Interferences of visual masks with semantic and perceptual priming effects
von Nicolas Becker
Datum der mündl. Prüfung:2018-06-26
Betreuer:Prof. Dr. Uwe Mattler
Gutachter:Prof. Dr. Uwe Mattler
Gutachter:Dr. Igor Kagan
EnglischMany studies demonstrate that visual stimuli can be processed even if they are not consciously perceived. These studies show that unconscious stimuli can be processed – amongst other things – in terms of their shape, color, and even semantic content. To prevent stimulus awareness, researchers present additional stimuli, visual masks, in close temporal and spatial proximity. Although different types of visual masks successfully reduce stimulus visibility it is not entirely clear in what way these masks interfere with information processing. Recent studies show that masks not only prevent stimulus awareness, but also reduce the extent of processing that occurs in the absence of consciousness. At the same time, priming studies that investigate the extent of processing in the absence of consciousness report that priming of low-level perceptual features is unaffected by masking, whereas priming of high-level semantic features is reduced under masking. This has often been interpreted as evidence that non-conscious processes are limited in depth. Here, we investigate an alternative explanation: Since different masks are used in semantic and perceptual studies, it is possible that the reduced effects in semantic studies are due to a confounding mask interference that is not present in perceptual studies. To clarify the role of visual masking for priming effects we examine the effects of conventional forward and backward masking techniques. To that end, we also introduce novel metacontrast masks for words. We find that different types of backward masks reduce stimulus visibility, but do not affect priming effects. Crucially, this was true for both priming of perceptual and semantic features. Forward masks, however, severely reduced perceptual priming effects. This suggests that reduced priming effects in semantic studies may be an artifact of the visual masking technique, since semantic priming studies typically involve forward masks, whereas priming studies on perceptual features often use only backward masks. When forward masks are avoided, semantic priming effects can be successfully dissociated from visibility. Our results bridge theories stating that priming effects are related to an early part of the neuronal response and neurophysiological masking studies stating that early neuronal responses are disrupted by forward masks, but not backward masks.
Keywords: Visual masking; Semantic priming; Color priming; Metacontrast; Paracontrast