Abstract
Contrary to the experience of a complete representation of our surroundings, representations of our visual surroundings are limited and constrained by the capacity limits of visual working memory (VWM). However, as long as visual information is readily available in the external visual world, there is no need for a complete internal representation of the outside world, as the world can act as an external memory source. Previous studies have therefore proposed a trade-off between storing information in VWM and making saccades. Here we directly tested this trade-off by using a copying task in which participants were instructed to remember, and copy a model consisting of an arrangement of many elements. If there is an adaptive trade-off between using the external visual world and VWM, the trade-off should be influenced by increasing the cost associated with using external information. We therefore increased the cost of a saccade by increasing the amount of time between saccade offset and external availability of visual information. Results show that participants made numerous eye movements between the work-space, model and blocks, often making one saccade per action performed. Increasing saccade costs resulted in more saccades towards the model and an increased dwell time on the model, suggesting a shift from making eye movements towards taxing internal VWM. Our study provides strong behavioral evidence for a trade-off between using the external world as a memory buffer, versus building complex internal representations. Prior research has shown that VWM and eye movements are directly linked and our results provide evidence that that the link between executing eye-movements and building an internal representation of our world is based on an adaptive mechanism governed by cost-efficiency. Lastly, these results indicate that the default mode of executing many eye-movements is more cost-efficient than storing information in memory.