Feature integration theory of attention explains attention has to be directed specifically to a certain stimulus in a demonstration whenever conjunctions of more than one separate feature is needed to establish or distinguish the object presenter.
(Treisman, Skyes, & Gelade, 1997) suggest in attention that features come first in perception i.e. when we see an object, the first thing we process is it’s feature.
In feature integration theory of attention the features are registered first, automatically and in parallel across the visual field, e.g. the colour, edges, orientation, location, movement etc. while objects are separately distinguished and identified later on in the process. Therefore a target among a distracter items are easier to find for some stimuli than others if the target has a unique colour or a unique orientation (Treisman & Gelade, 1980).
Treisman suggest that there are two types of search, simple feature search which is parallel and conjunction feature search which is serial.
Parallel theories assume that simple search features and conjunctive feature search are treated by a single mechanism that is regulated by attention and computational demands . however serial theories suggest that parallel processing may be an effective way to search but it’s in efficient searching requires an a additional serial mechanism an attentional “spotlight” ( Treisman A , 1991) that shifts and differ attention to different locations in the visual field.
Treisman suggest only basic features such as colour, size and orientation will support parallel search whereas all other stimuli requires a serial search i.e. serial search is required for targets defined by conjuctions of basic features e.g. a red X amongst green Xs and red Os.
When searching for an object certain targets are detected immediately if they are more unique i.e when they pop out, and the reaction times...