Attended images prime both themselves and their left-right reflections, whereas ignored images prime themselves but not their reflections (Stankiewicz, Hummel, & Cooper, 1998). These and other effects are predicted by the hybrid theory of object recognition (Hummel & Stankiewicz, 1996a) that the human visual system represents ignored images holistically (i.e., view-based), and attended images both holistically and analytically (i.e., part-based). In nine experiments using a naming task the predictions of the model were tested with split, plane-rotated and depth-rotated views of common objects. Consistent with the prediction of the hybrid theory, Experiments 1 and 3 demonstrated that split images primed their intact and split counterparts when they were attended but not when they were ignored, whereas intact images primed themselves whether they were attended or not. Experiment 2 demonstrated that a substantial component of the observed priming for attended split images was specifically visual. In Experiment 5, attended images primed themselves fnd their plane-rotated versions (90°) whereas ignored images only primed themselves but not their rotated versions. Experiment 6 tested whether rotated objects with a definite upright orientation prime themselves in the same view. Substantial priming was observed for attended and ignored objects when shown in their upright view. However, rotated objects with a definite upright orientation primed themselves only when attended but not when ignored. This result indicates that ignored images make contact with stored representations. Experiment 7 replicated the findings of Stankiewicz et al. for mirror images but with greylevel rendered 3D images. Experiment 8 tested priming for these objects using orientations in which parts change from study to test view. As before, there was substantial priming in all but the ignored-rotated condition. However, there was a greater reduction in priming for attended rotated objects than for ignored rotated objects. This result indicates that the representations mediating recognition of attended images are specifically sensitive to part changes. In Experiment 9, objects were rotated in depth such that equivalent parts were visible in both views. As in Experiment 7, the priming effects of view and attention were additive. These data provide strong evidence that one function of visual attention is to permit the generation of analytic (i.e., part-based) representations of object shape. At the same time these results show that object recognition is also mediated by additional holistic representations
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:404764 |
| Date | January 2003 |
| Creators | Thoma, Volker |
| Publisher | Goldsmiths College (University of London) |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0021 seconds