许多研究尝试探究应激水平、消极情绪、积极情绪与进食行为的关系，但是，这些研究存在明显不足。前人研究大多使用实验法。虽然能记录精准的进食指标，但是生态效度较低。在少部分使用问卷法的研究中，对于进食行为的记录完全依靠主观评价，无法记录饮食热量。极少研究使用生态瞬时评估法，而这种方法可以记录应激水平、情绪和饮食热量在短时间内波动的情况，是研究本问题的最佳方法。

本研究的目的是探讨瞬时应激水平、消极情绪、积极情绪与饮食热量的关系以及这些关系中的调节变量。本研究选取在校大学生67名，研究时长为8天。研究开始前，使用基本信息问卷收集人口学信息，采用荷兰进食行为问卷测量进食风格，采用中文版整体营养知识问卷测量饮食知识。研究开始后，主试每天在微信群中分别于五个时间段发布生态瞬时评估问卷，收集被试当前的应激水平、积极情绪、消极情绪以及从上一次测量以来摄入的所有食物的种类和热量。研究期间，被试在摄入任何食物后都需要使用“薄荷营养师”微信小程序记录饮食重量或分量，该程序可自动得到饮食热量，以便在生态瞬时评估问卷中填写相关内容。研究结束后，根据被试填答率发放被试费。

使用多层线性回归分析数据。结果显示，被试间消极情绪可以显著负向预测饮食热量，而应激、积极情绪和被试内消极情绪均不能显著预测饮食热量。至于调节效应，本研究发现，外部性进食风格可以调节被试内应激水平与饮食热量之间的关系。BMI、性别、饮食知识、情绪性进食风格、限制性进食风格均不能调节被试内应激水平、积极情绪、消极情绪与饮食热量之间的关系。

本研究表明，个体间与个体内应激/情绪对进食行为的影响存在差异，且此后探究应激、情绪与进食行为时应重点关注可能的调节变量。

Previous studies have explored the relationship between stress/emotions and eating behavior, but these studies have room for improvement. First, most of the studies used experimental methods. Although accurate indicators could be recorded, the ecological validity is low. Secondly, in a few studies using questionnaires, the recording of eating behavior completely depends on self-report, which may lead to potential bias. Third, few studies have used ecological momentary assessment, which can record fluctuations in stress, emotions and eating over a short period of time and is a good way to study this question.

The purpose of this study was to explore the relationship between momentary stress, negative emotion, positive emotion and dietary calories and the moderating variables between these relationships. This study recruited 67 college students for and 8-day study. Before the study, demographic information was collected using the basic information questionnaire; eating style was measured by the Dutch Eating Behavior Questionnaire (DEBQ); food knowledge was measured by the Chinese version of the General Knowledge of Nutrition Questionnaire (C-GNKQ). After the beginning of the study, we released EMA questionnaire in the wechat group for five times per day, collecting the participants’ momentary stress, positive emotions, negative emotions, and the names and calories of all the food they consumed since the last measurement. During the study, the participants were required to use the "Mint Nutritionist" wechat mini program to record the weight and calories after eating any food, in order to fill in relevant contents in the EMA questionnaire. At the end of the study, the participants will be compensated according to their response rate.

Multilevel linear regression was used to analyze the data. The results showed that the between-subject negative emotions negatively predicted caloric intake, while stress, positive emotions and within-subject negative emotions could not significantly predict caloric intake. As for the moderators, the study found that external eating style could moderate the relationship between within-subject stress and calorie intake. BMI, gender, dietary knowledge, emotional eating style and restrictive eating style could not moderate the relationship between stress, positive emotions, negative emotions, and calorie intake.

This study shows that there are differences between the between-subject and within-subject effect of stress/emotions on eating behavior. What’s more, the potential moderators should be focused on while exploring daily eating behavior.