Both Target Detection and Response Contribute to the Attentional Boost Effect

Detecting and responding to targets in an rapid serial visual presentation (RSVP) stream facilitates the processing of concurrently presented images. Theoretical accounts of this attentional boost effect (ABE) have emphasized the role of target detection, yet it is unclear whether the ABE originates from target detection or response. To examine this, we asked participants to search for the letter T among a rapid stream of other letters while encoding objects to memory. The nontarget letters, drawn randomly from the remaining letters of the Roman alphabet, comprised half of the trials. Across three experiments, participants responded only to the target letter T, to all letters but the target letter T, or to both T and not-T with two different keys. The large number of different nontargets ensured that participants preferred to search for the target letter T regardless of the response requirement. We found an ABE when participants responded to the target letter T, but not when they responded to the other nontarget letters, suggesting that response requirement modulated the ABE. Furthermore, making different responses to targets and nontargets led to a memory advantage for target-paired objects.