The constraints of object-based attentional selection

• Main Authors: Pei-luen Tsai, 蔡佩倫
• Other Authors: Gary C.-W. Shyi
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/24967178129470579927

博士 === 國立中正大學 === 心理學所 === 93 === The notion of object-based attention is substantiated by the same-object effect (SOE) in studies using divided-attention tasks and spatial-cueing tasks. The SOE refers to the fact that judging two attributes leads to better performance when they belong to the same object rather than when they belong to different objects. However, numerous studies have demonstrated that the SOE may not be a robust phenomenon inasmuch as it could be constrained under a number of circumstances. The main purposes of the present study were to (a) examine whether or not the SOE would be eliminated by increasing the complexity of displayed objects without eye movements; and (b) investigate whether or not the SOE could be recovered when observers were allowed to move their eyes freely. The stimuli used in the six experiments were all 3-D rendered objects, each comprising an elongated cylinder with one or two small attached 3-D parts. The stimulus display consisted of a pair of cylinders that were presented either horizontally or vertically. The divided-attention paradigm was used for all experiments and the participants were asked to determine on each trial whether two identical parts were presented or all displayed parts were different from one another. For the former, the two identical parts could attach either to the same cylinder or to two different cylinders respectively. The complexity of object was manipulated by adding additional parts to the cylinders in Experiments 1 to 3. Results of those experiments showed as the complexity of displayed objects increased, the SOE was constrained or eliminated. In Experiments 4 to 6, we found that SOEs based on the measurements of both response latency and eye movement could be recovered at least partially when participants were allowed to move their eyes under relatively free-viewing conditions (Experiment 5 and 6), but not under constrained viewing condition (Experiment 4). In addition, the results suggest that the scanning strategies of participants were influenced by the time allotted for inspecting the displays, which in turn affected restoration of the SOE. Taken together, the findings of the six experiments lead to the conclusion that (a) the SOE based on the traditional measurement of response latency was constrained by the complexity of displayed objects, and (b) SOEs based on both response latency and eye movement could be partially restored with the aid of eye movements, although still constrained to various degrees by top-down cognitive and strategic factors. Implications for the role of eye movement in object-based attentional selection are also discussed.