Abstract
Mental expansion of scenes’ spatial representation resulting in the memory error referred to as Boundary Extension had been demonstrated using multiple types of tests. People show a consistent tendency to remember having seen a larger expanse of a scene than they actually saw. As there is substantial evident that this robust memory error is a manifestation of underlying processing for spatial layout, it is interesting to examine what effects various concurrent tasks performed while perceiving scenes may have on this processing. In a series of experiments, we compared performance in a control condition, when participants just tried to memorize the scenes, to performance when additional tasks were performed. In these conditions, participants viewed pictures of scenes while performing secondary tasks, such as judging the layout of the scene (hilly/mountainous or flat), searching for people in the scenes, repeating a list of grocery items out loud (verbal task) and describing a list of groceries (verbal+ imagery task). After the initial presentation, participants were presented with versions of the same scenes again, randomly varying the expanse of the scene presented. They were asked to adjust the boundaries of these pictures until the image matched their memory from the original presentation. Replicating prior results with new images, significant boundary extension was obtained in the control condition, when participants simply studied the pictures. The addition of a search task (both for people and mountains) still resulted in Boundary Extension. However, when participants were instructed to also carry out the verbal tasks Boundary Extension was eliminated. Results suggest that, despite the apparent automatic expansion of scene layout, this expansion is affected by high levels of working memory load, including phonological loop load.