Next Article in Journal
Predicting the Impact of Academic Key Factors and Spatial Behaviors on Students’ Performance
Previous Article in Journal
Deep Transfer Learning Model for Semantic Address Matching
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Applied Sciences
Volume 12
Issue 19
10.3390/app121910111
applsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Research on the Influence of User and Graphic–Text Combined Icon Construal Level Fitting on Visual Cognition

by
Yanfei Zhu
1,
Ying Li
2,
Yun Lin
3,
Mo Chen
4,
Qi Guo
5 and
Zhisheng Zhang
1,*
1
School of Mechanical Engineering, Southeast University, Nanjing 211189, China
2
Department of Art and Design, Beijing University of Chemical Technology, Beijing 100029, China
3
School of Design Art and Media, Nanjing University of Science and Technology, Nanjing 210014, China
4
School of Art and Design, Nanjing University of Technology, Nanjing 210031, China
5
School of Art Design and Media, East China University of Science and Technology, Shanghai 200237, China
*
Author to whom correspondence should be addressed.
Appl. Sci. 2022, 12(19), 10111; https://doi.org/10.3390/app121910111
Submission received: 30 July 2022 / Revised: 1 October 2022 / Accepted: 6 October 2022 / Published: 8 October 2022
(This article belongs to the Section Applied Industrial Technologies)
Browse Figures
Versions Notes

Abstract

:
Based on the individual traits of the user’s construal level differences, this study examines the visual cognition differences of graphic–text combined icon concreteness in the interface between users with high and low construal levels. We classified four types of graphic–text combined icons, which are as follows: concrete graphic + concrete text (Ci + Ct), concrete graphic + abstract text (Ci + At), abstract graphic + concrete text (Ai + Ct), and abstract graphic + abstract text (Ai + At). According to the behavioral experiment results, there is no visual cognitive difference between the two types of traits users in Ci + Ct. In terms of Ci + At and Ai + Ct, the visual cognitive performance of high construal level users is slightly better than that of low construal level users. In Ai + At, there are obvious visual cognitive differences between the two types of users. Visual cognitive fluency plays a mediating role in response time and matching rates of the two types of users. Based on the construal level perspective, this study considers the visual cognitive differences based on the user’s stable traits. It provides a certain reference for the graphic–text combined icons’ design in the adaptive human–computer interaction interface.

1. Introduction

As the smallest symbol unit in the digital interface, the icon is the channel between the user and the computer target program or command search, and plays an important role in the interaction between the human and the computer [1]. A large number of user studies have shown that whether its design is reasonable or not has an important impact on users’ visual cognition [2,3,4]. The theoretical basis of interface usability design is affordance. Norman regards affordance as an information representation. He encourages designers to use affordance that can highlight interactive behaviors in specific designs, thereby improving the success rate of users’ perception of functions [5]. Nielsen proposed ten usability heuristic principles, such as matching and consistency in interaction design, to enhance the explanatory power of interface information to inspire users [6,7]. With the development of adaptive technology, the human–computer interface has gradually tended to be customized based on the individual characteristics of users, to more accurately serve the operational needs of user groups with different characteristics [8,9,10]. Therefore, establishing an organic relationship between the characteristics of users and icon design is a topic worthy of discussion in the future development of adaptive interface design.
In previous studies, the external features of icons were mainly discussed in terms of the size, color, density, text, spatial relationship, and different combinations, and how these affect the user’s cognition [11,12,13,14]. However, what is more important is how to effectively convey the semantics of icons, which is related to the internal attributes [15]. Research on the internal attributes of icons mainly focuses on visual complexity, concreteness, and semantic distance [16,17]. The concreteness of an icon refers to the authenticity of things in the real world, and related research show that users have a faster action response to more concrete icons [18,19,20]. However, concreteness does not exist independently. It interacts with other design attributes, tasks, and user types may produce differences in effective recognition, reducing or improving comprehensibility performance [21]. Rationalized visual representations that match the semantics of information will improve the effectiveness of user cognition, and whether the cognition can be matched by quick responses and fluency is an effective indicator of design usability [22].
Cross-modal relationship matching construction has always been an essential task in information systems, which requires the presentation of information from two different modalities into a shared space [23,24]. Compared with pure text or graphic icons, the combination of graphics and text has a higher degree of design appeal in improving user identification performance [25,26]. The high-performance matching composition representation of text labels combined with graphs has been extensively studied in the technical field [27]. The consistency of the bi-semantic clues between the image and the text label in the combination icon may affect the user’s performance in terms of information integration [28,29]. However, the ultimate server of information representation is the user, and the essential issue is whether the information is convenient for the user to understand. Relevant studies have shown that there are differences in internal user traits, and the way individuals understand tasks or objects has an important impact on their visual cognitive performance [30,31,32]. However, research on the relationship between the internal attributes of the icon and user traits in the combination of graphics and text is relatively sparse [33].
As a stable individual trait, the construal level mainly refers to the user’s preference for abstract or concrete representations in information coding [34]. This trait will affect the user’s cognitive state and influence decision-making [35,36]. Individuals with low construal levels usually use a specific information processing style, look closely at things, and initiate concrete mental representations. In contrast, people with high levels of understanding tend to present relatively abstract mental representations [37,38]. Fluency is an individual’s subjective experience of how easy it is to process information, and visual fluency is mainly affected by the objective characteristics of information and individual traits [39]. Visual fluency affects users’ preferences and decision-making. Studies have shown that processing speed and accuracy can be used as objective evaluation indicators [40,41]. Consistent integration between the construal level of user thought patterns and information clues could effectively initiate high operational performance, and fit can change user attitudes, trust, and attractiveness [42,43]. High inconsistency in cognitive conflict will affect users’ abstract construal level, especially for users with low construal levels, which will increase cognitive difficulty [44]. The information design of cognitive conflict will reduce users’ visual cognitive ability, which is mainly mediated by the participation of cognitive fluency in the visual process [45,46].
To the best of our knowledge, construal level user traits are an important factor that affects the visual cognition of information concreteness. In past research, relevant scholars have carried out extensive discussions and tests on the matching degree of user construal level traits and information concreteness in various fields [38,47], but as far as we know, in the interface environment, how the user’s construal level traits interact with the concreteness of the combined icon, and thus how they affect the user’s decision-making process, still needs to be verified. So, we pursued this question.
This study aims to explore the differences in the visual cognitive behavior of users with different construal levels in the graphic–text combined icon. We examined several points, which are as follows: (1) whether the concreteness of graphic–text combined icons and user construal level traits of visual cognition influence each other; (2) whether the coherence of clues of user construal level traits and graphic–text combined icon concreteness has a positive impact on visual cognition; (3) whether the behavioral performance of users’ different construal levels of decision-making in relation to visual cognition is mediated by cognitive fluency. We use behavioral experiments to capture the subjective and objective indicators of the visual cognition of users with construal level traits when judging graphic–text combination icons matching. The results of this research can help GUI designers by providing specific references in the custom design of adaptive icons for personalized users.

2. Materials and Methods

2.1. Experimental Design

This study adopts a 2 × 4 between-group experimental design and selects the following two independent variables: (1) individual user construal level traits—high construal level and low construal level; (2) graphic and text combination icon—concrete graphic + concrete text (Ci + Ct), concrete graphic + abstract text (Ci + At), abstract graphic + concrete text (Ai + Ct) and abstract graphic + abstract text (Ai + At); (3) text construal level—high construal level and low construal level. The dependent variables used in this study are as follows: (1) response time, the time from presenting the experimental material to judging consistency; (2) match rate, confirming the proportion of target icons and text match or their absence; (3) cognitive fluency, a subjective assessment of whether visual cognition feels fluid.

2.2. Experimental Material Design

We selected the common themes present in the following 176 interface icons from the internet, primarily concerning home appliances, animals and plants, vehicles, daily necessities, etc., for experimental simulation redesign. In extracting concrete and abstract graphics based on subject content, we mainly refer to Warell A’s product identification theory [48]. The theory proposes that the features of a complex product should be extracted sequentially from the gestalt (the outline of the product), the shape of the high-order parts, and the low-order parts. Therefore, outline features are mainly extracted in concrete icons, while in abstract icons, attribute features are mainly extracted.
Many previous studies have shown that the visual cognitive performance of icons may be affected by the following four dimensions: visual complexity, familiarity, semantic distance, and concreteness [17,49]. The visual complexity of a graphic symbol refers to the number of zigzag lines or details it contains; those with more are considered complex, those with less are considered simple [32]. The familiarity of a graphic symbol refers to the frequency with which users use the graphic symbol. The higher the frequency, the higher the familiarity, and vice versa [50]. The semantic distance of graphic symbols refers to the proximity of the relationship. The closer the relationship is, the stronger the relationship is, and vice versa [51]. Concreteness refers to the relationship between the meanings of graphic symbols and real things. The more obvious the connection, the more concrete, and vice versa [15]. Since concreteness is related to the degree of abstraction and representation of icons, semantic distance is associated with the construal level of words. Therefore, in establishing the relationship, we try to control the visual complexity and familiarity of all icons within a specific range and choose the common daily content as the theme.
The 176 themes were randomly divided into 2 groups for a redesign of the combined icons. Four student designers redesigned the icon representation part, and the text part was brainstormed by another four student designers. The 44 themes in each group consisted of a linear concrete or abstract graphic and a set of text. Line shapes and words were the same color, with RGB values of 255, 255 and 255. The outline pixels of the linear graphics were 4 px, the size was 30 × 30 mm, and the design style was kept uniform. The text size was 15 pt. The font was Boldface, with 2–3 Chinese characters. The background color RGB values were 0, 0 and 0. A total of 176 words were matched (44 concrete graphic + 44 concrete text, 44 concrete graphic + abstract text, 44 abstract graphic + concrete text, and 44 abstract graphic + 44 abstract text). The experimental material is shown in Figure 1.
We recruited 20 volunteers (10 males and 10 females, aged 19–26, mean = 23.1) to assess the visual complexity of 120 pairs of graphics and text (1 = very simple, 5 = very complex), familiarity (1 = very unfamiliar, 5 = very familiar), semantic distance (1 = not very closely related, 5 = closely related), and concreteness (1 = abstract, 5 = concrete). According to the scoring results, 160 graphic and text combination icons (40 for Ci + Ct, 40 for Ci + At, 40 for Ai + Ct, and 40 for Ai + At) were selected for formal experiments. As shown in Table 1, we used one-way ANOVA to analyze the evaluation results of the cognitive icon characteristics. The results show that there was no significant difference in visual complexity among the four groups of icons, with a mean of 3.13 and standard deviation of 0.725 (F (3, 159) = 0.822, p = 0.483). There was no significant difference in familiarity between the four icon groups; the mean was 3.24 and the standard deviation was 0.865 (F (3, 156) = 0.319, p = 0.812).
There was a significant difference between the four icon groups in terms of the semantic distance between the abstract and concrete graphics (F (1, 158) = 36.735, p < 0.05). There was a significant difference between the concreteness of the four icon groups with the abstract and concrete graphics (F (1, 158) = 11.988, p < 0.05). There was a significant difference between the high level and low level texts in the four icon groups (F (1, 158) = 52.31, p < 0.05). So, the manipulation of the experimental variables was successful. We eliminated the matching icons with abnormal scores in the four groups of icons. Finally, we screened out a total of 160 icons in 4 groups as the samples of this experiment.

2.3. Experimental Equipment and Participants

2.3.1. Apparatus

The experimental site was a human factor engineering lab without external interference. In the computer win10 system, the resolution of the display was 1080 × 1920. The experimental procedure used the psychological experimental software E-prime 3.0. The subjects sat in the human factors engineering lab without noise or soft light. The subjects were about 60 cm away from the display screen. The height of the display screen center was set at the same level as the subjects’ eyes, and the stimulus was presented in the center of the 1080 × 1920 resolution display screen.

2.3.2. Participants

The subjects first participated in the BIF Construal Level Personality Test (Vallacer and Wegner 1989) for classification. The BIF consists of 25 items, each with 2 abstract and concrete options that describe behavior. For example, the behavior “washing clothes” displays the following options: “looks clean (abstract explanation)” or “puts clothes in the washing machine (concrete explanation)”. The users choose the behavior they think is appropriate from 25 descriptions and select the option of an explanation for each question. An abstract answer scores 1, otherwise, it scores 0. The total score is a BIF index, with the median as the dividing line. The higher the score, the more abstract the user’s construal personality trait, and the lower the score, the more concrete the user’s personality trait.
As shown in Table 2, the subjects participating in the BIF are all undergraduates and postgraduates, totaling 80 people. After the experiment, users with fuzzy construal level traits close to the median were excluded, and 32 high construal level users were retained (16 males, 16 females), along with 32 low construal level users (14 males, 18 females). The subject age was between 24 and 28 years old (M = 25.72, SD = 1.27), and all had good computer experience. All were right-handed, with uncorrected or corrected visual acuity above 1.0. Before the experiment, the subjects were introduced to the experimental process, and the subjects were made familiar with the graphic materials (non-linguistic) of the experiment in advance, so that they could better complete the experimental tasks. The subjects gave gifts to express their gratitude after they completed the experiment.

2.4. Experimental Design and Procedures

As shown in Figure 2, the subjects were first asked to answer the BIF, and divided into two types of users with high and low construal levels. Then, the two types of users were correspondingly presented with four-category combined icons. The subjects remained comfortable, sitting with their eyes fixed on the stimulus object. After explaining the experimental precautions to the subjects, we opened the E-Prime running file; after this, the screen displays the experimental instructions, and the subjects read it carefully and pressed the space bar to start the practice. There were two sessions in the experiment for the subjects to practice. After the practice, they pressed the space bar to begin the experiment officially. The formal group were presented a total of 120 stimuli (see Figure 3).
After the experiment, three items adopted by Lee and Aaker [52] were used to measure the users’ cognitive fluency. The three items were “What do you think about the complexity of the information when you read it”, “What do you think about the complexity of the information when you try to understand it”, and “Do you agree that the information is easy to grab your attention”. Subjects rated their comprehension (1 = difficult to understand, 5 = easy to understand), processing (1 = difficult to handle, 5 = easy to handle), and attention (1 = hard to catch my attention, 5 = easy to catch my attention). We added and averaged these three items to arrive at the cognitive fluency index (see Appendix A).

2.5. Statistical Analysis

The behavioral data in this experiment included the three measures of response time, match rate judgment, and cognitive fluency. Response time refers to the average value of the response time of all stimuli presented to the subjects after removing outliers. Match rate refers to the proportion of instances in which the subjects believed that the pattern matched the word out of the total instances in which the stimulus appeared under this condition.
In this study, multivariate analysis of variance was used to measure the data and examine the differences between the effects of the combined icon type (i.e., Ci + Ct, Ci + At, Ai + Ct, Ai + At) and user construal levels (i.e., high construal level and low construal level) on visual cognitive response time, match rate judgments, cognitive fluency, and their interactions. The SPSS 22.0 software package (IBM, Armonk, NY, USA) was used to analyze the variance and mediation effect data statistically.

3. Results

Table 3 shows the mean, standard deviation of the reaction time, matching rate, and cognitive fluency of the forty subjects in the four-groups design. Table 4 shows the ANOVA results for combined icon types and user construal levels.

3.1. Response Time

Different combination icons had a significant impact on the response times of all the subjects (F (3, 248) = 332.055, p < 0.05, η2 = 0.801), and the response times of different combination icons were as follows: Ci + Ct, Ci + At, Ai + Ct, Ai + At. The response time of Ci + Ct was the shortest, which was significantly lower than that of the three other types of icons, which were as follows: Ci + At, Ai + Ct, Ai + At (p < 0.05). The response time of Ci + At was significantly less than those of Ai + Ct and Ai + At (p < 0.05). The response time of Ai + Ct was significantly more than that of Ci + At (p < 0.05), but significantly less than that of Ai + At (p < 0.05). The response time of Ai + At was significantly more than those of Ci + Ct, Ci + At and Ai + Ct (p < 0.05).
Figure 4 shows that the response times of the combined icon type and the construal level interact (F (3, 248) = 9.305, p < 0.05, η2 = 0.101). Compared with the other icon types, the response time of Ci + Ct was the shortest (p < 0.05), but there was no significant difference between the two groups (p = 0.848). With regard to the response time of Ai + At, between the two groups, that of the high construal level was significantly shorter than that of the low construal level; in addition, for the high construal level subjects, there was little difference in the response time between Ci + At and Ai + Ct, with two types of icons (p = 0.738). For the low construal level subjects, the response time of Ci + At was significantly longer than that of Ai + Ct (p < 0.05).

3.2. Match Rate

Different combination icons had a significant effect on the match rate cognition of all subjects (F (3, 248) = 1687.105, p < 0.05, η2 = 0.953). The recognition of the matching of combination icon types was ranked as follows: Ci + Ct, Ci + At, Ai + Ct and Ai + At. The match rate of Ci + Ct was the highest, followed by Ci + At, Ai + Ct and Ai + At (p < 0.05). The match rate of Ci + At was significantly lower than that of Ci + Ct (p < 0.05), but it was significantly higher than that of Ai + Ct and Ai + At (p < 0.05); the match rate of Ai + Ct was significantly lower than that of Ci + At (p < 0.05), but significantly higher than Ai + At (p < 0.05). The match rate of Ai + At was significantly lower than that of Ci + Ct, Ci + At and Ai + Ct (p < 0.05).
Figure 5 shows that there is an interaction between the match rate of the combined icon types and construal levels (F (3, 248) = 57.68, p < 0.05, η2 = 0.411). Compared with other icon types, in the two groups of subjects with different construal levels, the Ci + Ct match rate was the highest (p < 0.05), but there was little difference in matching cognition between the two groups of subjects (p = 0.422). There is a significant difference between the two groups of subjects’ visual cognition match rates between Ci + At, Ai + Ct and Ai + At (p < 0.05). In detail, the visual cognition match rate was better for Ci + At than Ai + Ct, while that for Ai + Ct was better than that for Ai + At (p < 0.05).

3.3. Mediation by Cognitive Fluency

As shown in Table 5, we tested the mediation effect with the Bootstrap sampling technique and used PROCESS model 7 on 5000 replicates. First, we set the Ci + Ct in the design type as the virtualized coding benchmark, and the results show that the design type (Ci + At, Ai + Ct, Ai + At) has a significant direct impact on the response time (β = 1.416, Boot-SE = 0.079, p < 0.05; β = 1.453, Boot-SE = 0.086, p < 0.05; β = 1.703, Boot-SE = 0.107, p < 0.05) of 95% of the three tests. The confidence intervals are in the ranges (1.272, 1.554), (1.349, 1.570) and (1.520, 1.895), respectively.
Secondly, the construal level feature has a moderating effect on some design types (Ai + Ct and Ai + At) and cognitive fluency (β = 0.899, Boot-SE = 0.192, p < 0.05; β = 1.123, Boot-SE = 0.224, p < 0.05), and the 95% confidence intervals of the two tests are in the ranges of (0.520, 1.267) and (0.681, 1.569), respectively. It is not surprising that the construal level feature has a moderating effect on the design type (Ci + At) and cognitive fluency (β = 0.184, Boot-SE = 0.152, p > 0.05). The 95% confidence interval is in the range (−0.108, 0.487).
Finally, the results show that cognitive fluency mediates the interactive effect of some design types (Ai + Ct, Ai + At) and explanation level on the response time, and has a significant impact (β = −0.303, Boot-SE = 0.076, p < 0.05; β = −0.378, Boot-SE = 0.096, p < 0.05). The 95% confidence intervals of the two tests were in the ranges (−0.455, −0.163) and (−0.584, −0.199), respectively. Cognitive fluency does not clearly mediate the interaction effect between design type (Ci + At) and construal level on response time (β = −0.062, Boot-SE = 0.053, p > 0.05). The 95% confidence interval is in the range (−0.172, 0.035). The mediation model is shown in Figure 6.
As shown in Table 6, we also used Ci + Ct in the design type as the virtualized coding benchmark. The results show that the design type (Ci + At, Ai + Ct, Ai + At) has a significant direct impact on the matching rate (β = −0.682, Boot-SE = 0.047, p < 0.05; β = −1.053, Boot-SE = 0.052, p < 0.05; β = −1.898, Boot-SE = 0.064, p < 0.05), and the 95% confidence intervals of the three tests are in the ranges (−0.732, −0.632), (−1.172, −0.944) and (−2.055, −1.761), respectively.
Secondly, the construal level feature has a moderating effect on some design types (Ai + Ct, Ai + At) and cognitive fluency (β = 0.899, Boot-SE = 0.192, p < 0.05; β = 1.123, Boot-SE = 0.220, p < 0.05), and the 95% confidence intervals of the two tests are in the ranges (0.521, 1.276) and (0.681, 1.547), respectively. It is not surprising that the construal level feature has a moderating effect on the design type (Ci + At) and cognitive fluency (β = 0.184, Boot-SE = 0.151, p > 0.05). The 95% confidence interval is in the range (−0.107, 0.490).
Finally, the results show that cognitive fluency mediates the interaction effect of some design types (Ai + Ct, Ai + At) and explanation level on the matching rate, and has a significant impact (β = 0.316, Boot-SE = 0.072, p < 0.05; β = 0.394, Boot-SE = 0.094, p < 0.05). The 95% confidence intervals of the two tests were in the ranges (0.178, 0.461) and (0.218, 0.585), respectively. Cognitive fluency does not clearly mediate the interaction effect of design type (Ci + At) and construal level on matching rate (β = 0.065, Boot-SE = 0.54, p > 0.05). The 95% confidence interval is in the range (−0.037, 0.173). The mediation model is shown in Figure 7.

4. Discussion

4.1. Effects of Combined Icon Types on Visual Cognitive Performance of Construal Level Traits Users

Among the four combined icons of Ci + Ct, Ci + At, Ai + Ct and Ai + At, all users achieved the best response time, match rate, and cognitive fluency with Ci + Ct, followed by Ci + At, Ai + Ct and Ai + At. Our findings are consistent with previous research. When the icons are presented in the form of Ci + Ct, because of the high reduction degree of concrete graphics to physical things and the straightness of the text, the graphics’ visual attributes and the text’s semantic attributes will be quickly encoded directly into the human brain [53]. However, the graphics and text in the combined icon are the same in the source and target domains, or at least belong to the same category. Therefore, the graphics–text semantic network is very close. Therefore, when concrete graphics match concrete text, it is easier to understand, and can lead to the best visual cognitive performance, which can be explained by the dual-coding theory [54]. In the two types of icons of Ci + At and Ai + Ct, the graphic is related to the texts’ source and target domain, but it is not in the same category. Hence, the user needs to convert the abstract concept of the textual semantics in the brain to match the concrete graphic, or convert the extended abstract graphic to the corresponding real object, and match this with the concrete text [41,55,56]. Among these two types of icons, either the source domain or the target domain needs to be converted, and this means that more cognitive resources are required relative to the type of Ci + Ct. However, the processing of graphic attributes in the human brain is more effective than that of text, so in the two types of icons, the cognitive performance of Ci + At is better than that of Ai + Ct [57]. In Ai + At, both the graphic and text attributes need to be retrieved and converted in the brain before matching, so it is easy to understand why the visual cognition of this type of icon is lower than that of the first three types, and more complex [47].

4.2. The Interactive Effect of Combined Icon Types and Construal Level Traits on Users’ Visual Cognitive Performance

In the visual cognition of the combined icon, the subjects judged whether the abstract or concrete graphic and the corresponding text clues were consistent. When both graphic clues and text clues are concrete, the response time, match rate, and cognitive fluency of the two types of users do not significantly differ. The cognitive load is minimal, and almost indistinguishable.
Construal level is a stable personality trait. Users with high construal levels are better at initiating abstract coding to represent brain information. On the contrary, users with low construal levels are more inclined to concrete coding. When the graphic clues are Ci + At or Ai + Ct, these are non-matching clues. For both types of users, Ci + At extraction is slightly better than Ai + Ct extraction; this may be because, compared with abstract graphics, texts use more cognitive resources in the process of semantic transformation [58].
In the two types of icon cognition, because they contain unimodal abstract information, in the cognitive processing of abstract modalities, the cognitive transformation of information in high construal level users is better, and they can more quickly and smoothly establish integrated relationships than those with a low construal level. When non-matching clues are presented, a difference is generated between the two types of users, and icons with abstract graphs outperform icons with abstract text [59].
However, with Ci + At and Ai + Ct type icons, in double-information clues, only one item needs to be converted and matched. When the type is Ai + At, although it is a matching clue, the abstract graphic needs to go through internal representation and semantic encoding to become extended semantics, and the abstract semantics also require an abstract concept to transcode the straight concept. Then, it is required to establish a matching relationship between the graphic attributes and the text semantic attributes, which prolongs the response time [60]. Therefore, even if the first two types of icons are non-matching clue icons, their cognitive performance is still better than that of the matching clues. When double abstract clues are present in matching clues, this further increases the visual cognitive load and difficulty encountered by low construal level users. Therefore, in the evaluation of the match rate and cognitive fluency of these three icon types, the visual cognitive performance of high construal level users is better than that of low construal level users, and the difference in performance between Ai + At extraction is more obvious.

4.3. Limitation and Future Research

There are some limitations to our study. First of all, the experimental materials selected in this study are abstract or concrete graphics and Chinese characters, and the research results can only serve as the terminal interface of Chinese applications. In future research, this should be expanded to other languages and combination icons to better serve the visual cognition of graphic and text combination icons by users with different cultural backgrounds. Secondly, construal level personality traits are the implicit and stable cognitive traits of users. How to quickly identify user types and accurately present a usable combination icon during the operation of the system remains a challenge. It is an urgent problem that needs to be solved in the future in order to combine advanced technologies, such as artificial intelligence, in actual program development. Thirdly, the design styles of icons have diversified, in terms of line shape, surface shape, flattening, and 3D or graphic–text combined icons. In addition, our experimental data assume that users are relatively equal in terms of their familiarity. When two types of icons have different levels of familiarity, and especially when users with low construal levels are familiar with abstract combination icons, in this way, their visual cognition can increase significantly. In this case, how will the cognitive performances of users with different construal levels change? The combined external design features and users’ internal cognitive characteristics will interact with each other, and so more research and exploration are needed. Furthermore, due to the limitation of the size of the cohort, this study is limited to the student designer group, and a larger socially oriented sample size may be required to test our findings in the future. All in all, the results of this study may be more applicable to system program development for user groups with different construal levels of Chinese traits.

5. Conclusions

Although many studies have shown that users have better levels of visual cognition of concrete icons, the visual cognition performance of graphic–text combined icons is unclear. The effect of the interaction between combined design style and user construal level traits on the behavioral performance of visual cognition is unclear. Therefore, this study uses subjective and objective data from behavioral experiments to measure the visual cognitive response time and match rate of combined icons by users with two types of construal level-related personality traits, and examines the mediating role of cognitive fluency. Through experimental research, we here propose that, in future design strategies for the adaptive interface of groups with personality differences, first of all, in the design of combined icons, the designers should choose the combination method where the graphics concretely match with the texts, in order to achieve the best possible results. It is possible to lighten the user’s cognitive load. Secondly, although matching across modal clues has been consistently emphasized in many designs, it is not applicable in some cases, especially compared with high construal level users, and presentation to low construal level users should be avoided—this rules out the combined icon method Ai + At. Finally, when the concrete clues of graphic and text are inconsistent, the combined icon-type design should be favored, namely, Ci + At rather than Ai + Ct.

Author Contributions

Conceptualization, Y.Z. and M.C.; methodology, Y.Z. and Y.L. (Ying Li); software, Y.Z. and Y.L. (Ying Li); validation, Q.G. and Z.Z.; formal analysis, Y.Z. and Y.L. (Yun Lin); investigation, Y.Z. and M.C.; resources, Z.Z.; data curation, Z.Z.; writing—original draft preparation, Y.Z.; writing review and editing, Y.Z. and Y.L. (Yun Lin); visualization, Y.Z.; supervision, Z.Z. All authors have read and agreed to the published version of the manuscript.

Funding

The paper is supported jointly by the Natural Science Foundation of China (No. 52205264, 71871056, 52005251), and the Jiangsu province graduate scientific research innovation project (KYCX18_0069), the Shanghai Pu jiang Program (21PJC032).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The data are contained within the article.

Conflicts of Interest

The authors declare no conflict of interest.

Appendix A

Applsci 12 10111 g0a1 550
Figure A1. User cognitive fluency scale.
Figure A1. User cognitive fluency scale.
Applsci 12 10111 g0a1

References

  1. Shneiderman, B.; Plaisant, C. Designing the User Interface: Strategies for Effective Human-Computer Interaction, 4th ed.; Pearson/Addison Wesley: Boston, MA, USA, 2004; ISBN 978-0-321-19786-3. [Google Scholar]
  2. Muter, P.; Mayson, C. The Role of Graphics in Item Selection from Menus. Behav. Inf. Technol. 1986, 5, 89–95. [Google Scholar] [CrossRef]
  3. Huang, S.-M.; Shieh, K.-K.; Chi, C.-F. Factors Affecting the Design of Computer Icons. Int. J. Ind. Ergon. 2002, 29, 211–218. [Google Scholar] [CrossRef]
  4. Gatsou, C.; Politis, A.; Zevgolis, D. The Importance of Mobile Interface Icons on User Interaction. Int. J. Comput. Sci. Appl. 2012, 9, 92–107. Available online: https://www.semanticscholar.org/paper/The-Importance-of-Mobile-Interface-Icons-on-User-Gatsou-Politis/414464f647492d6efca2144f611f4e4362105b48 (accessed on 29 July 2022).
  5. Norman, D. Affordance, conventions, and design. Interactions 1999, 6, 38–42. [Google Scholar] [CrossRef]
  6. Nielsen, J.; Molich, R. Heuristic evaluation of user interfaces. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, Washington, DC, USA, 1–5 April 1990; pp. 249–256. [Google Scholar]
  7. Nielsen, J. Enhancing the explanatory power of usability heuristics. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, Boston, MA, USA, 24–28 April 1994; pp. 152–158. [Google Scholar]
  8. Debevc, M.; Meyer, B.; Donlagic, D.; Svecko, R. Design and Evaluation of an Adaptive Icon Toolbar. User Model User-Adap. Inter. 1996, 6, 1–21. [Google Scholar] [CrossRef]
  9. Miraz, M.H.; Ali, M.; Excell, P.S. Adaptive User Interfaces and Universal Usability through Plasticity of User Interface Design. Comput. Sci. Rev. 2021, 40, 487–519. [Google Scholar] [CrossRef]
  10. Kolekar, S.V.; Pai, R.M.; Pai, M.M.M. Rule Based Adaptive User Interface for Adaptive E-Learning System. Educ. Inf. Technol. 2019, 24, 613–641. [Google Scholar] [CrossRef]
  11. Ming, C.; Wu, Z.; Gu, H.; Chen, C.; Zhang, Y. The Effects of Luminance Contrast and Color Combination on Icon Cognitive Performance. Color Res. Appl. 2022, 47, 498–506. [Google Scholar] [CrossRef]
  12. Lin, H.; Hsieh, Y.-C.; Wu, F.-G. A Study on the Relationships between Different Presentation Modes of Graphical Icons and Users’ Attention. Comput. Hum. Behav. 2016, 63, 218–228. [Google Scholar] [CrossRef]
  13. Shen, Z.; Zhang, L.; Li, R.; Hou, J.; Liu, C.; Hu, W. The Effects of Color Combinations, Luminance Contrast, and Area Ratio on Icon Visual Search Performance. Displays 2021, 67, 101999. [Google Scholar] [CrossRef]
  14. Huang, K.-C.; Chen, C.-F.; Chiang, S.-Y. Icon Flickering, Flicker Rate, and Color Combinations of an Icon’s Symbol/Background in Visual Search Performance. Percept. Mot. Ski. 2008, 106, 117–127. [Google Scholar] [CrossRef]
  15. Shen, Z.; Zhang, L.; Li, R.; Liang, R. The Effects of Icon Internal Characteristics on Complex Cognition. Int. J. Ind. Ergon. 2020, 79, 102990. [Google Scholar] [CrossRef]
  16. Ng, A.W.Y.; Chan, A.H.S. Visual and Cognitive Features on Icon Effectiveness. In Proceedings of the International Multiconference of Engineers and Computer Scientists, Hong Kong, China, 19–21 March 2008; Newswood Limited Hong Kong: Hong Kong, China, 2008; Volume 4. [Google Scholar]
  17. McDougall, S.J.P.; de Bruijn, O.; Curry, M.B. Exploring the Effects of Icon Characteristics on User Performance: The Role of Icon Concreteness, Complexity, and Distinctiveness. J. Exp. Psychol. Appl. 2000, 6, 291–306. [Google Scholar] [CrossRef]
  18. Stotts, D.B. The Usefulness of Icons on the Computer Interface: Effect of Graphical Abstraction and Functional Representation on Experienced and Novice Users. Proc. Hum. Factors Ergon. Soc. Annu. Meet. 1998, 42, 453–457. [Google Scholar] [CrossRef]
  19. Chajadi, F.; Uddin, S.; Gutwin, C. Effects of Visual Distinctiveness on Learning and Retrieval in Icon Toolbars. Ph.D. Thesis, University of Saskatchewan, Saskatoon, SK, Canada, 2022. [Google Scholar]
  20. Rahrovani, S.; Mirzabeigi, M.; Abbaspour, J. The Concreteness of Searching Module Icons and Their Effectiveness in Digital Library Applications. Electron. Libr. 2018, 36, 800–810. [Google Scholar] [CrossRef]
  21. Schröder, S.; Ziefle, M. Effects of Icon Concreteness and Complexity on Semantic Transparency: Younger vs. Older Users. In Computers Helping People with Special Needs; Miesenberger, K., Klaus, J., Zagler, W., Karshmer, A., Eds.; Lecture Notes in Computer Science; Springer: Berlin/Heidelberg, Germany, 2008; Volume 5105, pp. 90–97. ISBN 978-3-540-70539-0. [Google Scholar]
  22. Stammers, R.; Hoffman, J. Transfer between Icon Sets and Ratings of Icon Concreteness and Appropriateness. Proc. Hum. Factors Soc. Annu. Meet. 1991, 35, 354–358. [Google Scholar] [CrossRef]
  23. Wen, K.; Gu, X.; Cheng, Q. Learning Dual Semantic Relations with Graph Attention for Image-Text Matching. IEEE Trans. Circuits Syst. Video Technol. 2021, 31, 2866–2879. [Google Scholar] [CrossRef]
  24. Liu, R.; Zhao, Y.; Wei, S.; Zheng, L.; Yang, Y. Modality-Invariant Image-Text Embedding for Image-Sentence Matching. ACM Trans. Multimed. Comput. Commun. Appl. 2019, 15, 1–19. [Google Scholar] [CrossRef]
  25. Chi, C.-F.; Dewi, R.S. Matching Performance of Vehicle Icons in Graphical and Textual Formats. Appl. Ergon. 2014, 45, 904–916. [Google Scholar] [CrossRef]
  26. Dosso, C.; Chevalier, A. How Do Older Adults Process Icons during a Navigation Task? Effects of Aging, Semantic Distance, and Text Label. Educ. Gerontol. 2021, 47, 132–147. [Google Scholar] [CrossRef]
  27. Osman, A.H.; Barukub, O.M. Graph-Based Text Representation and Matching: A Review of the State of the Art and Future Challenges. IEEE Access 2020, 8, 87562–87583. [Google Scholar] [CrossRef]
  28. Jin, W.; Zhao, Z.; Zhang, P.; Zhu, J.; He, X.; Zhuang, Y. Hierarchical Cross-Modal Graph Consistency Learning for Video-Text Retrieval. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, Virtual Event Canada, 11 July 2021; ACM: New York, NY, USA, 2021; pp. 1114–1124. [Google Scholar]
  29. Spilcke-Liss, J.; Zhu, J.; Gluth, S.; Spezio, M.; Gläscher, J. Semantic Incongruency Interferes with Endogenous Attention in Cross-Modal Integration of Semantically Congruent Objects. Front. Integr. Neurosci. 2019, 13, 53. [Google Scholar] [CrossRef]
  30. Sojka, J.Z.; Giese, J.L. The In¯uence of Personality Traits on the Processing of Visual and Verbal Information. Psychol. Mark. 2006, 23, 995–1014. [Google Scholar] [CrossRef]
  31. Sarsam, S.M.; Al-Samarraie, H. Towards Incorporating Personality into the Design of an Interface: A Method for Facilitating Users’ Interaction with the Display. User Model User Adap. Inter. 2018, 28, 75–96. [Google Scholar] [CrossRef]
  32. Kocaoğlu, R.; Olguntürk, N. Color and Visual Complexity in Abstract Images. Color Res. Appl. 2018, 43, 952–957. [Google Scholar] [CrossRef] [Green Version]
  33. Satcharoen, K. An Investigation of Computer Icon Design. In Proceedings of the 21st Conference of Open Innovations FRUCT, Helsinki, Finland, 6–10 November 2017; pp. 489–491. [Google Scholar]
  34. Trope, Y.; Liberman, N. Temporal Construal. Psychol. Rev. 2003, 110, 403–421. [Google Scholar] [CrossRef]
  35. Durkin, C.; Hartnett, E.; Shohamy, D.; Kandel, E.R. An objective evaluation of the beholder’s response to abstracts and figurative art based on construal level theory. Psychol. Cogn. Sci. 2020, 33, 19809–19815. [Google Scholar] [CrossRef]
  36. Dogan, M.; Erdogan, B.Z. Effects of Congruence between Individuals’ and Hotel Commercials’ Construal Levels on Purchase Intentions. J. Hosp. Mark. Manag. 2020, 29, 987–1007. [Google Scholar] [CrossRef]
  37. Steinhart, Y.; Mazursky, D.; Kamins, M.A. The “Temporal-Processing-Fit Effect”: The Interplay between Regulatory State, Temporal Distance, and Construal Levels. SSRN J. 2013, 3, 315–335. [Google Scholar] [CrossRef]
  38. Yao, F.-S.; Shao, J.-B.; Zhang, H. Is Creative Description Always Effective in Purchase Intention? The Construal Level Theory as a Moderating Effect. Front. Psychol. 2021, 12, 1–12. [Google Scholar] [CrossRef]
  39. Oppenheimer, D.M. The Secret Life of Fluency. Trends Cogn. Sci. 2008, 12, 237–241. [Google Scholar] [CrossRef]
  40. Wurtz, P.; Reber, R.; Zimmermann, T.D. The Feeling of Fluent Perception: A Single Experience from Multiple Asynchronous Sources. Conscious. Cogn. 2008, 17, 171–184. [Google Scholar] [CrossRef]
  41. Reber, R.; Wurtz, P.; Zimmermann, T.D. Exploring “Fringe” Consciousness: The Subjective Experience of Perceptual Fluency and Its Objective Bases. Conscious. Cogn. 2004, 13, 47–60. [Google Scholar] [CrossRef]
  42. Duan, R.; Takahashi, B.; Zwickle, A. How Effective Are Concrete and Abstract Climate Change Images? The Moderating Role of Construal Level in Climate Change Visual Communication. Sci. Commun. 2021, 43, 358–387. [Google Scholar] [CrossRef]
  43. Sungur, H.; Hartmann, T.; van Koningsbruggen, G.M. Abstract Mindsets Increase Believability of Spatially Distant Online Messages. Front. Psychol. 2016, 7, 1–9. [Google Scholar] [CrossRef] [Green Version]
  44. Cancino-Montecinos, S.; Björklund, F.; Lindholm, T. Dissonance and Abstraction: Cognitive Conflict Leads to Higher Level of Construal: Dissonance and Abstraction. Eur. J. Soc. Psychol. 2018, 48, 100–107. [Google Scholar] [CrossRef]
  45. Li, X.; Chen, G.; Yang, C. How Cognitive Conflict Affects Judgments of Learning: Evaluating the Contributions of Processing Fluency and Metamemory Beliefs. Mem. Cogn. 2021, 49, 912–922. [Google Scholar] [CrossRef]
  46. Massara, F.; Scarpi, D.; Porcheddu, D. Can Your Advertisement Go Abstract Without Affecting Willingness to Pay?: Product-Centered versus Lifestyle Content In Luxury Brand Print Advertisements. JAR 2020, 60, 28–37. [Google Scholar] [CrossRef]
  47. Kuipers, J.R.; Jones, M.W.; Thierry, G. Abstract Images and Words Can Convey the Same Meaning. Sci. Rep. 2018, 8, 7190. [Google Scholar] [CrossRef] [Green Version]
  48. Warell, A. Design Syntactics: A Functional Approach to Visual Product form Theory, Model, and Methods. Ph.D. Thesis, Chalmers University of Technology, Göteborg, Sweden, 2021. [Google Scholar]
  49. Mcdougall, S.J.P.; Curry, M.B.; de Bruijn, O. Measuring Symbol and Icon Characteristics: Norms for Concreteness, Complexity, Meaningfulness, Familiarity, and Semantic Distance for 239 Symbols. Behav. Res. Methods Instrum. Comput. 1999, 31, 487–519. [Google Scholar] [CrossRef] [Green Version]
  50. Shen, Z.; Zhang, L.; Xiao, X.; Li, R.; Liang, R. Icon Familiarity Affects the Performance of Complex Cognitive Tasks. i-Perception 2020, 11, 204166952091016. [Google Scholar] [CrossRef]
  51. Silvennoinen, J.; Kujala, T.; Jokinen, J.P. Semantic distance as a critical factor in icon design for in-car infotainment systems. Appl. Ergon. 2017, 65, 369–381. [Google Scholar] [CrossRef] [Green Version]
  52. Lee, A.Y.; Aaker, J.L. Bringing the frame into focus: The influence of regulatory fit on processing fluency and persuasion. J. Pers. Soc. Psychol. 2004, 2, 205–218. [Google Scholar] [CrossRef] [Green Version]
  53. Ding, J.; Liu, W.; Yang, Y. The Influence of Concreteness of Concepts on the Integration of Novel Words into the Semantic Network. Front. Psychol. 2017, 8, 2111. [Google Scholar] [CrossRef] [Green Version]
  54. Sadoski, M.; Kealy, W.A.; Goetz, E.T.; Paivio, A. Concreteness and Imagery Effects in the Written Composition of Definitions. J. Educ. Psychol. 1997, 89, 518–526. [Google Scholar] [CrossRef]
  55. Hayashi, A.; Okamoto, Y.; Yoshimura, S.; Yoshino, A.; Toki, S.; Yamashita, H.; Matsuda, F.; Yamawaki, S. Visual imagery while reading concrete and abstract Japanese kanji words: An fMRI study. Neurosci. Res. 2014, 79, 61–66. [Google Scholar] [CrossRef] [Green Version]
  56. Papagno, C. The neural correlates of abstract and concrete words. Handb. Clin. Neurol. 2022, 187, 263–275. [Google Scholar] [CrossRef]
  57. Hemati, S.; Hossein-Zadeh, G.-A. Distinct Functional Network Connectivity for Abstract and Concrete Mental Imagery. Front. Hum. Neurosci. 2018, 12, 515. [Google Scholar] [CrossRef] [Green Version]
  58. West, W.C.; Holcomb, P.J. Imaginal, semantic, and surface-level processing of concrete and abstract words: An electrophysiological investigation. J. Cogn. Neurosci. 2000, 12, 1024–1037. [Google Scholar] [CrossRef]
  59. Wang, X.; Bi, Y. Idiosyncratic Tower of Babel: Individual Differences in Word-Meaning Representation Increase as Word. Abstractness Increases. Psychol. Sci. 2021, 32, 1617–1635. [Google Scholar] [CrossRef] [PubMed]
  60. Davis, C.P.; Altmann, G.T.M.; Yee, E. Situational systematicity: A role for schema in understanding the differences between. abstract and concrete concepts. Cogn. Neuropsychol. 2020, 37, 142–153. [Google Scholar] [CrossRef] [PubMed]
Applsci 12 10111 g001 550
Figure 1. Examples of four groups of experimental materials. Subfigure: (a) concrete graphic + concrete text; (b) concrete graphic + abstract text; (c) abstract graphic + concrete text; (d) abstract graphic + abstract text.
Figure 1. Examples of four groups of experimental materials. Subfigure: (a) concrete graphic + concrete text; (b) concrete graphic + abstract text; (c) abstract graphic + concrete text; (d) abstract graphic + abstract text.
Applsci 12 10111 g001
Applsci 12 10111 g002 550
Figure 2. Schematic of the experimental process.
Figure 2. Schematic of the experimental process.
Applsci 12 10111 g002
Applsci 12 10111 g003 550
Figure 3. Experimental scene image.
Figure 3. Experimental scene image.
Applsci 12 10111 g003
Applsci 12 10111 g004 550
Figure 4. Response time for four types of combination design icons in users of two types of construal level.
Figure 4. Response time for four types of combination design icons in users of two types of construal level.
Applsci 12 10111 g004
Applsci 12 10111 g005 550
Figure 5. Match rate for four types of combination design icons in two groups of users with different construal levels.
Figure 5. Match rate for four types of combination design icons in two groups of users with different construal levels.
Applsci 12 10111 g005
Applsci 12 10111 g006 550
Figure 6. The mediating model of cognitive fluency in response time. Note: * p ≤ 0.05.
Figure 6. The mediating model of cognitive fluency in response time. Note: * p ≤ 0.05.
Applsci 12 10111 g006
Applsci 12 10111 g007 550
Figure 7. The mediating model of cognitive fluency in match rate. Note: * p ≤ 0.05.
Figure 7. The mediating model of cognitive fluency in match rate. Note: * p ≤ 0.05.
Applsci 12 10111 g007
Table
Table 1. Mean scores and standard deviation for icon and text complexity, familiarity, semantic distance and concreteness.
Table 1. Mean scores and standard deviation for icon and text complexity, familiarity, semantic distance and concreteness.
Ci + Ct (n = 40)Ci + At (n = 40)Ai + Ct (n = 40)Ai + At (n = 40)
M±SDM±SDM±SDM±SDFp
Graphic complexity3.00±0.603.10±0.713.15±0.773.25±0.810.8220.483
Graphic familiarity3.25±0.903.13±0.793.30±0.913.27±0.870.3190.812
Concrete icons (n = 80)Abstract icons (n = 80)
M±SDM±SDtp
Graphic semantic distance3.85±0.792.95±0.737.9000.000
Graphic concreteness3.91±0.742.99±0.867.2330.000
Concrete text (n = 80)Abstract text (n = 80)
M±SDM±SDtp
Text concreteness3.625±0.682.92±0.766.0610.000
Table
Table 2. Demographic information of high construal level group and low construal level group.
Table 2. Demographic information of high construal level group and low construal level group.
High Construal LevelLow Construal Level
n32 32
SexMale = 16; female = 16Male = 14; female = 18
AgeM = 25.38; SD = 1.13M = 26.06; SD = 1.32
Table
Table 3. Mean (standard deviation) of response time and match rate for different combinations of design type icons and construal level.
Table 3. Mean (standard deviation) of response time and match rate for different combinations of design type icons and construal level.
Response Time(s)Match Rate
M±SDM±SD
Ci + Ct789.253±8.2794.699±0.277
Ci + At1036.310±94.5934.479±0.293
Ai + Ct1072.674±75.0983.984±0.587
Ai + At1161.170±101.9123.318±0.726
Ci + Ct × HCL787.580±7.9724.709±0.291
Ci + Ct × LCL790.927±8.3654.688±0.268
Ci + At × HCL1016.735±32.9734.479±0.294
Ci + At × LCL1055.884±127.6964.323±0.392
Ai + Ct × HCL1039.184±42.0084.323±0.392
Ai + Ct × LCL1106.164±85.9143.645±0.555
Ai + At × HCL1096.392±65.7343.739±0.492
Ai + At × LCL1225.948±90.1302.897±0.680
Table
Table 4. ANOVA results for combination design type and construal level.
Table 4. ANOVA results for combination design type and construal level.
Source of VarianceResponse TimeMatch Rate
Fpη2Fpη2
Combination design type332.0550.0000.8011687.1050.0000.953
Construal level46.7400.0000.159430.0620.0000.634
Design type × construal level9.3050.0000.10157.680.0000.411
Table
Table 5. The mediating role of the effects of cognitive fluency in combination design style and construal level on response time.
Table 5. The mediating role of the effects of cognitive fluency in combination design style and construal level on response time.
Direct Effect of Combination Design Type
CoeffBoot-S.EBootLLCIBootULCI
Constant−1.1430.033−1.207−1.079
Ci + At1.4160.0711.2721.554
Ai + Ct1.4530.0561.3491.570
Ai + At1.7030.0961.5201.895
Indirect Effect of Combination Design Type × Construal Level
CoeffBoot-S.EBootLLCIBootULCI
Ci + At−0.0620.053−0.1720.035
Ai + Ct−0.3030.076−0.455−0.163
Ai + At−0.3780.096−0.584−0.199
Index of Moderated Mediation
IndexBoot-S.EBootLLCIBootULCI
−0.3370.037−0.407−0.264
Table
Table 6. The mediating effect of cognitive fluency in combination design style and construal level on match rate.
Table 6. The mediating effect of cognitive fluency in combination design style and construal level on match rate.
Direct Effect of Combination Design Type
CoeffBoot-S.EBootLLCIBootULCI
Constant0.9080.0290.8510.966
Ci + At−0.682−0.682−0.732−0.632
Ai + Ct−1.053−1.056−1.172−0.944
Ai + At−1.898−1.899−2.055−1.761
Indirect Effect of Combination Design Type × Construal Level
CoeffBoot-S.EBootLLCIBootULCI
Ci + At0.0650.054−0.0370.173
Ai + Ct0.3160.0720.1780.461
Ai + At0.3940.0940.2180.585
Index of Moderated Mediation
IndexBoot-S.EBootLLCIBootULCI
0.3510.2820.2930.405
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Zhu, Y.; Li, Y.; Lin, Y.; Chen, M.; Guo, Q.; Zhang, Z. Research on the Influence of User and Graphic–Text Combined Icon Construal Level Fitting on Visual Cognition. Appl. Sci. 2022, 12, 10111. https://doi.org/10.3390/app121910111

AMA Style

Zhu Y, Li Y, Lin Y, Chen M, Guo Q, Zhang Z. Research on the Influence of User and Graphic–Text Combined Icon Construal Level Fitting on Visual Cognition. Applied Sciences. 2022; 12(19):10111. https://doi.org/10.3390/app121910111

Chicago/Turabian Style

Zhu, Yanfei, Ying Li, Yun Lin, Mo Chen, Qi Guo, and Zhisheng Zhang. 2022. "Research on the Influence of User and Graphic–Text Combined Icon Construal Level Fitting on Visual Cognition" Applied Sciences 12, no. 19: 10111. https://doi.org/10.3390/app121910111

APA Style

Zhu, Y., Li, Y., Lin, Y., Chen, M., Guo, Q., & Zhang, Z. (2022). Research on the Influence of User and Graphic–Text Combined Icon Construal Level Fitting on Visual Cognition. Applied Sciences, 12(19), 10111. https://doi.org/10.3390/app121910111

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop