2 – Attention 2 Flashcards
What is Feature Integration Theory (Treisman and Gelade, 1980) and how does it relate to attention?
FIT suggests we detect certain simple features of stimuli, such as colour, shape or orientation, BEFORE attention is even required.
What is the difference, in FIT, between a feature and a conjunction?
A feature might be a unique colour, shape or orientation. E.g. look for a red apple in a visual field of green apples.
A conjunction combines two or more features – it’s only unique because of its conjunction. E.g. look for a red circle in a visual field of green circles and green and red squares.
What is the difference between parallel and serial search, according to Treisman and Gelade (1980)?
Parallel search is searching every sector of the visual field at the same time. This is a fast process. Independent of set size or number of distractors, you’ll see target in pretty much the same time. It’s a pre-attentive process.
Serial search is searching for target one object at a time. If you do a serial search, number of distractors and set size begins to matter. This is a slow process that requires attention.
What two distinct processes did Treisman and Gelade suggest might be at work in visual searches?
1) A fast, initial, parallel processing of features (preattentive processing).
2) A slow, serial process of combining features to form objects.
Why, according to Treisman and Gelade, is attention needed for conjunction searches but not for feature searches?
In order to conjoin features –and identify a conjunction target – you need attention. Attention puts things together. Consciousness is also putting things together (coordinating different brain regions).
What are the three pieces of evidence, from visual search studies, that Treisman and Gelade put forward to support the notion that there are two processes involved in visual search?
1) Feature searches (parallel) are faster than conjunction searches (serial)
2) Texture segregation is easier when items separated by feature rather than by conjunction
3) Illusory conjunction – participants report seeing shapes made up of a combination of features from two different stimuli. e.g. if a display of a green X and a red O are flashed on a screen so briefly that the later visual process of a serial search with focal attention cannot occur, the observer may report seeing a red X and a green O.
What role does attention play in perceiving objects (rather than just features), according to FIT?
Attention binds together features with information about location to create objects. This requires effort/focused attention.
How does the attentional spotlight lead to the creation of an object file and recognition in FIT?
- Spotlight directed.
- Visual system activates feature maps.
- Features are conjoined together in an object file.
- The representation is matched to stored descriptions in LTM to allow recognition.
What kind of search slope is expected for single-feature targets vs. conjunctive feature targets as display size increases?
Single-feature targets –flat slope (indicating pre-attentive parallel search)
Conjunctive-feature targets –rising slope (for serial search)
What two confounds in the FIT visual search experimental model were observed by Duncan and Humphreys (1989)?
- Search time is reduced when distractors are more similar to each other. This may explain why feature search is fast
- Search time is increased when distractors similar to target. Conjunction search may be slower simply because target is harder to discriminate.
What three results in Wolfe’s analysis of visual search data (1 million data points) lead him to reject FIT’s distinction between parallel and serial search?
- Data was not bimodal. There should be one bump for feature searches and one for conjunction searches. Not found.
- There should be flat search slopes for feature searches (RT vs display size). But there was always a little slope. These two points suggest there is no distinct pre-attentive process.
- Search slopes for conjunction search shallower than expected (even taking into account that a serial search examines, on average, only half the set). So serial searches do not appear to be strictly serial.
The data suggest that conjunctive searches are harder than, but not qualitatively different from, feature searches.
What model did Wolfe devise to modify FIT?
Guided search theory (1998)
How does Guided Search Theory explain visual search?
The initial processing of basic features produces an activation map, with every item in the visual display having its own level of activation. Suppose someone is searching for red, horizontal targets. Feature processing would activate all red objects and all horizontal objects. Attention is then directed towards items on the basis of their level of activation, starting with those most activated. This explains why search times are longer when some distractors share one or more features with targets.
How does Wolfe differ from Treisman in his explanation of the role of feature search?
Treisman suggests that wefirst do a parallel search – then if no joy, we give up and do a serial search.
Wolfe posits a combination of these processes. Why chuck out the data from the feature search? If looking for a green triangle among green and red squares, Wolfe predicts two parallel searches, and the two feature maps are combined for a standard conjunction.
Why does Wolfe suggest a triple conjunction target is EASIER to find than a simple conjunction? And how is this different from FIT?
FIT would predict a serial search, but Guided Search Theory suggests a more efficient search, as there are more features in the target which are activated (salient). E.g. In a mountain range, a target that is 1) a mountain 2) of a certain height and 3) a certain colour is easier to find than mountain that’s just a certain height.
