We will write a custom Essay on Comparison of Guided Search Model and Feature Integration Model of Visual Search specifically for you
301 certified writers online
Scientific evidence supports the feature integration model. Experiments among patients with Balint’s syndrome, simulated experiments, and existence of the binding problem all testify to the relevance of attention in visual searches.
However, this model has weaknesses, which the guided search model solves. The guided search model explains why certain conjunction searches may be fast or slow and why feature searches may also be slow.
It accounts for distracters and solves the problem of the relevance of feature information from the parallel phase.
Using the activation map, the model successfully outlines why certain searches may be inefficient or inefficient. It is this explanatory power that makes the guided search model more effective than the feature integration model in visual searches.
A visual search is the examination of something within cluttered surroundings. The thing that one wants to find is called the target while the other items in the group are distractors. The qualities that one uses to find the target, such as colour or form, are known as features.
If the target possesses more than one feature, then it is a conjunction. Visual searches are important because many professions involve finding items in cluttered environments.
Understanding the visual search process can assist in controlling the distractors and making location of items more efficient.
Comparison of the two models
The feature integration model holds that visual searches occur in two distinct phases. The first phase is relatively fast and is the parallel phase.
Anne Treisman, who is the leading advocate of the model, explains that the parallel phase requires minimal attention and is not complicated (Sobel & Cave, 2002). Here, the subject will focus on basic features like size, orientation and colour. Other stimuli that are more complex will not be detected.
Thereafter, a visual search process enters the serial phase, which is rather sophisticated. Treisman (1988) explains that when features occur as conjunctions, then the visual search enters into this second stage.
For instance, if one wants to find red circles in a combination of green circles and red squares, then a serial process will come into play. The theorists also added that subjects will combine features by focusing their attention on the location of the target.
Attention thus acts as the bond that brings together features in the search (Palmer, 1995). However, the theorists also add that stored knowledge has an effect on how one combines features.
Therefore, if one does not focus their attention on object location or if one lacks stored knowledge on the features, then illusory results will arise. In this regard, features will be combined randomly, and will result in wrong conjunctions.
A series of experiments exist to validate the feature integration model.
Get your first paper with 15% OFF
However, the same may also be asserted for the guided search model. Scientists carried out experiments to prove that attention focusing is necessary when looking for conjunction items (Treisman and Gelade, 1980), (Treisman and Southern, 1986).
This supported the feature integration model. Here, targets’ detection speed changed when the set size altered if conjunction items were present. However, detection rates did not alter when the set size was changed for single-feature objects.
This indicates that an attention-demanding process came into play when looking for conjunctions (Kaniszewski, 1998). Another phenomenon that supports the feature integration model is binding problems.
Sometimes subjects may combine features from the target object in wrong ways (Schall, 2004). This presents a binding problem that can only be solved when one pays attention to the object location.
Therefore, the binding problem supports Treisman’s assertions that attention focusing is imperative in performing a visual search.
In the guided search model, experiments that Wolfe et. al. (1989) conducted among naive subjects looking for conjunctions found flat slopes. This proves that the feature search phase played a role in the serial search.
The flatness of their slope indicates that conjunction searches are indeed guided by information from the parallel phase. What makes experiments from the guided search quite compelling is that they rely on the same parameters that Treisman and her colleagues used.
For instance, in one case, the subjects searched for combinations of form and colour, and their search results were almost flat slopes (Wolfe, 2001). Furthermore, when the subjects looked for triple conjunctions of form, size and colour, their slopes were flatter than the double conjunctions.
This proves that a self terminating parallel search is unsatisfactory as suggested by Treisman. It also indicates that some information from the first phase is relevant in the second one.
Perhaps one compelling experiment that validates the feature integration model is evidence from patients with visual processing problems. Persons with Balint ‘s syndrome have a defect in their parietal cortex that causes them to have difficulties in focusing attention (Palmer et. al., 2000).
When these persons are asked to search for objects with combined features, they are not likely to find them. Treisman worked with a patient known as RM who had difficulties in processing visual information.
These experiments prove that attention focusing is critical in explaining visual searches as prescribed in the feature integration model.
Most experiments conducted by scientists advocating for the guidance search theory involve normal subjects who do not have damage to the parietal cortex. Therefore, the guided search model is less convincing in this regard.
In the feature integration model, the parallel phase has minimal impact on the serial phase. Advocates of this theory believe that subjects can find objects using only one feature (Treisman, 1988).
The theory does not explain the use of the information gathered from the parallel phase in subsequent phases. At this point, the guided search model becomes relevant because it accounts for the information that subjects gather in the first phase of a search.
If one cannot identify an object using features, then the brain still finds use for this information the latter theorists (Tsotsos, 1990). Take the example of a red square among green squares and red circles. The target object is a conjunction between colour and shape.
Therefore, the parallel phase cannot account for location of this object. However, this phase is still not useless because it detected red and green features.
It thus makes sense for the parallel process to transfer this information to the serial process so as to prevent the serial process from wasting time detecting the red and green symbols again (Lavie & De Fockert, 2006).
In this experiment, the parallel phase would tell the serial phase that squares are green and circles are red.
Therefore, the serial phase ought to restrict its criterion to identifying the exception. The guided phase model thus finds a way of accounting for previously gathered information that the feature integration model does not.
Another point of departure between the feature integration model and guided search model is the significance of serial processes in conjunction searches.
The feature integration theory holds that searches for conjunctions must go through both phases; however, the guided search model claims that sometimes a serial search may be unnecessary in conjunction experiments.
This is seen when the search for conjunctions occurs at a much faster rate than the feature integration model predicted (Neider & Zelinsky, 2006). This occurs when certain features fail to guide a person’s attention.
Take a case of a series containing a cross in the midst of other symbols as shown in the diagram below (Howe, 2013). The cross is the only parameter that consists of aligned intersections. However, this aspect is not useful in guiding a subject’s attention.
Feature maps as envisaged in the feature integration model are not always relevant in attention guidance. This was shown in the guided search model. Therefore, one may question the plausibility of the feature integration theory in terms of this parameter.
The guided search model fills some of the gaps prevalent in the feature integration theory. In the latter theory, Treisman claimed that a visual search is mostly determined by the characteristic of the targets; in that they could be either conjunctive or single (Nelson, 2001).
Therefore, she did not consider the role that distracters play in the visual search model. The guided search model accounts for this by including the role that distractors play in search process. In this model, a visual search can either be efficient or inefficient.
If it is efficient, then the number of distractors in the set has minimal effect on the search process because it is merely a parallel process with a flat slope (Nobre et. al., 2002). However, if it is inefficient, then the distractors play a significant role in determining the result.
Wolfe (2003) found that in efficient searches, a target yields very high activation peaks. When subjects were looking for red horizontal objects, they would get activated for horizontal and red items. Greater attention will go to those ones that have the greatest activation.
If distractors have common characteristics with the target, then search processes will take longer. Conversely, if the search is inefficient, then the reaction time will increase with an increase in distractors.
This occurs due to the presence of several high activation peaks, which surpass the activation peak of the target. Therefore, the guided search theory does something that the feature integration theory could not do; it accounts for the role of distractors in search outcomes.
It is evident that both theories have their relevance in explaining visual searches. The feature integration model is one of the most powerful schools of thought on visual processes. It is responsible for showing the relevance of neurological functions in visual searches.
The theory is also quite useful in showing the two-step nature of visual searches. Experiments among patients with Balint’s syndrome, simulated experiments and existence of the binding problem all testify to the relevance of attention in visual searches.
On the other hand, this theory has a lot of weaknesses.
First, the presumption that parallel and serial searches are separate and distinct has been disproved.
Furthermore, the theory does not account for faster detection in some conjunction searches than others. It does not consider the role of distractors and thus does not account for slower searches when distractors are increased.
Lastly, the model does not explain the role of the parallel phase in subsequent searches.
For these reasons, a more refined approach in the form of the guided search model is applicable. The latter is an improvement on the feature integration theory because it accounts for most of the gaps in this theory.
The guided search model explains why certain conjunction searches may be fast or slow and why feature searches may also be slow. It accounts for distractors and solves the problem of the relevance of feature information from the parallel phase.
Using the activation map, the model successfully outlines why certain searches may be inefficient or inefficient. Its key weakness is its inability to use real-life patients to demonstrate its findings like Treisman did.
Since the guided search model is a late development, it seems to have foreseen the problems encountered in the feature integration theory and solved them. Therefore, is more plausible for understanding visual processing.
Howe, P. (2013). Weeks 3 lecture: Selective attention and visual awareness. Melbourne: University of Melbourne Publication.
Kaniszewski, C. (1998). The influence of display characteristics on visual exploratory search behaviour. Journal of Consumer Research, 25 (3), 45-66.
Lavie, N., & De Fockert, J. (2006). Frontal control of attentional capture in visual search. Visual Cognition, 14(4), 863-876.
Neider, M. B., & Zelinsky, G. J. (2006). Searching for camouflaged targets: Effects of target-background similarity on visual search. Vision Research, 46(14), 2217-2235.
Nelson, C. (2001). The development and neural bases of face recognition. Infant and Child Development, 10 (1-2), 3-18.
Nobre, A., Sebestyen, G., Gitelman, D., Frith, C. and Mesulam, M. (2002). Filtering of distractors during visual search studied by positron emission tomography. Neuroimage, 16(4), 968–976.
Palmer, J. (1995). Attention in visual search: Distinguishing four causes of a set size effect. Current Directions in Psychological Science, 4(4), 118-123.
Palmer, J., Verghese, P., & Pavel, M. (2000). The psychophysics of visual search. Vision Research, 40, 1227-1268.
Schall, J. (2004). On the role of frontal eye field in guiding attention and saccades. Vision Research, 44 (12), 1453–1467.
Sobel, K. V., & Cave, K. R. (2002). Roles of salience and strategy in conjunction search. Journal of Experimental Psychology: Human Perception & Performance, 28(5), 1055-1070.
Treisman, A. (1988). Features and objects: the fourteenth Bartlett Memorial Lecture. Quarterly Journal of Experimental Psychology, 40, 201–236.
Treisman , A. and Gelade, G. (1980). A feature-integration theory of attention. Cognitive Psychology, 12, (1), 97–136.
Treisman , A. and Southern, J. (1986). Illusory words: The roles of attention and of top–down constraints in conjoining letters to form words. Journal of Experimental Psychology: Human Perception and Performance, 12(1), 3–17.
Tsotsos, J. K. (1990). Analyzing vision at the complexity level. Brain and Behavioural Sciences, 13(3), 423-469.
Wolfe, J. (2001). Guided search 4.0: A guided search model that does not require memory for rejected distractors. Journal of Vision, 1(3), 349.
Wolfe, J. (2003). Moving towards solutions to some enduring controversies in visual search. Trends Cognitive Science, 7(2), 70-76.
Wolfe, J., Cave, K. & Franzel, S. (1989). Guided search: An alternative to the feature integration model for visual search. Journal of Experimental Psychology, 15(3), 419-433.