Miller et al. have rightly identified serious and long-standing issues in social psychological research and have suggested means by which the availability of modern technology can go far toward resolving these issues. Their target article could not be more timely in light of the intense debate concerning the replicability of research findings and more generally, the status of social psychology as a mature science and its potential for real-world applications. The crux of their argument is that the historical tradeoff between the precision enabled by laboratory experiments and the generalizability of research findings to real-world contexts can be overcome with an approach referred to as Systematic Representative Design (SRD). The primary components of SRD are immersive environments such as Virtual Reality (VR) and computer games in which the social environment may be represented by real people (e.g., as avatars) or agents who may be equipped with artificial intelligence. In this approach, computer simulations may be used to identify crucial independent variables to be manipulated in experiments. They build a convincing case for the SRD approach, arguing that it allows for precise causal inferences that are not limited by the laboratory contexts in which such inferences are generated. VR, in particular, allows one to provide a context for participants that is highly realistic and immersive and also allows for precise control of the values of the independent variables, while limiting the influence of uncontrolled variables. The identification of cause-effect relations is the essence of theory construction and has been widely recognized as such as far back as John Stuart Mill, who developed the canons of causal reasoning. Although causality undeniably underlies social psychological processes, in many instances it may not be the best tool for the understanding and prediction of such processes because the interaction over time of many casual mechanisms can give rise to phenomena that are better described in other theoretical terms. In particular, recent advances in nonlinear dynamical systems, statistical physics, network science, and other approaches that can be collectively described as complexity science have broadened the scope of explanation in social psychology beyond the traditional focus on simple cause-effect relations (e.g., Butner, Gagnon, Geuss, Lessard, & Story, 2015; Guastello, Koopmans, & Pincus, 2009; Nowak & Vallacher, 1998; Vallacher, Read, & Nowak, 2002; Vallacher & Nowak, 1997). Several features of the complex systems perspective are especially relevant to understanding how intrapersonal and interpersonal systems operate. Our aim here is to highlight these features, some of which may be incorporated into the SRD approach. In combination with other recent developments in gathering and analyzing data, the SRD approach provides cause for optimism that a mature social psychological science is a feasible goal. As Miller et al. point out, the tradition in experimental psychology is to build theories in terms of cause-effect relations, in which the cause is operationalized as a change in an independent variable at Time 1 and the effect is operationalized as the resultant change in the state of a dependent variable at Time 2. This assumes unidirectional causality, such that one can clearly identity the cause and the effect in a causal relationship. While this may be necessary in a traditional laboratory experiment, it does not do justice to the essence of social psychological processes in real-world contexts. As noted by Bandura (1986) and others, social and psychological processes are commonly characterized by bidirectional or reciprocal causality, such that each variable both influences and is influenced by the other. In this scenario, Variable A at Time 1 influences the state of Variable B at Time 2, and Variable B at Time 1 influences the state of Variable A at Time 2. Multiple repetitions of this mutual influence may result in a prolonged temporal pattern of change, in which the values of each variable are adjusted in response to changes in the values of the other variable. In the complex systems perspective, the fact that each variable can play the role of both cause and effect can be reframed in terms of feedback loops that operate in an iterative fashion to promote patterns of stability and change in physical and psychological phenomena (e.g., Holland, 1995; Kelso, 1995; Schuster, 1984). These patterns, resulting from the mutual influence among two or more dynamical variables (i.e., variables that change in time), constitute the intrinsic dynamics of a psychological process (Vallacher, Van Geert, & Nowak, 2015). The ubiquity of intrinsic dynamics, in which there is often a sustained flow of thoughts, feelings, and actions, was accorded central status by early theorists such as James (1890), Cooley (1902), and Lewin (1935). In mental systems, for example, a thought with a particular value (e.g., negative evaluation) can trigger a feeling with a particular value (e.g., an intensity of anger), and that feeling

中文翻译：

---

乐观的原因？

米勒等人。正确地确定了社会心理学研究中严重和长期存在的问题，并提出了现代技术的可用性可以在很大程度上解决这些问题的方法。鉴于关于研究结果的可复制性的激烈争论，以及更普遍的社会心理学作为一门成熟科学的地位及其在现实世界中的应用潜力，他们的目标文章再及时不过了。他们争论的关键是，可以通过一种称为系统代表设计 (SRD) 的方法来克服实验室实验所实现的精度与研究结果对现实世界环境的普遍性之间的历史权衡。SRD 的主要组成部分是沉浸式环境，例如虚拟现实 (VR) 和计算机游戏，其中社交环境可以由真实的人（例如，作为化身）或可能配备人工智能的代理来表示。在这种方法中，计算机模拟可用于识别要在实验中操作的关键自变量。他们为 SRD 方法建立了一个令人信服的案例，认为它允许进行精确的因果推断，而不受产生此类推断的实验室环境的限制。特别是 VR，它可以为参与者提供高度逼真和身临其境的环境，还可以精确控制自变量的值，同时限制不受控制的变量的影响。因果关系的识别是理论构建的本质，早在约翰·斯图尔特·米尔（John Stuart Mill）就已被广泛认可，他开发了因果推理的经典。尽管不可否认，因果关系是社会心理过程的基础，但在许多情况下，它可能不是理解和预测此类过程的最佳工具，因为随着时间的推移，许多偶然机制的相互作用可能会产生可以用其他理论术语更好地描述的现象。特别是，非线性动力系统、统计物理学、网络科学和其他可以统称为复杂性科学的方法的最新进展扩大了社会心理学的解释范围，超越了对简单因果关系的传统关注（例如，Butner , 加格农, 盖斯, 莱萨德, & 故事，2015；Guastello、Koopmans 和 Pincus，2009 年；诺瓦克和瓦拉赫，1998 年；瓦拉赫、雷德和诺瓦克，2002 年；瓦拉赫和诺瓦克，1997 年）。复杂系统视角的几个特征与理解内部和人际系统如何运作特别相关。我们在这里的目的是强调这些特征，其中一些特征可能会被纳入 SRD 方法中。结合最近收集和分析数据方面的其他发展，SRD 方法为成熟的社会心理科学是一个可行的目标提供了乐观的理由。正如米勒等人。指出，实验心理学的传统是根据因果关系建立理论，其中原因在时间 1 时被操作化为自变量的变化，而结果在时间 2 时被操作化为因变量状态的结果变化。 这假设了单向因果关系，这样人们就可以清楚地识别原因和因果关系中的影响。虽然这在传统的实验室实验中可能是必要的，但它并不能公正地反映现实世界环境中社会心理过程的本质。正如 Bandura (1986) 和其他人所指出的那样，社会和心理过程通常以双向或互惠的因果关系为特征，这样每个变量都会影响另一个变量，也受另一个变量的影响。在这种情况下，时间 1 的变量 A 影响时间 2 的变量 B 的状态，时间 1 的变量 B 影响时间 2 的变量 A 的状态。这种相互影响的多次重复可能会导致长期的时间变化模式，其中每个变量的值都会根据另一个变量的值的变化进行调整。从复杂系统的角度来看，每个变量都可以发挥因果作用的事实可以根据以迭代方式运行的反馈循环来重新构建，以促进身体和心理现象的稳定和变化模式（例如，Holland , 1995; Kelso, 1995; Schuster, 1984)。这些模式由两个或多个动态变量（即随时间变化的变量）之间的相互影响而产生，构成了心理过程的内在动态（Vallacher、Van Geert 和 Nowak，2015 年）。内在动力无处不在，其中经常有持续的思想流动，早期的理论家，如詹姆斯（1890 年）、库利（1902 年）和勒温（1935 年），都将感情和行为置于核心地位。例如，在心理系统中，具有特定价值（例如，负面评价）的想法可以触发具有特定价值（例如，愤怒的强度）的感觉，并且该感觉