5.3. Correlation Results
In the process of data analysis, the selection of correlation coefficient needs to be determined according to the data type. Common indicators include Pearson linear correlation coefficient, Spearman rank correlation coefficient, and Kendall rank correlation coefficient. Based on the continuous variable characteristics of this study, SPSS software was used for bivariate correlation testing, and Pearson coefficient was selected as the main analysis method.
(1) Natural ecological analysis
The correlation analysis between the visual proportion of bamboo and wood elements and natural ecological evaluation is shown in
Table 14.
According to the analysis of
Table 14, the correlation coefficient between the visual proportion of bamboo and wood elements and natural ecological evaluation at the 0.01 level is 0.746, as determined by the bilateral Pearson correlation test, P=0.000, There is a significant positive correlation between the two, and the visual proportion of bamboo and wood elements can reflect and measure the natural ecological level of bamboo and wood landscape bridges. The larger the visual proportion of bamboo and wood elements, the higher the natural ecological level.
(2) Coordination and unification related analysis
The correlation analysis between visual entropy and coordination and unified evaluation values is shown in
Table 15.
According to the analysis of
Table 15, the correlation coefficient between visual entropy and coordination and unified evaluation values at the 0.01 level is -0.854, as determined by the bilateral Pearson correlation test, P=0.000, There is a significant negative correlation between the two, and the size of visual entropy can reflect and measure the coordination and unity of bamboo and wood landscape bridges. The larger the visual entropy value, the worse the coordination and unity between the bridge and the surrounding environment.
(3) Scale and proportion correlation analysis
The correlation analysis between proportionality, scale, and proportionality evaluation values is shown in
Table 16.
According to the analysis of
Table 16, the correlation coefficient between proportionality and scale and proportionality evaluation values is 0.869 at the 0.01 level, as determined by the bilateral Pearson correlation test, P=0.000, There is a significant positive correlation between the two, and the degree of proportionality can reflect and measure the scale and proportion of bamboo and wood landscape bridges. The larger the degree of proportionality, the more coordinated the scale of bamboo and wood landscape bridges.
(4) Color related analysis
The correlation analysis between color beauty value and color evaluation value is shown in
Table 17.
According to the analysis of
Table 17, the correlation coefficient between color beauty value and color evaluation value at the 0.01 level is 0.797 through bilateral Pearson correlation test, P=0.000.There is a significant positive correlation between the two, and the beauty value can reflect and measure whether the colors of bamboo and wood landscape bridges are harmonious. The higher the beauty value, the more harmonious the colors of bamboo and wood landscape bridges are.
(5) Visual Transparency Correlation Analysis
The correlation analysis between sky width and visual permeability evaluation values is shown in
Table 18.
According to the analysis of
Table 18, the correlation coefficient between sky width and visual permeability evaluation value at the 0.01 level was 0.876 as determined by the bilateral Pearson correlation test, P=0.000. There is a significant positive correlation between the two, and the size of the sky opening can reflect and measure the visual transparency of bamboo and wood landscape bridges. The larger the sky opening value, the wider the field of view of bamboo and wood landscape bridges.
5.4. Correlation Analysis
Based on the theoretical framework of psychophysics, the mathematical relationship model between external visual stimuli and subjective perception is mainly reflected in three functional forms: linear model, logarithmic model, and power function model. SPSS statistical analysis software is used to perform function model fitting tests on experimental data. Through linear regression analysis in the curve estimation method, the fitting of the three types of function models is as follows:
(1) The relationship between the visual proportion of bamboo and wood elements and natural ecological evaluation
On the basis of correlation analysis, further research was conducted on the functional relationship between the visual proportion of bamboo and wood elements and natural ecological evaluation.
Table 19 shows the fitting results of linear, logarithmic, and power function models.
According to the data analysis of
Figure 9 and
Table 19, the linear relationship between the visual proportion of bamboo and wood elements and the natural ecological evaluation value is stronger than the power function relationship, and the strength of the logarithmic function relationship is closer, which can be expressed as:
In the formula, y is subjective preference evaluation of the natural ecology of bamboo and wood landscape bridges by the test subjects, x is visual proportion of bamboo and wood elements.
This linear model reflects that for every 1% increase in the proportion of bamboo and wood visual elements, the natural ecological evaluation increases by approximately 0.008 points. This indicates that the visibility of bamboo and wood materials has a quantifiable promoting effect on enhancing the harmony between landscape bridges and the natural environment. However, the slope of the linear relationship is gentle, indicating that the visual proportion of bamboo and wood elements needs to reach a high proportion in order to significantly improve the natural ecological evaluation. It is recommended to prioritize increasing the visual proportion by exposing large areas of bamboo and wood materials and strengthening bamboo texture while balancing structural safety and functionality.
(2) The relationship between visual entropy, coordination, and unified evaluation
On the basis of correlation analysis, further research was conducted on the functional relationship between visual entropy and coordinated unified evaluation.
Table 20 shows the fitting results of linear, logarithmic, and power function models.
According to the analysis of the data in
Figure 10 and
Table 20, the linear relationship between visual entropy and the evaluation values of coordination and unity is stronger than the power function relationship, and is closer to the strength of the logarithmic function relationship. It can be expressed as:
In the formula, y is subjective preference evaluation of the subjects for the coordination and unity of bamboo and wood landscape bridges, x is visual entropy.
This linear model reflects a stable linear decreasing relationship between visual entropy x and coordination and unity evaluation y, indicating that the complexity and disorder of landscape elements significantly weaken participants’ perception of the unity between the bridge and the surrounding environment. Therefore, in the design of bamboo and wood landscape bridges, it is recommended to reduce the splicing of heterogeneous materials and prioritize the use of single or similar material combinations dominated by bamboo and wood; Control the contrast between hue and brightness, and weaken visual jumps through adjacent colors or gradient color schemes.
(3) The relationship between color beauty value and color evaluation
On the basis of correlation analysis, further research is conducted on the functional relationship between color beauty values and color evaluation.
Table 21 shows the fitting results of linear, logarithmic, and power function models.
According to the analysis of the data in
Figure 11 and
Table 21, there is a power function relationship between the color beauty value and the color evaluation value, which can be expressed as:
In the formula, y is subjective preference evaluation of the color of bamboo and wood landscape bridges by the test subjects, x is color beauty value.
The model shows a super linear growth relationship between color beauty values and color evaluation values (power exponent b=1.156>1). As the beauty value increases, the speed of color evaluation improvement gradually accelerates, and the beauty value needs to reach a certain threshold to produce a significant improvement in harmony. It is recommended to systematically improve the aesthetic value in the design of bamboo and wood landscape bridges through color saturation control or contrast enhancement with environmental vegetation.
(4) The relationship between sky width and visual transparency evaluation
On the basis of correlation analysis, further research was conducted on the functional relationship between sky width and visual permeability evaluation.
Table 22 shows the fitting results of the linear model.
According to the analysis of the data in
Figure 12 and
Table 22, it can be concluded that there is a linear function relationship between sky width and visual permeability evaluation value, which can be expressed as:
In the formula, y is subjective preference evaluation of visual transparency of bamboo and wood landscape bridges by test subjects, x is the width of the sky.
This model reflects a significant positive correlation between sky width and visual transparency evaluation. For every 1 unit increase in sky width, the average visual transparency evaluation improves by nearly 1 point, indicating that the expansion of the sky field of view has a strong amplification effect on visual transparency perception. In the design of bamboo and wood landscape bridges, the influence of sky width should be fully considered, and the sky width should be appropriately increased to enhance the visual transparency and aesthetics of the bridge.
(5) The relationship between proportionality, scale, and proportionality evaluation
On the basis of correlation analysis, further research was conducted on the functional relationship between proportionality, scale, and proportionality evaluation.
Table 23 shows the fitting results of linear, logarithmic, and power function models.
According to the analysis of the data in
Figure 13 and
Table 23, it can be seen that there is a power function relationship stronger than linear and logarithmic relationships between the proportionality and scale and the proportionality evaluation value, which can be expressed as:
This model reflects a super-linear accelerated growth relationship (power exponent b=2.115>1) between proportionality, scale, and proportional evaluation values. With the increase of proportion, the growth rate of bridge scale and proportion evaluation significantly accelerates. The proportional PM is in the range of 0.5-0.7, indicating a low level of evaluation, while in the high proportional range (0.8-1), the evaluation has significantly improved. When the proportional degree PM>0.8, the growth rate of scale and proportional evaluation will significantly accelerate.
Above all, except for the negative linear relationship between visual entropy and the evaluation of coordination and unity of bridges, the increase of the other four objective values (independent variables) has a positive promoting effect on the corresponding subjective evaluation (dependent variable), but the magnitude of the increase varies. And the rate of subjective perception improvement lags behind the changes in objective indicators, indicating that people’s aesthetic cognition has a delayed response characteristic. Regression analysis shows that there is no significant difference in the goodness of fit between linear relationships, power function relationships, and logarithmic function relationships, which may be due to the limited sample size making it difficult to fully reveal the morphological characteristics of the function curve between objective values and their corresponding subjective evaluations. However, the logarithmic relationship of Fechner’s law and the power function relationship of Stevens’ law are almost the same under certain experimental conditions and sample sizes, and can be replaced by each other.