Abstract:  Although visual working memory (VWM) has been studied for decades, the storage mechanism of VWM remains unclear. A strong object-based theory and a weak object-based theory have been proposed for the VWM storage mechanism. The arguments of these two theories focus on whether features from the same dimension can be integrated into the representation of objects stored in VWM. The strong object-based theory claimed that object was a unit of VWM, the capacity of which was not affected by the number of features. However, the recent studies have proved that weak object-based theory because the features from the same dimension (e.g., two colors) cannot be integrated simultaneously into the representation of objects in VWM, and the VWM capacity was constrained by the number of features. The aim of this paper is to explore whether features from the same dimension can be successfully integrated into a representation of objects stored in VWM. Based on the finding of previous study, we proposed that experimental paradigm and stimulus property might be the main factors that influence the integration of same-dimension information into a representation. Two experiments were conducted to test whether these two factors could affect the integration process respectively. Experiment 1 used a recall task to investigate whether experimental paradigm is a main factor to influence the integration of same-dimension features. The results show that, no matter which experimental paradigms, participants were unable to integrated the same-dimension features into a representation. These suggests that experimental paradigm is not a main factor for explaining the failure of the integration of same-dimension features into a representation stored in VWM. Experiment 2 was conducted to explore whether stimulus property could affect the integration of same-dimension features. Two types of stimulus were selected as experimental materials (e.g., meaningless and meaningful objects). We expected that, in meaningful object condition, participants would easily memorize two conjoint colors as one representation because of the integrated clues (the meaningful shapes); in contrast, participants were more likely to memorize the two features separately due to the absence of integrated clues. The results of Experiment 2 show that, the same-dimension features could not be integrated into one representation in both conditions (meaningful and meaningless objects). Thus, Experiment 2 suggests that the same-dimension features could not be integrated into a representation of objects stored in VWM, regardless of the integrated clues. The present study provides supporting evidences for the weak object-based theory by claiming that manipulation of variables such as experimental paradigm and stimulus property exerts no effect on the memorization of same- dimensional features.

Key words:  visual working memory, features integration, weak object-based theory