1. Research Background

With the prevalence of multimodal discourse analysis since the 1990s in Western countries, scholars from a variety of research fields have shown great interest in this field (Cheng, 2017, p. 1). Therefore, much progress with respect to multimodal discourse studies has been achieved worldwide.

However, based on the literature review of previous studies on multimodal discourse analysis, the following three points in the field of multimodal discourse studies have been very prominent. Firstly, virtually all scholars doing multimodal discourse analysis mainly adopt visual grammar formulated by Kress & van Leeuwen (1996/2006) as their theoretical foundation (Wei, 2009, pp. 8-9; Zhang, 2013, p. 29), which is grounded on systemic functional linguistics (hereafter abbreviated as SFL). However, multimodal discourse analysis has borrowed the thinking from linguistics, but it is not a linguistic study (Kress & van Leeuwen, 2006, p. 19). Just as the name of “multimodality” suggests,multimodal discourse analysis bears the interdisciplinary tendency from the very start,and it should be closely related with art graphic design. Secondly, as far as the nature of multimodal analysis is concerned, it is the case that viewers are actually decoding what designers have encoded. Nevertheless, “designers and viewers do not make contact with each other at any point, they are both monologues” (O’Toole, 1994, p. 3). It is maybe the case that “designers are just responsible for encoding while viewers are just responsible for decoding”. Besides, designers and viewers by all means have no way to communicate” (Cheng & Zhang, 2016, p. 7). Thus, what remains unclear is whether decoders’ interpretations are in accordance with encoders’ designing intentions or not.Finally, to date, scholars doing multimodal discourse analysis have mainly been limited to systemic functional linguists (Zhu, 2007, p. 85).

In order to address the above problems, this study was conducted to explore the gaps between decoders’ interpretations and encoders’ designing intentions. To be specific,is aimed at unveiling the similarities and differences in the interpretation of the same multimodal discourses made by encoders and decoders. Furthermore, possible underlying causes of both the similarities and differences will be discussed.

2. Research Design

Based on the research background, the present study was conducted primarily to find out the differences and similarities in the interpretation of the same multimodal products by contemporary primary decoders (i.e., linguistic scholars) with respect to where multimodal discourse analysis and art graphic designers are on earth.

First, the research question is posed: “what are the similarities and differences in interpretations of the same multimodal discourses between linguistic and art graphic design participants?” Second, considering the research objective, two kinds of participants,sixty in total—thirty linguistic participants and thirty art graphic ones—were chosen. The linguistic participants are all linguistic teachers from five universities across different regions in China. The participants selected this way are typical and representative to a large extent. In addition, all of them have had experience in doing multimodal discourse studies. Art graphic participants either have part-time job experience in advertising companies or are actual designers. The reason why art graphic participants were chosen this way is that their designing proficiency is equivalent or close to that of the designers designing the research data of the present study. Third, the instrument employed in the present study is an online questionnaire with thirty multiple choice questions, and short answer questions to be answered by art graphic participants only. The name of the questionnaire on sojump (http://www.sojump.com) is “Questionnaire regarding both the Similarities and the Differences in Comprehending Multimodality”. With respect to research data, they are thirty public multimodal advertisements (hereafter abbreviated as Ads) of investment and financial management (hereafter abbreviated as IFM). In addition,all data are characterized by the co-deployment of major image with big size, major hue with intense saturation and high differentiation, logo and explanatory verbal text (see Figure 1). Thus all the data are typical visual-verbal multimodal discourses in nature. In addition, the reason of choosing multimodal Ads of IFM as research data is that they are brief and striking in form, strong and appealing in effect, with harmonious and bright colors, and unique spatial arrangement.

中国神华煤制油化工有限公司工业控制系统有DCS，SCADA，PLC，其中DCS采用Honey well的E-PKS系统；SCADA采用爱克讯公司的SCADAPack RTU系统，与DCS有统一的操作界面；PLC则有Siemens，AB，GE等公司的系统。

3. Results of the Online Questionnaire

Major results regarding the online questionnaire will be depicted from the following four perspectives: general interpretations between participants, general interpretations within participants, interpretations in terms of color, and interpretations from the perspective of spatial arrangements.

图5中可以看出，当摩擦系数较小时，相对压强随泊松比的增大的增幅很小；当泊松比较小时相对压强随摩擦系数增大在较小范围内增大；当泊松比和摩擦系数增大到一定程度后，随着泊松比和摩擦系数的增大，相对压强急剧增大。表明物料特性能显著影响挤压制粒成型所需的挤出成型力，进而影响制粒能耗和设备的使用寿命，根据制粒原料的特性对制粒过程工艺参数和设备参数进行优化，以达到良好的制粒效果，同时有效降低制粒能耗和延长设备使用寿命。

3.1 General interpretations between participants

It is commonly acknowledged that multimodal literacy is indispensable for virtually anyone to understand both the inner and outside world. Yet the current state of multimodal literacy is far from satisfactory. Taking the interpretations by linguistic scholars for example, different linguistic scholars can give different or at least slightly different interpretations even for the same multimodal discourse.

This section is mainly concerned with the general interpretations, including both the differences and similarities between linguistic participants and art graphic ones. The descriptive statistics demonstrate that there are indeed remarkable differences in terms of both general interpretations and those only from the perspective of color. Furthermore,the interpretations of the spatial arrangements that present large similarities while certain differences are demonstrated as well. Table 1 demonstrates the general interpretations of the research data.

2012年3月6—9日在美国内华达州拉斯维加斯会展中心举行的2012年美国国际水处理展览会上，GE展示了2011年推出的ZCore深层滤芯过滤技术和其广泛的产品组合。这些产品组合可通过水处理设备、化学品药剂和配套技术服务帮助商业和工业领域的客户应对严峻的水资源挑战。

It can be seen from Table 1 that the complete unanimity in terms of both color and spatial arrangement among the interpretations between these two kinds of participants only registers thirteen research data points out of thirty in total. The overall unanimity rate just reaches 43.3%. In other words, among the thirteen research data points, all the sixty participants are consistent with each other at least in terms of the best major hues (namely color), the worst major hues as well as the best and the worst synergetic patterns. These descriptive statistics just confirm that there are large differences in the interpretations between linguistic and art graphic participants. However, similarities in the interpretations between participants do exist as well.

The unanimity in the interpretations from the perspective of color alone involves fourteen research materials of thirty in total, among which one research material with unanimity in terms of color but not in spatial arrangements is also included.The unanimity rate only reaches 46.7%. In this perspective, it can be inferred that interpretations by linguistic scholars have largely deviated from the meanings in the mode of color presupposed by the image designers. Or rather exactly, it can be seen that linguistic scholars do not decode or grasp all the meanings entailed in the colors used in these research data. What is more, the number of research materials that these two kinds of participants reach an unanimous agreement in terms of spatial arrangements but not in color is 8, which accounts for 26.70%. It is the same with the research materials with unanimity neither in terms of spatial arrangements nor in color. It can be calculated from the above data that the interpretations from the perspective of spatial arrangements present big unanimity in twenty-one research materials amid the thirty in total. The unanimity rate is up to 70%.

3.2 General interpretations within participants

As presented above, there are indeed similarities and differences in the interpretations of the same multimodal discourses between participants. According to the author’s analysis,it is the same with within-participants.

General interpretations from the perspective of both the most and least suitable colors in terms of representing themes of the multimodal ads of IFM have been profiled in Figures 2 and 3. It can be seen from the wavy tendency of these two Figures that differences do exist in both linguistic and art graphic groups. From the perspective of gaps between peaks and troughs of waves, however, differences within the group of linguistic participants are much bigger than those within the group of art graphic ones. In addition,although there are fluctuations with respect to the most suitable colors among art graphic participants, some regular tendencies can be found as well. For instance, golden yellow and blue are commonly acknowledged as the top two suitable colors. Nevertheless, no regular features at all can be concluded from the group of linguistic participants.

Contrary to the results exhibited from the perspective of the most suitable colors,interpretations of the least suitable colors by the linguistic participants present regular tendencies. It is commonly acknowledged by the majority of linguistic participants(63.33%, i.e., nineteen participants of thirty in total) that green is the least suitable one for embodying the themes of investment and financial management. By contrast, no regular features are exhibited among the interpretations by the group of art graphic design participants.

With respect to the similarities in the interpretations of spatial arrangement, there are big similarities whether from the perspective of between-participants or withinparticipants. These similarities are reflected in Table 2.

It can be concluded from Table 2 that center is the ideal location for major images in the multimodal genre of IFM, with upper left being the most suitable place for logos and lower parts of the framed area are ideal for explanatory verbal texts.

To sum up, color-collocation, color’s symbolic and schematized meanings, color preference order, and so on, are all universal. Being universal means being popular,general and historically successive, which are just some aspects of cultures. In other words, being universal means being accessible to public. Therefore, it is not surprising that similarities exist among participants.

3.3 Interpretations in terms of color

On the basis of the above color preference order and for the sake of facilitating the interpretation of general multimodal discourses of IFM, the present study intends to regulate the color preference pattern by assigning concrete values for the colors along the color preference order. Figure 6 demonstrates the values assigned for the mode of color.

In spite of this, in order to reveal an overall interpretation of all the research data,it is necessary to find out what the concrete interpretation of each research material is.Insomuch as these multimodal products are to express the same theme, i.e., investment and financial management, is there a general order of preference in terms of the major color? If so, in what order does the preference proceed? Are certain colors actually and usually preferred over others? If there is really an order of color preference, then is it possible to assign each concrete color a concrete value?

完全加权关联规则(I1→I2)提升度(all-weighted Association Rule Lift,awARL)的计算如式(10)所示.

开标必须得有招标人进行主持，然后邀请全部的投标人来参与。除了开标工作人员、投标人、监督人员和公证人员以外，其他不相关的人员应该予以回避。如果进行某一标段项目进行划分且对其唱标时，没有参与投唱标段的人员必须得回避。

The frozen serum samples were brought to room temperature before analysis.Serum concentrations of IL-6,OPG,and RANKL were measured using ELISA,in accordance with the manufacturer's instructions.

According to the above summarizations, the following conclusions can be reached.Firstly, linguistic scholars have not been able to decode or grasp all the meanings contained in the images presupposed by the designers. Secondly, big differences in the interpretations between linguistic participants and art graphic do exist. Thirdly, it is justified that researchers from other research fields are generally not familiar with the repertoire of designing knowledge, or lacking rich designing practices as well as systematic formal schooling education regarding art graphic design equipped by art graphic practitioners. In view of these facts, it is tentatively determined that the present study mainly employs the results regarding the mode of color summarized from the interpretations performed by art graphic participants. Meanwhile, opinions of the least suitable colors from linguistic scholars are taken into account as valuable references. In addition, these six colors, namely red, yellow, green, blue, black and white are commonly acknowledged as the six fundamental psychological visual colors within people’s feelings. Therefore, the present study chiefly takes the above six fundamental colors into consideration.

According to the above illustration, the following two points can be summarized.First, the between-participant unanimity rates are very low, i.e., 43.3% from the overall point of view and 46.7% from the perspective of color only. Second, the interpretations of the modes of color made by art graphic participants present regular features while no regular features have been found among the interpretations conducted by linguistic participants.

In view of the fact that this is only a pilot study along the road of the comparative empirical study which is meant to make encoders and decoders have a direct talk, and that no more previous studies of this kind have been available for reference, only the intensely saturated colors are included in the present study. Major hues employed in this study are shown in Figure 4.

In addition, in order to relieve visual burdens on the part of participants, colors with low differentiation are excluded from the present study. For instance, in Figure 5, image 1 is more differentiated than image 2, so image 2 is excluded. It is the same with image 3 and 4. Because purple is a cold-tuned color, the colors in image 3 are more appropriate than purple.

With regard to the similarities, it is commonly acknowledged by the participants that the most suitable colors to embody the theme of IFM are golden yellow or yellow, red and blue in turn. Of the thirty research materials in total, participants voted golden yellow or yellow, red and blue as the most suitable color for ten, ten and eight research materials respectively, which account for 93.33% of the total research materials (see Table 3). On the contrary, virtually no obvious regular features have been found, especially in terms of the most suitable colors for the theme of multimodal products of IFM among linguistic participants. This can prove that different scholars may have different opinions even for the same multimodal product.

Nevertheless, there are remarkable differences in the interpreting tendencies with respect to the least suitable colors between these two groups of participants. To be specific,interpretations of the least suitable colors for the multimodal genre of IFM within the group of linguistic participants present regular tendencies. It is commonly acknowledged by the majority of linguistic participants (63.33%, i.e., nineteen participants of thirty in total) that green is the least suitable one for embodying themes of investment and financial management. By contrast, no regular features with regard to the least suitable colors are exhibited among the interpretations by the group of art graphic design participants.

Based on the analysis and summarizations from the interpretations by art graphic participants and by reference to those of linguistic ones, it is found that there is indeed a general order of color preference. Moreover, during the process of determining the color preference, the present study refers to one of the three commonly-adopted methods listed in Kouwer (1949, p. 33) in conducting the studies on color preferences. To be specific, the present study chiefly adopted the method that “judging colors separately in accordance with a given scale of values” (ibid.).

The order of color preference in the multimodal genre of IFM is generalized as follows: starting from the most preferred color, the order of color preference ranges from yellow or exactly golden yellow, through blue and red, to the equally unpreferred colors(black and white), to the most unpleasant color (i.e., green). To summarize, the order of color preference in the multimodal genre of IFM is: green → black/white → red → blue→ yellow/golden yellow.

创新律师顾问服务形式。泰安市人大常委会共组建1个律师顾问团和8个律师顾问小组。律师顾问团由5名律师组成，为市人代会会议期间代表履职提供驻会法律服务。组建的8个律师顾问小组分别为全国人大代表、省人大代表和6个县（市、区）人大代表履职服务律师顾问小组，每个团组设两名联系人员，分别由市县两级人大常委会和司法行政机关工作人员担任。人大代表需要法律服务时，可以从公布的律师顾问小组中自由选取律师，也可以通过联系人随机选取律师。常委会明确了律师顾问服务的性质和服务期限，服务属于公益性活动，不收取任何费用；服务期限与人大代表任职期限相同。

One point worth noting is that the degree of preference increases as the right arrow moves forward. Actually, the order of color preferences forms a continuum and green,black and white, red, blue and yellow are five extremes along this continuum. In addition,the intermediate colors are considered more preferred than the left adjoining primary colors and naturally less preferred than the right adjoining primary color, for instance, a blackish green is more preferred than green but less preferred than black.

The general interpretations, including both the differences and similarities made by linguistic participants and art graphic ones have been profiled in the preceding section. As noted, Table 1 and the descriptive statistics in the preceding section have revealed how many pieces of research materials present the complete unanimity in terms of the overall interpretations, and how many pieces only present unanimity either in terms of color or from the perspective of spatial arrangements.

Major results with regard to the mode of color can provide answers for the first research question. The following three important points can be concluded from the above mentioned Figures: First, Figures 2 and 3 can confirm that there are really gaps between the interpretations conducted by the practitioners of art graphic design and linguistic scholars. Second, it is not difficult to find from Figures 1 and 2 that there are obvious regular features among the interpretations by art graphic participants while on the contrary, different linguistic participants may have different opinions even for the same multimodal products, or exactly they are ill-regulated to some extent. Finally, the fact can be verified that practitioners will inevitably, unconsciously or consciously take their designing practice or principles into consideration when interpreting multimodal products.Furthermore, their interpretations will produce some valuable implications for scholars from other research fields.

3.4 Interpretations in terms of spatial arrangements

Major results with regard to the mode of color have been demonstrated in the preceding section. These results cover both the similarities and differences within participants as well as between participants. Besides, the most suitable colors for the theme of multimodal genre of IMF have been summarized as well. This section is devoted to presenting the pairwise comparisons between the best synergetic patterns and the worst ones among all the redesigned multimodal discourses for each of the thirty research materials one by one on condition that the major hues keep unchanged.

Inasmuch as there is indeed a general order of color preference in this financial genre, some colors are really preferred to others and the major colors can be granted with concrete values according to the color preference order, then the following question will arise naturally: is it the same with the spatial arrangements? To be specific: 1) What are the overviews in terms of the best and the worst synergetic patterns of all the thirty research materials regarded by the majority of the subjects? 2) Are there general orders of preference in terms of spatial locations for major images, logos and the explanatory verbal texts? 3) Actual co-occurring patterns among all the modes can be of great variety,yet is there a best synergetic pattern for the major modes to work together? If so, what is it? These questions will be addressed in the subsequent part.

3.4.1 Similarities and differences in terms of spatial arrangements of images and verbal texts

另一个重要的技术难题是：长江属于多泥沙河流，虽然含沙量不算大，但输沙量很大。在汛期利用来水来沙多的特点，需要尽量多地排走上游来的泥沙或者冲走淤积在库底的泥沙，水库采用“蓄清排浑”的运行方式是保证大型水库长期运行的重要措施之一。所以，大部分控制性水库都需要在汛后蓄水。

According to the results outlined in Table 5.3 and statistical analysis of the other nine research materials with different evaluations by linguistic and art graphic participants,the holistic opinions of the spatial arrangements of major images and explanatory verbal texts are as follows: 56.67% linguistic participants (seventeen in total) and 53.33% of art graphic ones (sixteen in total) view the patterns where the major image is placed on the left and the explanatory verbal text is placed on the right as a more ideal visual-verbal arrangement. The pattern where the explanatory verbal text is placed on the upper part and the major image placed on the lower framed area (see Figure 7) is regarded as more ideal by both linguistic and art graphic design participants as well.

Although there is high unanimity of spatial arrangements of images and verbal texts between these two groups of participants, the number of linguistic participants with this opinion is slightly larger than that of art graphic ones. It is likely that scholars from other research fields are mainly internalized with traditional visual-verbal relations while scholars in the field of art graphic design will take a lot of unconventional factors into consideration, like the insights about effects of both conventional and unconventional designing to be achieved by format design (e.g., Liu & Li, 2016; Tao et al., 2015; Chen,2009), special requirements for art design practitioners within omnimedia environment(Leng, 2016, pp. 105-106). In other words, it is essential that art graphic practitioners not only possess knowledge like formal beauty in esthetics, typology, color, rhythm, and so on, but also take other factors such as consumer psychology (e.g., Solomon et al., 2012),marketing strategy, etc. into account in the context of the omnimedia.

Secondly, the objective of multimodal visual grammar is in agreement with that of format design and visual flow. Scholars doing multimodal discourse analysis are well equipped with knowledge of visual grammar (Kress & van Leeuwen, 1996/2006).According to Machin (2007, pp. ix-x), what multimodality primarily concerns is just the rules and principles adopted by image designers. Exactly speaking, multimodality involves the analysis of fundamental principles utilized by designers to arrange visual elements. Thus, viewers can be able to grasp meaning potentials transmitted by the arrangements of salience, color saturations, framing and so on. Meanwhile, the basic requirement of format design (Liu & Li, 2016, p. 62) is first to determine the habitual reading modes on the part of target readers, and then to arrange the co-occurring patterns among visual elements like images, verbal texts, colors and so on according to target readers’ habitual reading modes in a reasonable, orderly and vivid way.

After analyzing the results demonstrated in Figures 2, 3 and Table 2 thoroughly, I interviewed five art graphic subjects interviewed with the results in Figures 2, 3 and Table 2 presented to them. This in-depth interview was employed to explore what co-occurring patterns formed by the core modes are the most and the least ideal in a concrete, thorough and penetrating way. After that, both the most and the least ideal synergetic patterns have been tentatively concluded.

Specifically, when the major hue of multimodal products or major hue of major images is golden yellow, the major image happens to be placed in the center, and the explanatory verbal text happens to be placed on the lower left part or the lower right section, these patterns are called the best or the most ideal synergetic patterns. In addition,when the four major modes, namely the logo, the major image, the major hue, and the explanatory verbal text work together to form this ideal pattern, then the specific multimodal discourse with this pattern is the most informative. Figure 8 portrays an overview of the best synergetic pattern(s).

In addition, the results presented in Table 2 reveal that there are general orders of preference in terms of the spatial locations for major images, logos and the explanatory verbal texts. At the same time, it serves as an overview of the agreement of evaluations on the best spatial arrangements of major pictures, logos and explanatory texts by linguistic and art graphic design participants. In addition, there are big similarities between these above two kinds of participants in terms of spatial arrangements. To be specific, it is commonly acknowledged that the center or middle lower part of the framed area is the most ideal location for the major pictures. With respect to the logo, the participants commonly perceive that the upper left position is the most ideal for it in that this position is the starting point for reading multimodal products. Moreover, the left and right inferior parts are much more ideal locations for explanatory texts (Cheng & Zhang,2016, p. 6).

In addition, these two kinds of participants are highly consistent with each other with respect to the worst synergetic pattern in terms of representing meanings. According to statistics, the worst synergetic pattern formed by the core modes is this situation in which the major picture is placed in the upper right part, with the logo lower left and the explanatory texts in the upper part. In addition, it is found that there are no obvious regular patterns among the ill-arranged research materials as in the well-arranged ones from the perspective of spatial arrangements. Therefore, the results as regards the agreement evaluations on the worst spatial arrangements of major pictures, logos and explanatory texts by linguistic and art graphic design participants will not be illustrated in detail.

4. Similarities

As noted, results with regard to similarities and differences in the interpretations have already been reported in the preceding sections. So, what are the underlying causes for these similarities and differences? This section is devoted to discussing the major results by referring to theories and studies mentioned previously.

与其他星座一样，人马座这片黑暗天区中同样散布着点点繁星，只是它们普遍没有那么亮，显得灰蒙蒙的。在那里，一颗名为S2（有时也称S-02）的小恒星正在永恒的边缘巡航。每16年，S2就会穿越一个神秘黑暗天体的势力范围，后者大概有400万个太阳那么重，正正好好占据着银河系的中心位置。

2.3.5 重复性试验 取“2.1”项下米索硝唑pH敏感脂质体适量，共6份，按“2.2.4”项下方法制备经预处理的米索硝唑pH敏感脂质体溶液。以甲醇为空白，于322 nm波长处测定吸光度并计算主成分的含量。结果，米索硝唑含量平均值为99.20%，RSD为1.55%（n＝6），表明本方法重复性良好。

According to the analysis of the major results, the complete unanimity rate of the overall interpretations is 43.3% (i.e., thirteen research materials). In other words, as far as these thirteen research materials are concerned, all the sixty participants are consistent with each other at least in terms of the best major hues (namely color), the worst major hues, and the best and the worst spatial arrangements. The unanimity of the interpretations from the perspective of color alone involve fourteen research materials of thirty and the unanimity rate reaches 46.7%. The unanimity rate with regard to the spatial arrangements among core modes amounts to 70%. From this standpoint, it can be concluded that certain similarities do exist in the interpretations of the same multimodal discourse between linguistic scholars and art graphic designers.

In this section, major results will be further discussed in the hope of exploring what might lie behind these above results from the perspective of cultural traditions,practical experience and professional knowledge reserve possessed by these two kinds of participants. In addition, the discussion will be specially conducted to address the feasibility of the multimodal visual-verbal discourse analytical framework constructed in the present study. Meanwhile, potential applications of these results will be explored as well.

4.1 Cultural tradition

These similarities in the interpretations made by “decoders” (i.e., linguistic participants)and “encoders” (i.e., art graphic participants) can be provided with evidences from longterm cultural traditions. First, the employment of color during the process of art graphic design is not really random and intuitive. Instead, it is relatively fixed experience during long-term practice. As already noted in numerous previous studies (e.g., Ge et al., 2005;Saito, 1996), certain regular features with respect to color characteristics, employment and color preference have been existing although some changes are bound to happen with the evolution of times. This very experience can in turn be transmitted from generation to generation. In addition, the relations among different colors are not at random, either.Instead, they are the consequences of color-collocating experience of the process of longterm practice.

目前，高邮湖管理已经形成了从省水利厅到高邮湖流域性管理单位到市、县水行政主管部门到各基层湖泊管理单位的四级管理体制。要深入贯彻落实高邮湖管理与保护联席会议制度，充分利用联席会议的平台，切实加强与环保、渔业等其他行业主管部门的协调沟通，形成管理合力，积极探索综合管理的新模式。

Second, the Chinese traditional “five color system” (i.e., cyan, yellow, red, white and black) is a valuable cultural heritage as suggested by the name. The traditional “five color system” is used to summarize the relationships between human beings, between human beings and society, and between human beings and nature. These five colors are used to generalize all the complicated colors in the natural world. Gradually, the complicated colors are gradually regularized, symbolized and schematized, and more gradually folk color views are formed deep-rooted and transmitted from generation to generation(Zhao, 2012, pp. 229-230). In addition, it is found that although the formation of color’s symbolic meanings is subjected to certain social and cultural backgrounds, the application of color’s symbolic meanings is not specific to certain people or certain cultures, rather, it is universal (ibid.).

Third, as mentioned by Kouwer (1949, p. 34), Cohn (1894) was the first scholar to conduct the study on color preferences. He then conducted the study again by increasing the number of subjects from the original 14 to about 500 or even 1000. The results were summarized by Eysenck (noted from Kouwer, p. 34). The results of Cohn’s study reveal that there is indeed a certain universal order of color preference according to the affective value of the colors. Being universal means being popular among people, whether professional designers or common people.

Based on the above analysis, it is justified that there are similarities in the interpretations of colors among participants because of tradition, transmission and universal color preference.

4.2 Similarities in knowledge structure

As already elaborated, any multimodal discourse will involve specific sign-makings against semiotic potentials. However, signs are by no means pre-existing combinations of signifiers and signifieds in social semiotics, nor are they always “available” or“ready-made” to be selected and employed. Instead, sign-makings, to a larger extent,are consequences of motivation relating to sign-makers and a certain specific context(Kress & van Leeuwen, 2006, p. 8). Specific sign-makers will choose the most apt and plausible signs available to them to achieve their specific purposes under a specific context. Therefore, from this standpoint of view, signs are “functional units”. Thus,selected signs can be presented in many possible material ways so long as they can achieve the representational function (Larsen, 1994, p. 3824). Meanwhile, there is no contact at all between sign-makers and viewers (O’ Toole, 1994, p. 3; Cheng & Zhang,2016, pp. 4-5). Based on the above analysis, it is justified that there are differences between linguistic scholars and art graphic designers even in interpreting the same multimodal discourses.

In the first place, meanings conveyed by visual elements placed in the visual center are similar to the field of discourse in contextual configuration. Visual center is a common term utilized in the field of art graphic design while context is what linguistic scholars are familiar with. It is found in many previous studies (e.g., Tatler, 2007;Roberts-Breslin, 2003) that the major information of top hierarchy is bound to be placed in the visual center. Field of discourse (Halliday & Hasan, 1985/1989, p. 12) concerns what is going on in a specific text. In addition, scholars doing multimodal discourse analysis are well informed of the compositional parameter “salient” (Kress & van Leeuwen, 2006, p. 202). They commonly determine what the field of discourse concerns by means of salience of modes. In this respect, it is natural that certain similarities in the interpretation of primary meanings transmitted by certain multimodal discourse exist between them.

3.4.2 Opinions of the best and the worst synergetic patterns

Finally, the similarities in the interpretation lie in art graphic designers’ designing intentions. According to Halliday and Hasan (1985/1989), a text is an outcome of both process and product. During the designing process, what designers are primarily concerned with is to transmit information to target viewers in an accurate and smooth way(Liu & Li, 2016, p. 62) and along a definite channel (Watson & Hill, 1997, p. 139). As a consequence, designers will efficiently address their viewers in easily “recognizable and socially appropriate way” (Halliday & Hasan, 1985, p. 45). For this purpose of design,designers are supposed to choose the “most apt and plausible forms” in a specific context so as to make their information understandable (Zhang, 2013, p. 42). In addition, context of creation will chiefly relate to the “author’s conception of the viewers”. Furthermore,his or her perception as well as attitudes towards target viewers will in turn have much to do with how these perceptions and attitudes are built into the text. From this standpoint,designers attach great importance to the context in which multimodal discourses are created and to the appropriate usage of signs. However, context and signs are what systemic-functional linguists are well equipped with. In other words, when decoding the “product”, i.e., text, scholars doing multimodal discourse analysis will focus on how designers construct meaning systems from the perspective of social semiotics (Halliday,1994).

“秦月”由“秦富1号”和“嘎拉”杂交育成。果实呈圆形或长圆锥形，大小均匀，果个与嘎拉接近，果皮鲜红，果肉黄白，肉质细脆多汁，酸甜适口，较耐贮存。在渭北高原南部地区8月底成熟，渭北高原中部地区9月上中旬成熟，刚好赶上中秋节和国庆节，是一个优良的中熟苹果新品种。

5. Differences

Apart from the similarities in the interpretations of the same multimodal discourse exhibited by both linguistic and art graphic participants, what is notable is the differences between them. This section is dedicated to discussing the results of interpreting differences from the perspective of the sign-making process in social semiotics, properties of multimodal discourse, and professional knowledge reserves possessed by these two groups of participants. Besides, mismatching between participants is also viewed as a possible reason for the interpreting differences.

在12日上午举行的座谈会上，部分住青岛的全国和省人大代表结合自身履职实际作了发言，围绕海洋强省建设、工业互联网资源整合、乡村振兴、基础教育发展等热点问题提出了意见和建议。于晓明对代表们忠实履行职责、充分发挥代表作用给予了高度肯定。他鼓励代表就自己关心的问题形成调研报告，积极提出意见建议，交有关部门研究办理。

5.1 Sign-making process in social semiotics

As has been elaborated, as far as the primary meanings conveyed are concerned,interpretations by both scholars doing multimodal discourse analysis and art graphic designers demonstrated large similarities. In addition, linguistic scholars’ opinions of spatial arrangements are in high agreement with those of art graphic practitioners. This can be attributed to similarities in knowledge structure between these two kinds of participants.

It can be found that the most apt and plausible signs are the consequences of an analogous process, which is actually a process of classification. Analogy or classification, in turn, is conventional (Kress & van Leeuwen, 2006, p. 8). So it is highly proposed by Kress and van Leeuwen (ibid., p. 11) that sign-makings are always motivated and conventional. Or rather they form a continuum between motivation and convention.

To be specific, many sign-makings are conducted in a mediated way between convention and specific motivation. In other words, the selecting signs are, by and large,somewhere in between these two extremes, or exactly in correspondence with a certain point along the convention-motivation continuum. In this respect, the conventionmotivation continuum can justify both similarities and differences in the interpretations made by linguistic and art graphic participants.

5.2 Properties of multimodal discourses of IFM

As far as the remarkable interpreting differences among linguistic scholars as well as between the linguistic scholars and art graphic practitioners are concerned, these differences can be attributed to the properties of multimodal discourses of IFM. First,the design of print advertisements is a strongly comprehensive mental process in that a very wide span that falls between the intuitive visual images and the meanings to be expressed by means of inner logics (Cheng & Zhang, 2016, p. 5). Thus, the originalities of print advertisements are polysemous, hierarchical and floating. Second, there is lack of communication, or rather exactly, no contact at all between art graphic designers and scholars. The lack of communication is supposed to shoulder responsibilities for the interpreting differences. Just as O’Toole (1994, p. 3) put it, the viewers (decoders)and the designers (encoders) of arts “do not make contact with each other at any point.They are monologues, i.e., no intersection with each other, and they fail to help their speakers relate to each other”, which are our long-term and habitual cultural traditions.This was detected by Cheng and Zhang (2016, pp. 4-5) as well. They note that “designers and decoders have never interacted and communicated about designing concepts before” and that “decoders in the linguistic field have not virtually communicated with encoders majoring in art graphic design. This contributes to the situation where designers are just responsible for designing while decoders interpret whatever they like individually” (ibid.). Third, O’Toole (1994) also ascribed this to the lack of standard languages for describing arts. When we are confronted with multimodal products,which are ubiquitous in people’s practical lives, people are accustomed to adopting their individual decoding ways and methods. As a matter of fact, they have no choice but to conduct decoding individually in that there is no handy standard to adopt. These meanings to be expressed here can be confirmed by O’Toole (1994, p. 4). It is stated in his study that different viewers with certain pieces of art are privately involved; how to interpret a piece of art when encountered comes completely out of individual viewer’s vision, knowledge reserve, and mood. Viewers are always making their associations.In a sense, it is their eyes that do the analysis indeed. In addition, viewers often find that the transformation of arts into words has little to with the perceptual reality and intentions of designers.

One important point to note is that there are likely to be conflicts between the drives to present designers’ creativity and those to reach an audience under a market-oriented economy. It is generally contradictory for designers to design multimodal ads under a market-oriented economy. In doing so, designers have motivations to design ads which will realize their desire for self-expression on the one hand, while on the other hand,designers will be manipulated by marketplace as well where the value of a certain multimodal ads is primarily judged by economic criteria. As a consequence, it is highly possible that the motivation to present one’s creativity and the one to reach an audience are sometimes compatible, and sometimes not (Roberts-Breslin, 2003, p. 8).

In addition, although there are remarkable differences in the decoding conducted by these two kinds of participants, still quite large proportions of interpretations by them present similarities to a large extent. However, similarities in the interpretations from the point of view of spatial synergetic pattern are much larger compared with those demonstrated both by the overall interpretations and colors. These phenomena can also be justified from the perspective of cultural traditions. Even in ancient times, people can familiarly have the reading capacity of spatial orientations. Furthermore, the culture about Yin and Yang, the eight diagrams as well as the five elements is deeply rooted in people’s minds and transmitted from generation to generation. Plus, spatial orientations are relatively familiar and easier to understand than colors. This is because in designing,lots of complicated factors apart from the six kinds of basic colors need to be taken into consideration, such as the opposition between cold and warm colors, the contrast between light and dark colors, between purity and clarity, the harmony among different hues,saturations, color brightness, color gradients, etc. These are all professional knowledge possessed by art graphic practitioners and are out of reach of scholars in other research fields. From this perspective, it can be concluded that this professional knowledge is out of the reach of cultural traditions and transmissions. Therefore, it is not surprising that the similarities concerning the interpretations from the point of view of spatial synergetic pattern are much larger compared with those demonstrated both by the overall interpretations and colors.

5.3 Professional knowledge reserves

As already elaborated, some common sense like color-collocation, colors’ symbolic and schematized meanings, and color preference orders, are universal and can be transmitted from generation to generation. However, knowledge with respect to concrete designing practice are far more complicated than common sense and tradition. Taking color as an example, lots of factors relating to it are by no means simple to grasp, such as the opposition between cold and warm colors, the contrast between light and dark colors,the harmony between hues, saturations, color brightness, color gradients and so on.It is obvious that they are by no means a matter of perception out of common sense and cultural tradition. Instead, they are acquired and accumulated in long-term formal school education and designing practice. Obviously the knowledge belongs to part of professional knowledge reserves possessed by art graphic practitioners. Therefore, in this sense, it is not linguistic knowledge alone that can settle the issues of multimodal decoding.

5.4 Mismatching between subjects

The differences in the interpretations between the art graphic design participants and linguistic ones can be attributed to mismatching between subjects. It is conceivable and logical that designers of any advertisement will take the potential viewers’ comprehending capacities into account when designing. This can be proven by Hasan (1996, p. 52) that creation will be related to the “author’s conception of the audience”, which relates to the designers’ perception and attitudes towards the readership and how they are built into the text (Royce, 1999, p. 131). Additionally, it is highly possible that designers will by no means expect that their designing will be evaluated, interpreted or even studied by linguistic scholars for academic purposes one day when they conducted the act of designing.

The present study seeks to endow between-subject mismatching with the following definition by referring to the opinions elaborated by Chen (1999, pp. 31-32). Mismatching between subjects refers to the fact that the designing subjects (i.e., encoders or designers) will be likely to exclude certain kinds of groups as their potential readers when they conduct the action of selecting certain modes in the process of designing.As a consequence, these certain kinds of groups will be naturally excluded from the communication. Obviously, the contexts generated by a group of subjects with similar cultural and knowledge backgrounds will be relatively independent of those yielded by another group of subjects with different cultural backgrounds and knowledge structure as well as reserves. Furthermore, it is definite that there will be differences in terms of terms, knowledge structures as well as logics between these above two kinds of groups as they have never interacted and communicated with each other before. What is more, to be certain, the involvement by other groups instead of the designing subjects will definitely impose their individual moods, attitudes, and so on upon comprehending the original speech actions (O’Toole, 1994, p. 4). In view of the above analysis, it is justified that there are differences between different subjects.

6. Summary

This paper is dedicated to reporting the results with regards to both the similarities and the differences in the interpretations between groups of linguistic participants and art graphic design ones. Meanwhile, the results of within-participants have been demonstrated as well. Major results are generalized with regards to preliminary study, online questionnaire and retrospective interview, respectively.

The statistical analysis of the preliminary study has shown that there are indeed significant differences between linguistic scholars and art graphic designers in terms of both interpreting foundations and interpreting sequence. To be specific, when decoding the main meanings conveyed by multimodal discourses and tapping potential attitudes of image designers as well as investors, they are grounded on the following four explicit criteria: the elements placed in the visual center or the best visual region, visual process,the affective characters of the main colors, as well as the image-verbal relations. In addition, it is found that their reading path is fixed and clear-cut. The interpreting sequence adopted by art graphic practitioners is from images to logo, then to explanatory verbal texts, and ultimately to images again. By contrast, the interpreting sequence adopted by linguistic scholars is from logo first to explanatory verbal texts, and then to major images.

The analysis of the online questionnaire demonstrates:

● With respect to the general interpretations of the research data, there are similarities and differences both between and within participants.

● The unanimity rate of color interpretation is 46.60% whilst the divergence rate of color interpretation between participants is up to 53.40%.

● There are high similarities (i.e., 70%) in the interpretation of spatial arrangements,though there are slight variances in terms of visual-verbal arrangements. In addition, two groups of participants are in agreement with each other about the most ideal synergetic pattern formed by core modes involved in the multimodal genre of IMF. To be specific, the best synergetic patterns refer to situations where the major hue is golden yellow, the major image happens to be placed in the center, and the explanatory verbal text happens to be placed on the lower left part or the lower right section are generalized.

Both the quantitative and qualitative analyses of the retrospective interview show that there are slight similarities and remarkable differences between participants.More specifically, besides primary meanings conveyed, there are differences in the interpretations from the perspective of the following dimensions: interpreting foundation,reading path, terms employed, reading “attract”, synergetic effects formed by core modes,and so on.

References

Chen, J. (1999). On the four distinctive features of context. Shandong Foreign Languages Journal,(4), 30-33.

Chen, Z. N. (2009). Dynamic typography design in video media. Packaging Engineering, (6), 99-102.

Cheng, R. L. (2017). An interdisciplinary study on the synergetic patterns formed by major modes in multimodal discourse of IFM (Doctoral dissertation). Shandong University, Jinan.

Cheng, R. L., & Zhang, D. L. (2016). An interdisciplinary study on the synergetic patterns of major modes in multimodal text. Technology Enhanced Foreign Language Education, (5), 3-8.

Cohn, J. (1894). Experimentelle Untersuchungen über die Gefühlsbetonung der Farben,Helligkeiten und ihrer Combinationen. Philosophy Study, (10), 562-603.

Ge, L. Z. et. al. (2005). A research of color preference. Psychological Science, (4), 849-851.

Halliday, M. A. K. (1994). An introduction to functional grammar. London: Arnold.

Halliday, M. A. K., & Hasan, R. (1985/1989). Language, context, text: Aspects of language in a social-semiotic perspective. Oxford: Oxford University Press.

Hasan, R. (1996). Ways of saying, ways of meaning: Selected papers of Ruqaiya Hasan (C. Cloran,D. Butt, & G. Williams, Eds.). London: Cassell.

Kouwer, B. J. (1949). Colors and their character: A psychological study. Springer: Berlin.

Kress, G., & van Leeuwen, T. (1996/2006). Reading images: The grammar of visual design.London: Routledge.

Larsen, S. E. (1994). Semiotics. In R. E. Asher (Eds.), The encyclopedia of language and linguistics (pp. 3821-3832). Oxford: Pergamon Press.

Leng, J. Q. (2016). The development of new media art design in the field of communication. News and Writing, (1), 104-106.

Liu, C. X., & Li, P. (2016). Pedagogical research and practice of interactive format cases in format design. Higher Education Exploration, (S1), 62-63.

Machin, D. (2007). Introduction to multimodal analysis. London: Hodder.

O’ Toole, M. (1994). The language of displayed art. London: Leicester University Press.

Roberts-Breslin, J. (2003). Making media: Foundations of sound and image production.Massachusetts: Focal Press.

Royce, T. D. (1999). Visual-verbal inter-semiotic complementarity in the economist magazine(Unpublished doctoral dissertation). University of Reading Center for Applied Language Studies, Reading.

Saito, M. (1996). Comparative studies on color preference in Japan and other Asian regions, with special emphasis on the preference for white. Color Research and Application, (1), 35-49.

Solomon, M. R. et al. (2012). Consumer behavior: Buying, having, being. Sydney: Pearson Higher Education.

Tao, H. F. et al. (2015). A study on format design in the context of new media. Art Education, (3),253-255.

Tatler, B. W. (2007). The central fixation bias in scene viewing: Selecting an optimal viewing position independently of motor biases and image feature distributions. Journal of Vision, (7),1-17.

Watson, J., & Hill, A. (1997). A dictionary of communication and media studies. London: Arnold.

Wei, Q. H. (2009). Studies on multimodality and multimodal discourse in visual surroundings.Beijing: Science Press.

Zhang, Z. (2013). Multimodal literacy development in classroom ecosystem. Jinan: Shangdong University Press.

Zhao, J. (2012). Traditional color in Chinese graphic design. Hundred Schools in Arts, (3), 229-230.

Zhu, Y. S. (2007). Theory and methodology of multimodal discourse analysis. Foreign Language Research, (5), 82-86.