在s – o – r模型中，有机体体现了情感和感知的中介状态。它可以调解刺激与反应之间的关系。Mehrabian和Russell(1974)将中介国家分为以下三个维度:快乐、觉醒和支配。快乐-不愉快的维度被用来描述享受和满足的感觉。不觉醒衡量的是个体受到刺激和活动的程度。支配性代表了一个人受到限制或自由的程度。虽然这三个维度在相关研究中经常被使用，但很少有实证结果支持支配-顺从维度的有效性。罗素和普拉特(1980)删除了主导因素，因为它不能代表个体对所有情况的情绪反应。罗素(1980)进一步认为，支配地位几乎没有什么适用的价值，而不是纯粹的预测。在实践中，Donovan和Rossiter(1982)揭示了买家的反应是在两个主要的方面——在零售商店里对mehrabian – mussell模型进行实证测试后的觉醒和愉悦。因此，最近对“PAD模式”应用二维方案的研究，而非最初的三维理论:快感和激励。因此，本文不考虑主导类别。s – o – r模型响应的最后一部分是中介状态的结果。Mehrabian和Russell(1974)将行为反应分为接近或回避行为。方法指的是想要停留并进一步探索存储环境，从而增加重复购物频率和货币支出。相反，一个有回避行为的人要么避免与环境互动，要么表现消极。
应用s – o – r模型和PAD分类，Donovan和Rossiter(1982)提供了第一次零售氛围的实证检验，表明大气线索，作为“刺激”，对购物者的规避行为有显著的影响，被概念化为“反应”。类似地，Arora(1982)用LISERL分析方法评估了s – o – r模型，并证实了该框架的有效性和可靠性，以及这三个组件的因果关系。
In the S-O-R model, organism embodies the affective and perceived intermediary states and. It serves to mediating the relationship between stimuli and response. Mehrabian and Russell (1974) classified the intermediary states into the following three dimensions: pleasure, arousal and dominance. Pleasure-displeasure dimension is being used to describe the feelings of enjoyment and satisfaction. Arousal-nonarousal measures the extent to which individual is stimulated and active. Dominance-submissiveness represents the degree to a person feels restricted or free. Although theses three dimensions are frequently used in relevant researches, there is little empirical findings supporting the effectiveness of dominance-submissiveness dimension. Russell and Pratt (1980) deleted the dominance factor since it failed to represent individual’s emotional response to all situations. Russell (1980) further argued that dominance provided little applicable value is not purely predictive. In practice, Donovan and Rossiter (1982) revealed that buyer’s response are in terms of two major aspects—arousal and pleasure—after empirically testing the Mehrabian-Mussell model in retail stores. Therefore, more recent researches of the PAD schema applied two-dimensional scheme instead of the original tridimensional theory: pleasure and arousal. Accordingly, the dominance category was not considered in this article. The last part of the S-O-R model—Response—is the outcomes of the intermediary states. Mehrabian and Russell (1974) dichotomised the behavioral responses into approach or avoidance behaviors. Approach refers to the desire to stay and further explore the store environment, which in turns increase repeat-shopping frequency and money expenditures. On the contrary, a person with avoidance behavior tend to either avoid interacting with environment nor perform negatively.
Applying both the S-O-R model and PAD classifications, Donovan and Rossiter (1982) provided the first empirical test of retail atmosphere, suggesting that atmospheric cues, as the ‘stimuli’, have a significant impact on shoppers’ approach-avoidance behavior which is conceptualized as ‘response’. Similarly, Arora (1982) assessed the S-O-R model with the method of LISERL analysis, and substantiated the validity and reliability of this framework as well as causality of these three components.