Abstract
Feature-based attention is a critical skill for exploring both real-world and artificial displays. We can select items based on a single feature dimension, such as color, shape, orientation, or motion. Here we tested whether selection performance is higher for simultaneous selection of multiple dimensions (color and shape) that specify the same set, relative to selection of either dimension alone. That is, is conjunctive selection helpful, even when the extra dimension is redundant? We constructed a novel paradigm requiring parallel visual selection on a large set of objects. Targets formed a partial ring embedded among distractor objects. Participants were asked to indicate the quadrant of the screen where the target object ring was missing some elements. After previewing target and distractor objects, the display was rapidly flashed. Display time (M=88ms, SD=33ms) was staircased to halfway between ceiling and chance (25%) performance. Target objects (e.g., blue circles) were identical to each other, and differed from distractors in color only (color trials), shape only (shape trials), or in both color and shape dimensions (conjunction trials). Experiments 1-3 replicated the findings across three target color-shape combinations: blue asterisk, red triangle, and blue circle. Across all three, conjunctive (but redundant) feature-selection showed a massive benefit, an average of 88% performance across experiments, compared to 66% for color alone and 58% for shape alone. Selection of multiple dimensions appears to increase the likelihood of successful selection for that set, suggesting that both feature values can be selected in parallel, and combined. This finding has implications for the way that we encode features in data visualization (e.g., when graphing software such as Microsoft Excel defaults to redundant shape/color conjunctions in graph glyphs).
Meeting abstract presented at VSS 2014