Prolonged focal attention without binding: Tracking a ball for half a minute without remembering its color.

Conventional theories of cognition focus on attention as the primary determinant of working memory contents. However, here we show that about one third of observers could not report the color of a ball that they had just been specifically attending for 5-59 s. This counterintuitive result was obtained when observers repeatedly counted the passes of one of two different colored balls among actors in a video and were then unexpectedly asked to report the color of the ball that they had just tracked. Control trials demonstrated that observers' color report performance increased dramatically once they had an expectation to do so. Critically, most of the incorrect color responses were the distractor ball color, which suggested memory storage without binding. Therefore, these results, together with other recent findings argued against two opposing theories: object-based encoding and feature-based encoding. Instead, we propose a new hypothesis by suggesting that the failure to report color is because participants might only activate the color representation in long-term memory without binding it to object representation in working memory.