Visual marking for search: behavioral and event-related potential analyses

Abstract

Visual marking is an attentional mechanism of prioritizing the selection of new static or moving objects via top-down intentional inhibition of old objects [ Psychol. Rev. 104 (1997) 90–122]. The present study investigated the operation of visual marking using event-related brain potentials (ERPs). Participants searched a display for one of two target letters among randomly selected letter distractors. Two item set sizes were used in a two-stage display. Half of the items appeared for 1000 ms in the first stage followed by the remainder in the second stage. Three blocked conditions were used. In the Preview condition, targets appeared as one of a set of new items in Display 2. Items in Display 1 could thus be excluded from search by visual marking. In the Control condition, old items changed when new items appeared, and the target was equally likely to be at an old or a new position. In this condition marking is unlikely to be engaged since old locations could contain the target. In the Search condition, targets appeared either in Display 1 (80%) or Display 2 (20%), and Display 1 items did not change. The results showed condition-contingent attentional modulations in the N1/N2 range. In addition, a broadly distributed centrally-preponderant sustained negativity was obtained in the 350–750 ms time range after the onset of Display 1 for the Preview–Control comparison. This latter negativity is interpreted to reflect the processes involved in setting up and maintaining of visual marking. It demonstrated an effect of visual marking well prior to Display 2 onset, thereby ruling out an account of the preview benefit solely due to attentional capture by Display 2 item onset.