The processing architectures of whole-object features: A logical-rules approach

Abstract: In this article, we examine whether dimensions comprising the entirety of an object (e.g., size and saturation) are processed independently or pooled into a single whole-object representation. These whole-object features, while notionally separable, sometimes show empirical effects consistent with integrality. A recently proposed theoretical distinction between integral and separable dimensions that emphasizes the time course of information processing, can be used to differentiate whether whole-object features are processed independently, either in serial or in parallel, or pooled into a single coactive process (see, e.g., Little, Nosofsky, Donkin, & Denton, 2013). The current research examines this theoretical distinction in the processing of 3 sets of whole-object-featured stimuli that vary on any pair of the dimensions of saturation, size, and orientation. We found that a mixture of serial and parallel architectures underlies the processing of whole-object features. These results indicate that whole-object features are processed independently.