Limits...
Fast and Conspicuous? Quantifying Salience With the Theory of Visual Attention.

Krüger A, Tünnermann J, Scharlau I - Adv Cogn Psychol (2016)

Bottom Line: Only a few studies systematically related salience effects to a common salience measure, and they are partly outdated in the light of new findings on the time course of salience effects.With this procedure, TVA becomes applicable to a broad range of salience-related stimulus material.A 4th experiment substantiates its applicability to the luminance dimension.

View Article: PubMed Central - PubMed

Affiliation: Faculty of Arts and Humanities, Paderborn University.

ABSTRACT
Particular differences between an object and its surrounding cause salience, guide attention, and improve performance in various tasks. While much research has been dedicated to identifying which feature dimensions contribute to salience, much less regard has been paid to the quantitative strength of the salience caused by feature differences. Only a few studies systematically related salience effects to a common salience measure, and they are partly outdated in the light of new findings on the time course of salience effects. We propose Bundesen's Theory of Visual Attention (TVA) as a theoretical basis for measuring salience and introduce an empirical and modeling approach to link this theory to data retrieved from temporal-order judgments. With this procedure, TVA becomes applicable to a broad range of salience-related stimulus material. Three experiments with orientation pop-out displays demonstrate the feasibility of the method. A 4th experiment substantiates its applicability to the luminance dimension.

No MeSH data available.


Related in: MedlinePlus

Hierarchical Bayesian graphical model of the data of the saliencecondition. The salience condition is indicated by the indexs . The same model applies for the neutralcondition n. The group level, the variables in thehighest layer, estimate TVA parameters for a particular condition. Thislayer was compared to the neutral condition (see Table 1). SOA = Stimulus Onset Asynchrony.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4862317&req=5

Figure 2: Hierarchical Bayesian graphical model of the data of the saliencecondition. The salience condition is indicated by the indexs . The same model applies for the neutralcondition n. The group level, the variables in thehighest layer, estimate TVA parameters for a particular condition. Thislayer was compared to the neutral condition (see Table 1). SOA = Stimulus Onset Asynchrony.

Mentions: To estimate the TVA parameters introduced in this section, a suitable statisticalmodeling is needed. We use Bayesian statistics for modeling and data analysisbecause Bayesian methods are particularly well-suited for inference under anassumed model (Little, 2006). Weimplemented a generative model based on the mathematical description of TVA,visualized in the hierarchical graphical Bayesian model of Figure 2. Table 1shows how the variables (nodes) are formally defined. The graphical modeldescribes the relation between the raw data and the TVA parameters on the grouplevel. As an intermediate step, the TVA parameters are estimated perparticipant. The graphical model depicted in Figure 2 belongs to one group or condition in an experiment. Eachfurther condition is modeled analogously. If there are at least two groups,their group parameters represented at the very top can be compared. On the grouplevel, the mean of attentional weight is represented by node ωspm. Because of technical reasons the variance of the estimatedattentional weight is represented as a separate variable node ωspτ. Similarly, the capacity mean and variance are represented bythe upper two C nodes. Additionally, we can infer thegroup-level processing speed for both targets as represented by the upperυ nodes. However, they do not provide additional information because theydepend on the weight and capacity, as indicated by the direction of the arrows.For further information on the exact nature of the Bayesian parameter estimationprocess, please refer to Appendix A.


Fast and Conspicuous? Quantifying Salience With the Theory of Visual Attention.

Krüger A, Tünnermann J, Scharlau I - Adv Cogn Psychol (2016)

Hierarchical Bayesian graphical model of the data of the saliencecondition. The salience condition is indicated by the indexs . The same model applies for the neutralcondition n. The group level, the variables in thehighest layer, estimate TVA parameters for a particular condition. Thislayer was compared to the neutral condition (see Table 1). SOA = Stimulus Onset Asynchrony.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4862317&req=5

Figure 2: Hierarchical Bayesian graphical model of the data of the saliencecondition. The salience condition is indicated by the indexs . The same model applies for the neutralcondition n. The group level, the variables in thehighest layer, estimate TVA parameters for a particular condition. Thislayer was compared to the neutral condition (see Table 1). SOA = Stimulus Onset Asynchrony.
Mentions: To estimate the TVA parameters introduced in this section, a suitable statisticalmodeling is needed. We use Bayesian statistics for modeling and data analysisbecause Bayesian methods are particularly well-suited for inference under anassumed model (Little, 2006). Weimplemented a generative model based on the mathematical description of TVA,visualized in the hierarchical graphical Bayesian model of Figure 2. Table 1shows how the variables (nodes) are formally defined. The graphical modeldescribes the relation between the raw data and the TVA parameters on the grouplevel. As an intermediate step, the TVA parameters are estimated perparticipant. The graphical model depicted in Figure 2 belongs to one group or condition in an experiment. Eachfurther condition is modeled analogously. If there are at least two groups,their group parameters represented at the very top can be compared. On the grouplevel, the mean of attentional weight is represented by node ωspm. Because of technical reasons the variance of the estimatedattentional weight is represented as a separate variable node ωspτ. Similarly, the capacity mean and variance are represented bythe upper two C nodes. Additionally, we can infer thegroup-level processing speed for both targets as represented by the upperυ nodes. However, they do not provide additional information because theydepend on the weight and capacity, as indicated by the direction of the arrows.For further information on the exact nature of the Bayesian parameter estimationprocess, please refer to Appendix A.

Bottom Line: Only a few studies systematically related salience effects to a common salience measure, and they are partly outdated in the light of new findings on the time course of salience effects.With this procedure, TVA becomes applicable to a broad range of salience-related stimulus material.A 4th experiment substantiates its applicability to the luminance dimension.

View Article: PubMed Central - PubMed

Affiliation: Faculty of Arts and Humanities, Paderborn University.

ABSTRACT
Particular differences between an object and its surrounding cause salience, guide attention, and improve performance in various tasks. While much research has been dedicated to identifying which feature dimensions contribute to salience, much less regard has been paid to the quantitative strength of the salience caused by feature differences. Only a few studies systematically related salience effects to a common salience measure, and they are partly outdated in the light of new findings on the time course of salience effects. We propose Bundesen's Theory of Visual Attention (TVA) as a theoretical basis for measuring salience and introduce an empirical and modeling approach to link this theory to data retrieved from temporal-order judgments. With this procedure, TVA becomes applicable to a broad range of salience-related stimulus material. Three experiments with orientation pop-out displays demonstrate the feasibility of the method. A 4th experiment substantiates its applicability to the luminance dimension.

No MeSH data available.


Related in: MedlinePlus