Purchase this article with an account.
Lisa R. Fournier, Jennifer Nelson, Matthew Wiediger; Conjunction benefits can occur for dimensions within an object but not between objects. Journal of Vision 2005;5(8):1031. doi: 10.1167/5.8.1031.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Conjunction benefits refer to the case in which discriminating the presence of multiple features within an object (feature conjunctions) can be faster than discriminating the presence of the less discriminable feature alone (Fournier et al., 1998, 2000; 2004). An asynchronous priming model assumes that conjunction benefits result from early partial decision activation by more discriminable features that are combined with activation by less discriminable features to meet a single decision criterion. Conjunction benefits occur if task-relevant dimensions differ in discriminability and features on the task-relevant dimensions are consistently mapped to a response. However, it is unclear whether they occur when task-relevant dimensions occur across two different objects (or parts of an object). This study investigated whether conjunction benefits can occur for feature discriminations across two conjoined, 2-D objects. Observers judged whether one or two features were present or absent in a stimulus consisting of a geometric object bisected by a line. There were three dimension judgment conditions: object shape and/or line orientation (between object), object shape and/or line color (between object), line color and/or line orientation (within object). Discriminability difficulty between the task-relevant dimensions was varied in each condition. Results showed that conjunction benefits occurred only for dimensions within the same object. This suggests that decision activation from features within an object, as opposed to across different objects, can asynchronously prime a common response decision in parallel. When relevant dimensions occur across objects, attention may be serially allocated to each dimension, preventing response decisions from activating a common decision in parallel. The implications of these results for visual attention models, as well as the asynchronous priming model, will be discussed.
This PDF is available to Subscribers Only