The Influences of Object Type and Visual Field Projection on Reviewing Effects in Recognition of Moving Objects
wang xiaofei,zhang kan,guo sumei
1995-01-01
Abstract:Using a research paradigm of single hemisphere projection and apparent motion,this study examined the influences of stimulus type and hemisphere projection on reviewing effects. Reviewing effects were proposed by Treisman and Kahneman to explain the recognition of a moving object. When an object moves, its shape, size, color and other features change continuously, but it can be recognized as the same one without much difficulty. This implies that the visual system of human beings are able to integrate information of time and space simultaneously. In 1986, Treisman and Kahneman proposed the concept of object files as the theoretical basis to explain the moving object recognition. According to Treisman and Kahneman, an object file is a temporary “episodic” representation of a real world object which is different from the representation of the same object stored in a long term recognition network. The object file collects, integrates, and updatesinformation about the moving object. The perceptual continuity of a moving object is a reviewing process, which connects the current image and previewed image of the object. Reviewing is considered as the process in which a current target item evokes or retrieves an item previewed in an earlier visual field, and it can facilitate recognition when the current and previous states of the object match; hamper it otherwise. The present experiment was designed to explored three issues. First, whether the types of visual stimuli would affect the size of reviewing? If reviewing was guided mainly by the spatial temporal feature, rather than by the shape, or content of the target, the size of reviewing from the three types would be the same when the spatial temporal relationship and the number of objects were kept constant. Second, were there any differences of reviewing between the two hemisphere projection? It was predicted that the reviewing effect would not be influenced by its hemisphere because of its spatial temporal specificity. Third, the difference between reviewing and priming effects was also of our interest. A 3(Visual Stimuli)×3(Successive Objects Matching)×4(Visual Field Projection) within subject design was used in the experiment. Visual stimuli include figures, letters and Chinese characters. Matching types of successive objects includes same object matching (SO), different object matching (DO) and no matching (NM). Visual field projection takes four sets: L L(R), R R(L), L R(R), and R L(L), representing four possible hemisphere locations of the two successive displayed objects. 16 subjects (12 male, 4 female, all college students) participated in the experiment. Each subject was tested in 36 blocks of 20 trials in random order, forming 720 trials, lasting for about 2 hours. The subjects were trained to ensure the error was lower than 5% before formal testing sessions. Response time was recorded as dependent variable. The object specific reviewing effect was the difference of naming RT between the SO and DO conditions(DO-SO). The priming effect was increased by the difference of RT between NM and DO conditions (NM-DO). MANOVA was employed to analyze the data. The results indicated that the naming latencies of simple figure graphs, English letters and Chinese characters were significantly different. The reviewing effects were observed on all three stimulus types while the priming effects were found significantly different among the three stimulus types. This result highlights the difference between reviewing effect and priming effect. The reviewing effect seems object-specific, and was determined by the temporal-spatial pattern of stimuli, while the priming effect was determined by features and semantic relationship between stimuli. The influence of hemisphere location on reviewing effects seems much complicated and need further discussion. In the general discussion, a influence matrix was proposed to analyze the relationship of projecting hemisphere of target stimulus, reviewing effects and priming effects. The experimental data met the theoretical hypot