人类使用媒介，媒介延伸人类的历史驰骛至今，留下一道道演进轴线如车辙印一般历历可考。尤其从大众媒体开始，媒介结构与自身逻辑逐渐清晰，版面铺展了报刊，频道串联了广播电视，窗口（windows）打开了桌面互联网，应用程序（applications）连接了移动互联网……

智能手机兴盛以来，用户的日常生活已经与数以万计的应用程序相互匹配、深度嵌合。在使用行为上突出表现为用户愈加频繁地在多种应用的勾连、流转中登录网络。因此越来越多的研究者呼吁，必须认真对待当前传播过程的“复媒介”或“跨媒介”现象，既将应用视为网络媒介系统中的基础单元，又从整体关系分析用户行为，避免以管窥豹，只见一斑。然而传统、单一、静态的媒介使用时间统计、内容分析已经无法有效反映动态而丰富的手机使用方式与连接状态。

如何击中当前人们使用手机应用的真正命意？问题的关键在于把捉应用之间的关系，以及应用如何进行通信转换——本文提出“切换”（switch）的研究视角和测量方法，用以分析手机应用的连接与交互方式。本研究中的“切换”是指用户跳转或并行使用多种应用的行为。

研究对45名青年群体用户的切换使用（以及15名平台开发者）展开深度访谈，并通过第三方应用收集了45名用户15天的手机行为数据，利用示能（affordances）理论分析框架与复媒介（polymedia）理论逐条递进地解析了核心研究问题。首先，从大众媒介的发展过程分析，切换使用从何时开始？形成了什么样的演进脉络？其次，从应用种类、使用时长、切换次数等指标分析用户使用应用的总体情况，即碎片化程度如何？再次，深入“碎片”勘探其中的连接远近关系，即切换中的应用群落；复次，深入“群落”发掘其中的时序先后关系，即应用切换的流动方向；最后，研究不止于数据描述，更深入用户使用场景与日常探询切换模式与原因。

研究在文献、数据与资料分析的基础上有如下发现：第一，切换行为在电子媒介发展过程中有三个阶段：一是从广播电视媒体开始的起始阶段，其特点是操作简单，缺乏交互，切换之间少有意义连接；二是从桌面互联网开始的加速阶段，用户基于台式电脑窗口的切换使用，任务、数量明显增加，交互程度愈加深入；三是从移动互联网开始的无序阶段，用户与网络信息深度交融，难以分离，使用行为非线性、多线程且迅速流动。研究进一步区分了大众媒体与网络媒介之间的切换差异，认为大众媒体的切换是中断，而网络媒介的切换是连接。

第二，文章通过媒体多任务指数与辛普森指数推导出了适用于应用切换的指数计算公式（ASI），用以全面反映应用使用的碎片化程度。统计数据显示极其碎片使用的同时应用分布高度集中，文章从用户与平台角度予以了解释。

第三，研究分析了应用之间的群落分布和流向。微信、QQ、微博等社交媒体的中心节点位置最为明显。社交类应用与音视图类应用的群落结构最为突出。社交类应用之间的信息吞吐量最大。研究从复媒介理论进一步探讨了应用之间存在的六种关系：配合关系、互补关系、增强关系、并行关系、冲突关系、中断关系，并结合访谈资料认为各种应用构建的社交示能强化了用户的切换行为。

第四，研究分别从网络突发新闻、压力型任务、娱乐型任务、网络购物、用户生产内容、无目的性使用等不同场景探究了可能存在的切换模式，与上述六种切换关系相互对应。

第五，研究在示能理论的分析框架下，解释了多应用切换行为的生成过程及其影响。研究认为，在媒介维度上，技术与设备的兼容造就了信息流动的环境，广泛连接与协调增强了信息流动，移动与便携致使信息更加无序，它们共同构成了切换的生成基础；在用户维度上，切换是为了“跟上网络信息的节奏”，即随时随地保持对网络空间信息刷新、变化的感知，并倾向于将线下生活发布、分享到网络平台，研究将之概念化为“信息同步感”（information synchronization），切换正是现实社会向网络社会转型的行为表征；在关系维度上，媒介与用户相互读取数据，共同迭代更新，背后的人机关系并不对等。切换的影响主要体现在效率、放松、倦怠、内疚与遗忘等方面。研究最后讨论了手机应用的社交属性、用户动态使用、信息同步感等关键概念及其延伸意涵。

With the use of human media, the media has extended human history. So far, it has left a rut-like evolutionary axis. Especially from the mass media, the media structure and its logic are gradually clear. The layout spreads newspapers and periodicals, the channels are connected to radio and television, the windows open the desktop Internet, and the applications are connected to the mobile Internet.

Since the prosperity of smartphones, users' daily lives have been matched and deeply embedded with tens of thousands of applications. The outstanding performance in the user behavior is that users log in to the network more and more frequently in the connection of a variety of applications. Therefore, more and more researchers appeal that we must seriously deal with the “multi-media” or “cross-media” phenomenon in the current communication, not only regard the application as the basic unit in the network media system but also analyze the user behavior from the overall relationship. However, traditional, single and static statistics of media usage time and content analysis can’t effectively reflect dynamic and rich network usage and connection state.

What is the real purpose of people using mobile apps? The key to the problem lies in grasping the relationship between applications and how to communicate with them. This paper puts forward the observation angle and measurement method of switch to study the connection and interaction of mobile applications. In this study, switching refers to the behavior of users jumping, crossing, or using multiple applications in parallel.

The research conducted in-depth interviews with 45 young users (and 15 platform developers), collected 15-day mobile phone behavior data of 45 users through third-party applications, and analyzed the core research problems one by one by using the analysis framework of affordances theory and the theory of poly media. First, from the analysis of the development process of mass media, when did the switching start? What kind of evolution context has been formed? Secondly, analyze the overall situation of users using applications from the indicators such as application types, use duration, and switching times, that is, how fragmented is it? Thirdly, explore the connection between far and near in applications, that is, the application community in switching; Fourth, explore the time sequence relationship in the community, that is, the flow direction of application switching. Finally, the research is not only data description, but also in-depth user usage scenarios and daily switching patterns and reasons analysis.

Through the analysis of literature, data and materials, the research has the following findings: First, the research puts forward three switching stages of the development of electronic media: First, from the mass media period represented by radio and television, it is characterized by simple operation, lack of interaction, and little meaningful connections between switching; Second, in the acceleration stage starting from the desktop Internet, the number of tasks has increased significantly and the degree of interaction has deepened based on the switching and use of desktop computer windows; The third is the disorderly stage starting from the mobile Internet. It is deeply integrated with users' network information, and it is difficult to separate it. Its usage behavior is nonlinear, multi-threaded, and fast-flowing. The research further distinguishes the switching difference between mass media and network media and holds that the switching of mass media is an interruption, while the switching of network media is a connection.

Secondly, this paper deduces the index calculation formula (ASI) suitable for application switching through the media multitasking index and Simpson index, which can fully reflect the degree of application fragmentation. Statistics show that users use extremely fragmented applications while the distribution of applications is highly concentrated. This article explains it from the perspective of users and platforms.

Thirdly, the community distribution and flow direction between applications are studied and analyzed. The central node attribute of social media such as WeChat, QQ, and Weibo is the most obvious. The community structure of the social application and the type of audio, video and picture application is the most prominent. The flow probability between social applications is the largest. This study further discusses the six kinds of relationships between applications in poly media Theory: cooperative relationship, complementary relationship, enhanced relationship, parallel relationship, conflict relationship and interruption relationship. Combined with the interview data, it is considered that the social display constructed by various applications can strengthen the switching behavior of users.

Fourthly, the study explores the possible switching modes from different scenarios, such as network breaking news, pressure tasks, entertainment tasks, online shopping, user generated content, and aimless use, and corresponds to the above six switching relationships.

Fifthly, under the framework of affordances theory, the generation process and influence of multi-application switching behavior are explained. The research shows that in the media dimension, the compatibility of technology and equipment creates the environment of information flow, extensive connection, and coordination enhance the information flow, and mobile and portable make the information more disordered, which together constitute the generation basis of switching; In the user dimension, switching is to “keep up with the rhythm of network information”, that is, to keep the perception of the refreshing and changing of network space information anytime and anywhere, and tend to publish and share offline life on the network platform. Conceptualized as “information synchronization”, switching is the behavioral representation of the transformation from real society to network society. In the dimension of relationship, the man-machine relationship behind reading data and iteratively updating with media users is unequal. The impact of switching is mainly reflected in four aspects: efficiency, relaxation, burnout, and forgetfulness. Finally, the key concepts and extended meanings of mobile applications are discussed, such as social attributes, users' dynamic use, information synchronization.