**3.2 Possible confounders**

The natural secretion of abovementioned biomarkers has a diurnal change: the highest level is in the morning and gradually decreases afterwards to the lowest level in the night time (it should be noted that only alpha-amylase act controversy; Strahler, 2010). Therefore saliva sampling should be conducted depending on the objective of a study: (1) it should be corrected in the afternoon where the secretion of biomarkers is expected to be stable in the case of laboratory stressor studies by which a level of transient increase of these biomarkers are focused, and (2) in the morning or just after awakening in the case of chronic stress study since basal secretion level is expected to show a remarkable difference among subjects in the morning (e.g. Steptoe, 2005). Because of its remarkable diurnal change, repetitive sampling by a distinct time point of a day is recommended even in a chronic stress study.

Postprandial effects on the secretion of biomarkers are considerable. In addition pH of saliva could affect the quantitative determination of biomarkers. Subjects should not take any food or drink except for water at least an hour prior to the experiment. Hard exercise should be avoided prior to the experiment since it could make biomarkers temporary elevate. Moreover it is strongly recommended to take 5 to 10 minutes of an initial rest period before conducting experiment. If subjects were not familiar with or nervous to an experiment, most of biomarkers would increase by such a negative feeling or strain. Subjects who take any medications, especially oral contraceptive, suffer from any disease, are pregnant, or other cases in which physiological states were considered to be unusual should be excluded. Age and gender should be balanced in groups of between-subjects study since it is known to be confounders, especially for the study assessing hormones.

Subjects must be well informed about the objective and method of the study. Any experimental design targeting on human mental or physical stress should be approved by a local ethics committee or equivalent organization.
