Relationship between environmental factors and facial expressions of visitors during the urban forest experience

Abstract A forest environment can promote the mental state of people and the effect can result within short frames (e.g., minutes of visit). Most studies to date have used questionnaire investigation or medical examination to study how forest affect people. However, these methods were not rapid enough to assess real-time emotion of visitors who were immersed in a forest experience. In this study, twenty-four university students (21 years old, 11 females and 13 males) were recruited to visit four urban forest-park locations or to walk along an urban street in two early-summer days. Participants were asked to take selfies every 30 min between the hours of 09:00 am and 14:00 pm by posing with their natural facial expressions and real-time emotion. Environmental factors were monitored simultaneously at every photographing phase. Photographs were analyzed using the FireFACE™ software to obtain scores for happy, sad, and neutral expressions, which were also used to calculate the positive response index (PRI) which is an expression of happy minus sad facial expressions. The forest experience evoked higher happy and PRI scores, but lower neutral scores compared to that in urban environment. Analysis of variance using a mixed model indicated a higher green-light spectrum-ratio in the sunlight spectrum, higher relative humidity (RH), lower temperatures, and lower noise in forests especially at noon and all were significant explanations of facial expression. Stepwise regression indicated that the green-light spectrum-ratio had a positive relationship with happy scores and the combination of lower noise and higher humidity contributed to a higher PRI in the forest. In conclusion, the forest experience can result in more positive emotional expressions on visitors’ faces, and the understory conditions of green colored light, less noisy peaceful atmosphere, and moist air measured through RH likely led to this experience.