9° diameter. The target was presented 17.6° above fixation reference with two radial flankers. We added random jitter (±1.6° vertical axis, ±1° horizontal) to the target position to prevent colour adaptation. The centre-to-centre distance between target and flankers was 2.1°, corresponding on average to 0.12 times the eccentricity. The luminance pattern was a cowhide pattern generated from random matrix, low-pass filtering it with a cutoff of 2.5 cpd and rounding it to two values (light and dark). Luminance contrast between light and dark regions of the patches was 0.30. Hues were defined as angles in the Derrington-Krauskopf-Lennie (DKL) colour space. Target hues to be judged were selected from 25 equally spaced hues between 176° (green/turquoise) and 349° (pink/red). The two flankers always had the same hue, which covaried along with the target. Specifically, the difference (Δ) between their hues and the target were: Δ = 0°, ±36°, or ±72° (see Fig. 1 for examples). Observers were required to report in two-alternative forced-choice whether the target seemed to be “pinker” or “greener” than a previously learned standard, a violet hue, 263° in DKL space. Conditions with coloured flankers had 0.34 saturation ("full sat"). Targets either had the same saturation, or 0.12 (2.8 times less - "low sat"). In a baseline condition we employed grey flankers with zero saturation (by definition). Stimuli were displayed for 500 ms on a linearized 27” LCD monitor (resolution 1920 x 1080 pixels, refresh rate 60 Hz, mean luminance 99.9 cd/m2, ΔE<3). Observers were positioned 57 cm from the monitor, and maintained fixation on a black dot (0.2 deg diameter) against a mean-grey screen. Observers were first familiarized with the violet colour standard stimulus (263° in DKL space), together with examples of the target hues. They were shown the standard again before each session as a reminder. Observers judged the target as “greener” or “pinker” than the violet standard by keyboard press. They were told that all the colours spanning green to blue should be reported as “greener than the reference”, whereas violet, purple and pink should be reported as “pinker than the reference”. Once the observer had responded, the next stimulus was displayed. No feedback was given. In both the experiments, grey flanker trials were intermingled in the same blocks with the five coloured flanker conditions, while the “no flankers” trials were always in independent blocks, measured either before, after or between subsequent “flanked” blocks.

Institutions

Categories

Funding

Licence