Design and procedure
A pretest-posttest control group design was used to evaluate the psychological benefits of Mediterranean FB in different forests of Catalonia compared with guided hiking. Most of the Catalan territory lies on the northeast of the Iberian Peninsula, to the south of the Pyrenees Mountain range. The capital and largest city Barcelona is the second-most populated municipality in Spain and the fifth-most populous urban area in the European Union (Demographia 2021).
The intervention followed the procedure employed in a pilot study carried out in May 2020 in the general population (Muro et al. 2022). The current study included two experimental condiditons distributed in different locations:
- Experimental group. It included participants that took a FB guided by a specialized therapist and a nature guide, in two different main different allocations: 1) Vallès Occidental, and 2) la Garrotxa area, Both spaces are located at a linear distance of 70 km, they represent the Mediterranean forest, woolands, and scrub biome, and according to the World Wide Fund for Nature they are included in the Northeastern Spain and Southern France Mediterranean forests Olson and Dinerstein (1998).
1) The first group was recruited in collaboration with the Diputació de Barcelona (a public administration that promotes the progress and well-being of the citizens and local governments of its territorial area) and the Network of Natural Parks of Catalonia (https://parcs.diba.cat), announcing the activity for the general population during the first waves of the pandemic: one for October 2020 (https://parcs.diba.cat/web/agenda/-/sant-lloren%C3%A7-obac-banya-t-al-bosc-experi%C3%A8ncia-immersiva-a-la-m%C3%A0gia-de-les-arenes) and the other for May 2021 (https://parcs.diba.cat/web/agenda/-/sant-lloren%C3%A7-obac-banya-t-al-bosc-experi%C3%A8ncia-immersiva-a-la-m%C3%A0gia-de-les-arenes-1). The activity was framed within the Ginesta project of environmental education and the European Charter for Sustainable Tourism, which is a set of good environmental practices for the management of tourism in protected natural areas, promoted by the Europarc Federation (Canals 2014; Europarc Federation 2010).
The forested area chosen for the FB was located in Les Arenes de Sant Llorenç (geographical coordinates: N 41º 38.799 N, E 2º 3.457): it is an area of a mountain Mediterranean forest in Sant Llorenç del Munt i Serra de l’Obac Natural Park (Catalonia, Northeasteast of the Iberian Peninsula), a territory included in the central pre-coastal Mediterranean climate and close to urban centers. The base of the massif is occupied by white pine groves (Pinus halepensis), very resistant to water scarcity and up to 600 masl, often being replaced in the darkest and/or tallest areas by Pinus sylvestris and Pinus nigra. These pine forests, for the most part, are the result of the human transformation of the primitive forest and in many areas appear mixed with holm oaks (Quercus ilex) and Mediterranean shrubs such as heather (Erica) and strawberry (Arbutus unedo). The holm oak groves are the characteristic and most widespread vegetation of the natural park, which above 800 masl is enriched with species typical of wetlands such as whitebeam (Sorbus), boxwood (Buxus) and oak (Quercus humilis); Lorenzo and Fernández (2009). The walk included 2 stops (one at the beginning and one in the middle of the walk) to practice short mindfulness-meditations of 15 minutes each to train deep breathing and mindful awareness of the 5 senses.
2) The second experimental group took a FB in different forests of la Garrotxa area: Les Olletes (geographical coordinates: N 42° 9.886; E 2º 31.831) Serra d'Heures (geographical coordinates: N 41° 53.968; E 2º 31.683), , and Can Fornaca (geographical coordinates: N 41° 48.200; E 2º 38.979). The region of La Garrotxa is located at the eastern end of the Catalan Pre-Pyrenees and is known as the volcanic area of Catalonia. Its geographical location makes it attractive both for its proximity to the seasuch as proximity to the mountainous regions of the country. This whole areaforms part of the Natura 2000 Network, an European initiative to protect the most unique, diverse, rare, well conserved, representative, fragile or vulnerable natural spaces and which share with each other similar characteristics. The climate of La Garrotxa is medium mountain Mediterranean. Rainfall is abundant throughout the year, and the winter is the driest season. The frequent showers keep summers cool, while the influence of the Pyrenees makes the winters very cold. The vegetation reflects the climate. While in the Alta Garrotxa and the east of the county there is an area of typically Mediterranean vegetation, the rest of the county is covered with sub-mediterranean vegetation that becomes Atlantic in the most humid parts. Holm oaks are typical on the sunny hillsides of the eastern sector, with relatively dry and hot summers, the oak tree predominates in other parts, and the beech is found in the most humid sectors. On the Olot plain, the most characteristic natural species is the common oak, although nowadays its presence is limited to small woods found around the city. This group of participants was recruited in collaboration with Sèlvans association (https://selvans.ong/). Sèlvans Association is a non-profit organisation that was born in 2007 with the main aim of conserving the Catalan forests.It organises activities related to environmental education and nature conservation, such as FB.
- Control group It included two hiking-groups that did not take a FB, but practiced a group-guided activity such as hiking. Hiking is a walk, usually on trails or footpaths in the countryside. The activities were organised in collaboration with the Network of Natural Parks of Catalonia and the School of Nature La Muntada (https://www.lamuntada.cat/ca), located in the Vall d’Horta of Sant LLorenç Savall (geographical coordinates: N 41º 40.638; E 2º 1.864). One grup hiked around 8km. from la Vall d’Horta to a small-mountain called els Emprius-la Roca Foradada with an unevenness/inclination of 250m (https://parcs.diba.cat/web/agenda/-/sant-lloren%C3%A7-obac-els-emprius-el-queixal-corcat-i-el-cau-dels-emboscats). A second control group hiked around 3 km. without inclination, along a plain and flat route in la Vall d’Horta https://parcs.diba.cat/web/agenda/-/sant-lloren%C3%A7-obac-passejada-teatralitzada-joan-oliver-al-marquet-de-les-roques-1. The two groups of activity that are reported as control condition were selected from among the cathalog of recreational and sports activities offered in the catalogue of activities of the institutions that have promoted the forest bathing experiences analysed in this study. For the selection of these activities, the researchers set the following inclusion criteria a priori: (1) Being a guided activity, (2) practiced in the same geographical spaces, (3) qualitatively different from forest bathing, (4) with the same duration and (5) intensity. Of the two selected groups, second one met the 5 criteria and the first one 4 of them (all except low intensity). This gave rise to a statistical control of this possible confounding variable that is explained later.
A call for participation was made through the website of the corresponding organisations. Prior registration was requested to participate in all the activities. All participants consented to collaborate voluntarily and provided informed consent before participating in the study. The same pre-post surveys were administered immediately before and after all the activities online via Google Forms©, but also using paper-pencil questionnaires in the cases in which no internet-connection was available. The ethics committee of the Autonomous University of Barcelona (CEEAH-UAB) approved the study in advance (reference code UAB5339).
Participants
95 people signed up for the events. A final sample of 86 volunteers participated in the study, including 57 participants in the FB session (Sant Llorenç n = 34; la Garrotxa n = 23), and 29 in the control group of hiking practice (with inclination n = 18; plain n = 11). The final sample analysed recorded an average age of 49.6 years (ranging from 19 to 78). Most of participants were residents in urban areas (75.6%), were women (69.8%) and had higher education (76,7%) (see Table 1).
Instruments
Both FB and hiking participants responded 4 standardised tests before and after the FB session. These questionnaires are widely used in the study of the psychological effects of FB and are cross-culturally validated for both the general and clinical populations (Hansen et al., 2017; Oh et al., 2017; Timko-Olson et al., 2020; Wen et al., 2019):
1. State-Trait Anxiety Inventory (STAI; Guillén-Riquelme and Buela-Casal (2011); Spielberger et al. 1982): This test measures the levels of state (at the moment) and trait (global personality) anxiety with a total of 20 items in each scale and format of Likert type answers. In the present study, only state-anxiety was measured. High scores warn of altered states related to anxiety, low scores indicate emotional stability and absence of stress. In the original Spanish validation, STAI-state showed an internal consistency of Cronbach’s α = .90. In the study sample, STAI-state showed an internal consistency of Cronbach’s α = .88.
2. Positive Affect and Negative Affect Scale (PANAS; Watson et al. 1988; López-Gómez et al. 2015): It includes two subscales of 10 items each that assess the experience of positive emotions related to psychological well-being and experiencing negative emotions related to psychological distress and mental health problems. In the original Spanish validation, positive affect subscale showed an optimal internal consistency of Cronbach’s α = .88, and negative affect subscale showed an internal consistency of Cronbach’s α =. 87. In the study sample, positive affect subscale showed an optimal internal consistency of Cronbach’s α = .91, and negative affect subscale showed an internal consistency of Cronbach’s α =. 910.
3. Profile of Mood States (POMS; McNair et al. 1971; Andrade et al. 2010). This test measures 6 moods from 30 items: Anger, fatigue, vigour, friendship, tension, and depressive mood; each item can be scored in a scale ranging from 0 to 4. In the original Spanish validation, the subscales showed an internal consistency of Cronbach’s ranging from α = .77 to Cronbach’s α = .92 for the six subscales. In the study sample, the subscales showed an internal consistency of Cronbach’s ranging from α = .92 to Cronbach’s α = .96 for the six subscales.
4. State Mindfulness Scale (M-E; Tanay and Bernstein (2013)): It is composed of 21 items with a Likert-type response scale to indicate whether the sentences describe well their experiences in the last 15 minutes. It assesses two dimensions: 1) Mindfulness state of mind (eg, "I realized thoughts coming and going") and 2) Mindfulness state of the body (eg "I realized physical sensations coming and going"). In the original Spanish validation, mind mindfulness subscale showed an optimal internal consistency of Cronbach’s α = .95, and body mindfulness subscale showed an internal consistency of Cronbach’s α = .90. General score of mindfulness also showed an optimal internal consistency of Cronbach’s α = .95. In the study sample, mind mindfulness subscale showed an optimal internal consistency of Cronbach’s α = .94, and body mindfulness subscale showed an internal consistency of Cronbach’s α = .90. General score of mindfulness also showed an optimal internal consistency of Cronbach’s α = .96.
Data preparation and statistical analyses
Statistical analyses were conducted using the IBM-SPSS 26.0 statistical package. No participants had to be discarded due to significant data loss. The omission of answers in some of the participants in the little group of paper-and-pencil data gathering procedure (less than 0.5%) explains the slight differences in the degrees of freedom indicated in the hypothesis tests. Prior to the analyses, an evaluation of the quality of the raw data obtained was carried out in two steps. In the first step, a principal components analysis (PCA) was carried out with a VARIMAX rotation, an extraction of the theoretical factors reported in the respective psychometric validations, and a selection of factor loadings greater than .40. In all cases, a satisfactory factorial structure (matching the theoretical one) was obtained. In the second step, a reliability analysis was performed using Cronbach's alpha internal consistency index for all scales and subscales. As reported in the instruments section, internal consistency indices obtained were in the range of Cronbach’ α = .77 and Cronbach’ α = .96, thus very satisfactory and comparable to those of the original validations of the respective instruments. The empirical structure of all the measurement scales was identical to the theoretical, except for the friendship subscale of the POMS, from which one of the items was eliminated due to a significant reduction in reliability; this subscale values have been rescaled to the original range [0-20] for a better interpretation of the results.
Once the data quality tests had been carried out, the scales’ scores were calculated, as an identification of a possible confounding variable, the homogeneity of the two control groups was checked regarding the level of intensity of the hiking carried out. An analysis with a general linear model of mixed design revealed the absence of differences between both groups in the pre-post evolution for state-anxiety, positive affect, negative affect, the six POMS subscales and body mindfulness. Significant differences were found between both groups in mindfulness of mind (Wilk' λ(1,27) = 4.55; p = .04; η2 = .14) and in the general mindfulness scale (Wilk' λ(1,27) = 4.73; p = .04; η2 = .15), so that the activity intensity variable was generated, which was included as a covariate in the following statistical analyses of these two variables.