Keywords

1 Introduction

As globalization and technological innovations continue to support the rise of multinational corporations, it has become increasingly pressing for business to cater to a worldwide audience. Particularly, websites from these companies must be both navigable and easily comprehensible to individuals across cultures to cater to global audiences. Interestingly, however, according to the authors’ knowledge, few studies have explored how individuals from different cultural backgrounds may process information differently and, thus, interact with websites distinctively. In this study, the researchers explore how cross-cultural differences influence user experience and interface interactions across cultural environments.

Specifically, the researchers investigate how individuals from China, who tend to process information holistically, differ from Westerners, who tend to process content analytically, in navigating two news sites. To examine distinctions in navigation patterns, two interactive websites were built reflecting Western (i.e., BBC.com) and Eastern (i.e., QQ.com) designs. Further, participants were subject to tests measuring response times to search queries in addition to surveys measuring their evaluations to the sites. Specifically, we had participants respond to questions gauging their self-reported perceptions of the news sites.

2 Literature Review

There currently exits and extensive body of research on holistic versus analytical information processing, particularly in the cross-cultural literature. Prior studies have found that individuals from different cultures tend to process visual stimuli distinctively. In particular, the influence of culture on information processing appears most relevant when considering how Easterners (e.g., Chinese, Japanese, etc.) and Westerners (e.g., American, Swedish, etc.) perceive information differently [2]. Specifically, research suggests Westerners, who tend to be individualists, interpret content analytically while Easterners, who tend to hold collectivists values, generally process information holistically [3]. The difference between Western analytical thinking and Eastern holistic thinking is that holistic thinking emphasizes the interconnectedness of perceptual stimuli, while analytical thinking involves the interpretation of objects independently (e.g., in isolation of surrounding cues) [1].

It is believed that Easterners are able to synthesize and consider more pieces of information relative to their Western counterparts. For example, in eye-tracking studies, East Asians tended to scan through an entire image, bouncing between focal and background objects. Conversely, eye-tracking heat maps indicate that Western subjects did the opposite; instead of taking account of all the objects in an image, Westerners tended to gaze extensively at focal objects. As such, Westerners were able to recall more details surrounding central objects, while Easterners made saccades between various information cues. As such, holistic processing “involves an orientation to the context or fields as a whole” [4] whereas analytical processing “involves a detachment of the object from its context, a tendency to focus on attributes of the object” [1]. These findings suggest that Eastern collectivists (e.g., Chinese, Korean, Japanese people), will be more likely take in more information content than Westerners. Additionally, they may be more inclined to incorporate non-central cues for information processing. Conversely, it is suggested that Western analytical thinkers may prefer to focus on fewer objects and may tend to use only central information.

2.1 F-Shaped Pattern

The F-shaped pattern refers to findings made in a study on user eye movements [7]. This pattern has been dubbed the “F-shaped pattern” since prior research suggests that users typically scan pages starting with a horizontal movement, usually across the upper part of the content area. Users then tend to read across in a second horizontal movement further down on the page that typically spans a shorter area. Finally, users scan the content’s left side in a vertical movement. When measuring users’ eye gazing as a heat map, this creates a pattern that resembles an F. Web developers, either consciously or unconsciously, often design their websites according to this pattern. The F-shaped pattern is not an absolute law and several other scanning patterns exist, but the F-shaped pattern remains the most prevalent in Western cultures [8]. If a developer designs a page without knowledge about this pattern, they risk putting important information in places where users might miss it.

3 Methodology

The aim of the investigation phase is to obtain a more nuanced understanding regarding the differences in perception and navigation of Western versus Chinese websites. This information facilitated the decision-making process concerning which features to keep, which features need to be analyzed, and what usability metrics should be implemented in the study. As an initial step, Chinese websites, along with their Western equivalents, were identified and curated. After obtaining a corpus of websites from China and Western countries (e.g., United Kingdom), the key design differences between these respective sites were documented. Once the primary design variances were explored, this information was used to decide which design patterns should be tested in the study. Finally, the specific metrics to measure the results of these different designs were concluded. In reviewing the websites, it was evident that Chinese pages differed significantly from Western counterparts. The differences between sites extended from the look and feel to the UX design. A few of China’s most popular browsing sites were then analyzed and compared to similar Western counterparts to further investigate these design variations.

BBC is a very popular Western new site. The design of the site follows similar trends, in terms of information design and density, of other Western news media sites. The layout of various landing pages in the BBC appear to closely follow the F-shaped pattern. Each news section has a large news figure in the top left corner and two rows. Simply looking at the large picture and the first row then glazing down to the next news section would have the users quite naturally follow the F-shaped pattern. The site itself is quite information sparse with large images taking most of the space. Without hovering over any content on a standard computer screen, there are roughly 48 clickable elements [10].

QQ is one of the most visited media websites in China [11, 12]. QQ, like many other Chinese websites, does not focus on one thing but has multiple functions. Some of the functions that QQ supports are instant messaging, online games, music, shopping, microblogging, news, movies, group chat software, and etc. On the QQ homepage, users are greeted by the site’s news page, which is highly information dense. Without hoovering over any content on a standard computer screen, there are roughly 147 clickable elements. One element that is notably common on Chinese websites, including QQ, is the menu bar design. On the QQ page, the menu bar contains two rows with a total of 40 clickable options. This format of menus is typical in China and is shared by multiple other Chinese sites (e.g., Sina, Taobao).

4 Analysis

For the purposes of this research two interfaces were created, with one following a Western design (e.g., similar to BBC) and the other one using a Chinese layout (e.g., similar to QQ). Both these sites were translated to both English and Chinese. The translation process was iterative and two bilingual researchers performed the meaning translations. The goal of implementing these interfaces were to explore how fluidly users from different cultural backgrounds can navigate sites containing high information density (i.e., copious amounts of images and texts). Both interfaces contain roughly equivalent levels of material and clickable elements; the primary difference is that the Western site became longer (due to translations), forcing users to scroll down the page. Additionally, some of the information were mapped in sub-menus using natural mapping for the Western site [5].

Conversely, the Chinese inspired site provided most of the material directly on the screen for users to view without any nested menus. The interfaces, cable of measuring what actions users take in responding to certain tasks, was developed given the feedback from these prototypes. Main measurements that were used are task-success, time-on-task and a modified System usability scale [9]. Research from Nisbett and Miyamoto [1] suggest perception differs in Western and Eastern cultures. Dong and Lee [13] further indicate that this perception difference holds true in the case of people observing websites. Specifically, users using analytical processing follow the F-shaped pattern when browsing sites [7]. Holistic thinkers, conversely, do not follow the F-shaped pattern when browsing through a website [13]. In light of these past studies, it will be interesting to explore how these perception differences will influence users’ abilities to navigate and perceive web pages.

To investigate this question, we select elements both in accordance to the F-shaped pattern and elements outside of this pattern. By testing the performance on analytical and holistic thinkers, an indication of the differences and how well people follow the F-shaped pattern when looking for specific elements will be derived. The test was unsupervised, as a larger test audience was necessary in order to obtain significant results. To measure variations in perception, tests for the sites BBC and QQ were created. On these pages, participants were requested to find elements following an F-shaped pattern and elements that not following this pattern.

Results on time differences and survey responses were stored in a database for significance testing. The corpus of collected data were applied to statistical significance analyses of differences between groups. Our findings indicate that (1) there is a significant difference between how Chinese and Western users process information (t = −3.1606, p < 0.05), (2) Westerners are more reliant on the F-shaped pattern than Easterners (t = −3.1606, p < 0.05), and (3) Chinese individuals are comfortable with Western pages (t = 5.3301, p < 0.05), suggesting that a universal interface can be deployed cross-culturally. This study offers contributions both to theoretical understandings of cross-cultural user experience as well as industry practitioners. Findings from this study suggest that culture significantly influences information processing which subsequently effects how users evaluate and navigate websites

5 Discussion

The aim of this research was to examine (1) how differences in interface design may be due to differences in information processing styles or simply trends, (2) how do different processing styles in Western (analytical) versus Chinese (holistic) users affect performance on different interfaces, and (3) if one global interface should be created or if web designers should focus on creating separate user interfaces for different cultures. Concerning the question of how interface design may be due to different information processing styles or trends, my research suggests that a mix of both factors have contributed to the different web designs from China and its western counterparts. Specifically, Chinese users preferred the Western based design over the Chinese one, according to self-reports (M = 4.11 for QQ and M = 5.04 for BBC).

Interestingly, this suggests that perhaps the reason why Chinese news sites tend to be more information dense may simply be due to a trend in web design rather than fundamental perception differences. For instance, Chinese users performed as well as English users navigating both the BBC site (t = −0.49, p > 0.05) and QQ site (t = 0.42, p > 0.05) indicating that there is no significant difference between the two groups. Regarding the question of how information processing styles (analytical versus holistic) would affect user performance on different interfaces, the findings suggest that information processing style does significantly influence performance. Unsurprisingly, the research indicated that English, or analytical-thinking, users who used the BBC site were much quicker at finding important objects within the F-shaped pattern compared to outside of it (t = 2.8479, p < 0.05). On average, these participants took 12 s longer outside the F-shaped pattern compared to within. This is congruent with the existing literature on UX, in which eye tracking studies suggest Western users tend to scan through a website using an F-shaped pattern.

However, it was surprising to find that Chinese users were actually also able to answer questions more quickly when questions were within the F-shaped pattern versus outside (t = 5.3301, p < 0.05). It appears that for both Chinese and English users of the BBC site, objects that fell within the F-pattern were easier to find that those outside. However, the effect size was significantly larger for English users, with a 12 s difference inside and outside the F-pattern, than for Chinese users, who only had a 4 s difference. Overall, it appears that a Western based design seems easily navigable for both Eastern and Western users. The F-shape appears to increase performance for both groups. The main difference, however, is that English users were significantly better within the F-pattern (taking 12 s less) than outside, whereas Chinese users only saw a slight improvement (taking only 4 s less within the F-shape).

On the effects of information processing and website navigability, the results suggest that English speakers using the QQ site were quicker inside the F-shaped pattern than outside of the pattern (t = −3.1606, p < 0.05), with a 3 s difference. For Chinese users on the QQ site, conversely, they were marginally slower inside the F-pattern (using 22.6 s) than outside (22.4 s) the pattern (t = −5.907, p 0.05), English users were still slightly more dependent on the F-pattern, like the findings form the BBC site. Interestingly, however, according the questionnaire self-reports, it appeared that Chinese users found both the QQ and BBC websites to be similarly likeable (M = 4.11 for QQ vs. M = 5.04 for BBC) and did not seem to be overwhelmed by the information density of the QQ site (M = 2.8 on the question “I felt overwhelmed using this site”). For the English users, however, self-reports indicated that they were highly uncomfortable with the layout of the page, finding the information overwhelming. They had an average of 4.23 from 1 to 5 on the question “I felt overwhelmed using this site”. Western users also tended to give higher ratings for BBC than QQ (M = 4.46 for BBC vs. M = 2.28 for QQ).

Taken together, these results suggest that information processing does influence how users interact with different interfaces. It appears that while Western users can use a Chinese designed site, they find the experience to be extremely unpleasant (M = 2.28 for QQ). Conversely, the findings indicate that Chinese users found both sites similarly likeable (M = 4.11 for QQ vs. M = 5.04 for BBC). Furthermore, English users were more reliant on the F-shaped pattern than Chinese users, which is in-line with existing research on user perception. Concerning the question of whether or not one global interface should be created or if web designs should be tailored to different cultures, these results suggest that one global interface can be deployed to maximize efficiency. Chinese users found BBC to be just as likeable as QQ, whereas English users disliked the information density of QQ, suggesting that a site with lower information density is favorable. Accordingly, it appears that web designers can focus on creating one global interface to maximize efficiency.