Abstract
Previous literature has suggested visual short-term memory (VSTM) has a capacity for approximately four objects. New research shows the complexity of these object also plays an important role. The specific number of objects stored within VSTM were previously suggested to be independent of complexity of the objects. However, many studies have challenged this idea. In the current study, we hypothesized object memory capacity will decrease as object complexity increases, thereby leading to lower accuracy in change detection. Conversely, as object complexity decreases, change detection accuracy should increase since each object requires less VSTM storage space. We measured change detection accuracy between four levels of complexity using the following stimuli: a square (Complexity 1), a square with an added line (Complexity 2), a square with multiple added lines (Complexity 3), and a cube (Complexity 4). Complexity 2 and 3 were essentially partial cubes containing more contours than the square but less than the cube. Participants viewed an array of four of these stimuli at study. Then a blank screen was shown, followed by a test array of the same (No Change) or different (Change) stimuli. Trials were randomized between conditions for all participants. A 4 (Complexity) x 2 (Change Type) repeated-measures ANOVA found significant main effects of Complexity F(3,27) 71.9, p < 0.001, ηp2 = 0.889, and Change Type, F(1,29) = 5.426, p < 0.001, ηp2 = 0.159, in addition to an interaction between these two factors, F(3,27) = 6.707, p < 0.001, ηp2 = 0.427. These data provide important evidence supporting the importance of complexity in VSTM accuracy and perhaps capacity. This study is the first of its kind to explore the impact of stimulus complexity on VSTM using two- and three-dimensional appearing stimuli in a change/no-change paradigm.