Costs of storing colour and complex shape in visual working memory: Insights from pupil size and slow waves

• Maintenance effort in VWM as a function of set size and relevant feature was estimated.
• Pupil sizes and ERPs were assessed during change detection of coloured shapes.
• Memory for simple colours was better than for complex shapes.
• Pupil size and early N1 correspond to set size related attentional effort.
• Slow waves during maintenance correspond to the difficulty of relevant features.

We investigated the impact of perceptual processing demands on visual working memory of coloured complex random polygons during change detection. Processing load was assessed by pupil size (Exp. 1) and additionally slow wave potentials (Exp. 2). Task difficulty was manipulated by presenting different set sizes (1, 2, 4 items) and by making different features (colour, shape, or both) task-relevant. Memory performance in the colour condition was better than in the shape and both condition which did not differ. Pupil dilation and the posterior N1 increased with set size independent of type of feature. In contrast, slow waves and a posterior P2 component showed set size effects but only if shape was task-relevant. In the colour condition slow waves did not vary with set size. We suggest that pupil size and N1 indicates different states of attentional effort corresponding to the number of presented items. In contrast, slow waves reflect processes related to encoding and maintenance strategies. The observation that their potentials vary with the type of feature (simple colour versus complex shape) indicates that perceptual complexity already influences encoding and storage and not only comparison of targets with memory entries at the moment of testing.