Measurement Type What is Measured / Instrument Pros & Cons Physiological Sensors Measures like galvanic skin response (GSR) , skin temperature , heart rate variability (HRV) , sometimes cortisol etc. These capture bodily arousal (which is often part of feeling uneasy). Example in Salzburg paper: wristband sensors. + Objective; less biased by self-report. − Might be noisy / influenced by many factors (heat, hydration, movement). − Need baseline / calibration. Video / Computer Vision Using cameras to capture crowd density, proximity of people, obstacles, etc. In this paper, Mask R-CNN to identify humans, vehicles, bikes etc. + Can quantify features of the environment objectively (density, crowd flow etc.). − Privacy issues; processing complexity; might not capture subjective feeling. Self-Report / Diaries / Surveys Asking subjects how they felt (e.g. “Did you feel stressed?”, “Did you feel crowded?”, “How comfortable/uncomfortable were you?”), often via experience sampling (diaries or apps) or brief post-exposure surveys. E.g., the Salzburg study used an eDiary app for negative feeling reports. + Captures subjective experience directly. − Bias possible (memory bias, social desirability, etc.). − Hard to get frequent reports without being burdensome. Behavioural Observations Observing how people behave in crowded settings: do they avoid the center, do they move slower, are there more overstepping of boundaries (like stepping onto dangerous zones), more fidgeting, maybe body posture etc. E.g., another study looked at stepping into danger zones on train platforms when crowd density is high. + Taps into what people do , not just what they feel. − Interpretation sometimes ambiguous: is stepping forward because of discomfort, necessity, or ignorance? − Needs good observational tools / video, possibly coding. Perceived Crowd / Density Estimation Subjective estimation: “How many people do you think are around you?”, “How dense does the crowd feel?”, “How close do people feel physically?”, “How much personal space do you feel is being violated?” Also measures of safety perception, risk, etc. E.g., in train station platform study, they had surveys about safety perception and density estimation. + Measures subjective perception, which often matters more for feelings of unease than objective density. − Subjective perceptions can vary widely. − Requires well-worded questions. Mixed Methods Combining above (e.g., physiological + video + survey) to triangulate: do objective signals correlate with self-report? This gives stronger validity. The Salzburg paper is a good example. How Behaviour / Discomfort is Measured in This and Related Studies From this and similar crowd / crowding-studies, here are the common measurement methods, i.e. how uneasiness in crowd is operationalized and quantified.