Directing spatial attention to locations within remembered and imagined mental representations

Spatial attention enables us to enhance the processing of items at target locations, at the expense of items presented at irrelevant locations. Many studies have explored the neural correlates of these spatial biases using event-related potentials. More recently some studies have shown that these ERP correlates are also present when subjects search visual short-term memory. This suggests firstly that this type of mental representation retains a spatial organisation that is based upon that of the original percept, and secondly that these attentional biases are flexible and can act to modulate remembered as well as perceptual representations. We aimed to test whether it was necessary for subjects to have actually seen the memoranda at those spatial locations, or whether simply imagining the spatial layout was sufficient to elicit the spatial attention effects. On some trials subjects performed a ‘visual’ search of an array held in visual short-term memory, and upon other trials subjects imagined the items at those spatial locations. We found ERP markers of spatial attention in both the memory-search and the imagery-search conditions. However, there were differences between the conditions, the effect in the memory-search began earlier and included posterior electrode sites. By contrast the ERP effect in the imagery-search condition was apparent only over fronto-central electrode sites and emerged slightly later. Nonetheless, our data demonstrate that it is not necessary for subjects to have ever seen the items at spatial locations for neural markers of spatial attention to be elicited; searching an imaginary spatial layout also triggers spatially-specific attention effects in the ERP data