Skip to main content

Table 2 Summary of regression models at the scene category level

From: Toward viewing behavior for aerial scene categorization

Estimate

t value

Std. Error

p value

Coefficient information (Organized as shown to the left; Intercept omitted)

Image statistic

Eye movement

Fixation duration

Number of fixations

FDM entropy

Saccade amplitude

GTE (Scanpath randomness)

Low

Low S.F

  

− 19.49

− 2.23

8.7

= 0.076

      

High S.F

        

− 1.12

− 2.17

0.51

= 0.073

Block_1 PCA

          

Mid

Homogeneity

      

− 1.74

− 2.92

0.60

*

  

Block_5 PCA

  

− 1.8e−03

− 4.5

3.9e−04

**

− 1.7e−03

− 4.22

3.9e−04

**

− 0.002

− 4.86

0.0004

**

1.8e−05

2.41

7.4e−06

= 0.053

High

Within Sim

      

− 2.24

− 2.73

0.82

*

  

Across Sim

  

7.03

2.81

2.51

*

      

FC_2 PCA

  

2.5e−02

4.93

5.0e−03

**

0.014

3.26

0.004

*

0.017

3.872

0.004

**

  

Object

Critical object saliency

14.31

3.72

3.85

**

− 1.32

− 6.92

0.19

***

− 0.47

− 2.87

0.16

*

  

− 0.018

− 3.09

6.0e−03

*

Adjusted R2

0.66 (*)

0.96 (***)

0.82 (**)

0.81 (**)

0.57 (p = 0.062)

  1. ***p < 0.001, **p < 0.01, *p < 0.05