Participant recruitment. A convenience sample of 25 male and 25 female participants was recruited through word of mouth. Many were either students or working professionals who had completed their studies. Demographic information was collected (age, gender, education level), and inclusion and exclusion criteria were reviewed during a phone screening. Inclusion criteria were age (18-49), very good oral and written German language skills, and physical and mental health. Exclusion criteria were acute or chronic physical illness (e.g. incontinence, cold symptoms), acute or chronic psychiatric or neurological disorders (e.g. depression, epilepsy, ADHD), medication that impairs physical or mental functioning and the autonomic nervous system, addiction disorders, open or bleeding wounds, as well as contraindications due to a medically diagnosed illness, pregnancy, and menstruation at the time of the floating session. The participants' prior experiences with floating and other contemplative practices were also recorded. The exclusion criteria were asked again using a checklist before each REST session.
A total of 53 participants were initially recruited. Two subjects reported falling asleep during the Bed-REST condition. In the Floatation-REST condition, one subject did not turn off the light as she felt uncomfortable in the dark. These three subjects were therefore excluded. The participants were randomized into either the experimental condition (60 minutes of Floatation-REST) or the control condition (60 minutes of Bed-REST) in the two subsequent sessions following the acclimatization session in both conditions. The order was balanced across all three sessions to control for sequence effects (the order within the acclimatization session, and the order of the two measurement conditions). The time interval between sessions 1 and 2 as well as between sessions 2 and 3 was typically one week.
The study was pre-registered at the University Clinic of Freiburg (https://www.uniklinik-freiburg.de/zks.html; FRKS004085). The study was approved by the Ethics Committee of the University Freiburg. All participants were informed about the confidentiality and anonymity of the data in a participant information sheet and gave written informed consent. Participation in the study was voluntary, and a session could be terminated at any time. Each participant received a compensation of ten euros per session. The study was financed by a Hanns Seidel Foundation scholarship for Helena Hruby from funds of the Federal Ministry of Education and Research (BMBF), Germany.
Trait measures
Absorption (TAS): The German version39 of the TAS (Tellegen Absorption Scale40) consists of 34 items with answer categories on a 5-point Likert scale (0-4) ranging from not applicable/slightly/partly/mostly/absolutely applicable. Example items include: "I like to watch how clouds change their shape" (item 6), "My thoughts often appear as images rather than words" (item 22), and "I think that different smells have different colors" (item 33). Absorption is defined as ‘‘…a disposition for having episodes of ‘total’ attention that fully engage one’s representational (i.e., perceptual, enactive, imaginative, and ideational) resources.”40. The internal consistency value is a Cronbach's alpha of 0.8939.
Perceived stress (PSQ): The Perceived Stress Questionnaire (PSQ) by Levenstein et al.41 measures retrospective individual stress perception during the preceding two years. The questionnaire was validated on a German sample42 and comprises 20 items that are assigned to seven scales in the German version. Participants can indicate whether the statements in the items apply "almost never", "sometimes", "often", or "mostly" on a 4-point Likert scale of 1-4. The internal consistency value is a Cronbach's alpha of 0.8642.
State measures
Emotional reaction (SAM, pre-post-measurements): The Self-Assessment-Manikin Scales (SAM43) measures both (1) positive and negative valence and (2) the current arousal level with five figures, the Manikins, by answering the two questions: "How positive/negative do you feel right now?" and "How aroused/calm do you feel right now?".
State anxiety (STAI, pre-post-measurements): The German short version of the State-Trait-Anxiety Inventory (STAI) by Spielberger44 contains ten items measuring subjectively experienced current anxiety. Participants answer how well the following descriptions of emotions apply to themselves in the present moment: "How much do the following descriptions of emotions apply to you right now?" on an 8-point Likert scale. An overall score with a minimum of 10 and a maximum of 80 points is calculated. The internal consistency of the State-Anxiety Scale has a Cronbach’s alpha of 0.9045.
Tension and stress level (VAS, pre-post-measurements): Participants rated their individual relaxation and stress levels on a 100-mm visual analog scale (VAS) by making a vertical line. The relaxation question asked, "How relaxed do you feel right now?" (endpoints: "extremely relaxed" (0), "extremely tense (100)") and the stress question asked, "How high do you estimate your current stress level?" (endpoints: "extremely low", "extremely high").
Fatigue (VAS, pre-post measurement): Level of fatigue before the intervention was indicated on a 100-mm VAS that asked, "How sleepy do you feel?" (endpoints: "extremely awake: 0"; "extremely sleepy: 100"). Additionally, we asked participants after the session how sleepy they felt during the REST conditions: "How sleepy did you feel?" with the same answer endpoints of “awake” to “extremely sleepy” on a VAS.
Mind-wandering (VAS, post-measurement): Mind wandering is assessed after the intervention using a 100-mm VAS by asking the question, "How many thoughts are currently going through your head?" (endpoints: "none at all", "extremely many").
Subjective time perception (STSS, post-measurement): The Inventory on Subjective Time, Self, Space (STSS46) assesses subjective time perception by using mostly 100-mm VAS. The scales consist of the following questions, which are answered as spontaneously as possible: "How often did you think about time during REST?" (Thinking about time: endpoints: "not at all" to "extremely often"); "How fast did time seem to pass for you during REST?" (Subjective passage of time: endpoints: "extremely slow" to "extremely fast"). "Did you lose your sense of time during REST?" (Time distortion: endpoints: "not at all" to "extremely strong”). The question "How long did the 60-minute REST session subjectively feel?" assesses the subjectively felt duration of the exposure to the intervention. Subjects indicated their sense of duration by writing the respective number of minutes into a field.
Body boundaries (PBBS, post-measurement): Changes in perceived body boundaries during the intervention were assessed using the Perceived Body Boundaries Scale (PBBS47). Participants can choose between the drawings of seven human bodies with different levels of visible body boundaries. 1 represents a weak body boundary, and 7 represents a strong body boundary.
General change in states of consciousness (PCI, post-measurement): The German version of the Phenomenology of Consciousness Inventory (PCI)48 consists of 53 items that can be answered on a 7-point Likert scale ranging from 0 to 6. The following 12 dimensions are given: self-awareness (D15), altered state (D16), internal dialogue (D17), rationality (D18), voluntary control (D19), memory (D20), arousal (D21), positive (D22) and negative affect (D23), altered experience (D24), imagination (D25), and attention (D26). The PCI is a valid measurement instrument for the assessment of subjective phenomenological experiences in different contexts (e.g., hypnosis, relaxation, meditation, and Ganzfeld)1.
Floatation-REST. The float tank, a “Cabin for two” built by Floataway (Norfolk, United Kingdom), is located in the Prana Health Practice in Freiburg, Germany (see Fig. 1a). The room contains the float tank (width 180 cm, length 237 cm, height 225 cm) and a shower. The pool was filled with 1100 liters of water mixed with 550 kg of Epsom salt (magnesium sulfate) creating a water height level of 27 cm. The water and air temperature was matched to human skin temperature (~35°C) and the air temperature at the rim of the pool was around 33.4°C, slightly lower than the water temperature based on the relative humidity in the air (84%). The water pH values ranged from slightly acidic to neutral between 6.8 and 7.2 (total alkalinity: 150 ppm). The density of the saltwater was maintained at a specific gravity of around 1.25 for all float sessions, allowing for effortless floating with about 1/3 of the body above the water surface. The float tank was placed on a bed of butyl rubber floor springs, effectively isolating the pool from the building and preventing all structure-borne noises from entering the water. The entire floating room had non-slip flooring and handholds for getting in and out of the tank. The room was illuminated while the participant changed and showered. The external lighting in the room was switched off by the experimenter only after the subject had entered the pool. A blue LED light inside the float cabin could be switched on and off by the participant using an air switch, and an alarm button could be pressed to trigger an audible alarm signal in case of an emergency. The exact instruction the experimenter gave was: You will now float for 60 minutes. Please lie on your back in the middle of the pool and do nothing else. Please try not to fall asleep during this time, as measurements are taken after the resting phase. After the 60 minutes have elapsed, a gong will sound which signals the end of the session and the light will turn on again.
Bed-REST. In the control condition, participants were situated in a separate, quiet room where they lied supine on a 90 x 220 cm waterbed (Classic Wasserbett SOLO) built by Aqua Comfort GmbH (Paderborn, Germany) while wearing clothing (see Fig. 2b). The waterbed has a calming level of 90%, meaning that the post-swing time of water movement when you lie down in the bed is maximum 2 seconds. The surface temperature of the imitation leather cover was held constant across subjects at 28° C, which is the manufacturer's recommended comfort temperature. When the lights were turned off by the experimenter, the room was completely dark and a low salt content could be perceived in the air due to salt stones that were placed around the perimeter of the room. The average room temperature was ~23°C. Participants were instructed to do nothing and lie on their backs for 60 minutes without falling asleep. The exact instruction the experimenter gave was: You will now rest for 60 minutes. Please lie on your back and do nothing else. Please try not to fall asleep, as measurements will be taken after the resting phase. After the 60 minutes have elapsed, I will turn on the light and open the door.
Procedures. At the beginning of the first session (acclimatization session), participants received information about the experiment and signed an informed consent. Then they filled out questionnaires related to their trait characteristics (TAS, PSQ). The subjects were exposed to 30 minutes of Floatation-REST and Bed-REST (balanced succession across participants) in randomized order. Comprehensive written and verbal information instructions were given regarding the exact procedure and all relevant safety aspects before each REST session.
Participants were requested to shower and wash their hair with the provided shampoo before entering the pool naked in the Floatation-REST condition. They were reminded to remove or store all valuable items, such as jewelry, cell phone, hair clips, makeup, and clothing, and to plug their ears with earplugs while they were still dry to provide better acoustic isolation and protect from salt crystals becoming lodged in the ear canal (of note, the same earplugs were also used during the Bed-REST condition). Urgent needs, such as using the restroom, were taken care of before the REST conditions. The room was lit when the participants entered the float cabin; the ceiling light was turned off by the experimenter once the participant had entered the tank. The participants stepped into the pool using the handles and laid down on their backs in the middle of the pool. The participants were instructed to turn off the pool light when they were in their preferred resting position. The participant indicated through an intercom when he/she turned off the light so that the experimenter could accurately time the floatation session. The end of the session was indicated when the pool light went on and a gong sounded. The participants left the pool to shower and dress. To indicate the end of the 60 minutes session in the Bed-REST condition the experimenter opened the door of the salt room and switched on the light. The post-condition questionnaires were completed in a nearby room on a table (of note, the pre-condition questionnaires were completed in the same room). Participants were briefly asked if any adverse experiences had occurred before they filled out the questionnaires.
The two 60-minute REST sessions took place after the acclimatization session. Either the Floatation-REST condition was completed first followed by the Bed-REST condition or vice versa. In both cases, participants filled out state-related questionnaires during the pre-measurement, followed by the 60-minute REST intervention, and then the completion of the post-measurement questionnaires. Overall, all three sessions lasted a maximum of two hours each.
Statistical analysis. The main intention of this study was to determine the differences in experience between Floatation-REST and Bed-REST. Statistical calculations were conducted using SPSS (Version 24). After testing for normal distribution (Kolmogorov-Smirnov and Shapiro Wilk), it was found that 28 according to the Kolmogorov-Smirnov-test and 30 according to the Shapiro Wilk test out of a total of 62 variables were not normally distributed (2 trait variables, 30 variables in the Floatation-REST condition, and 30 variables in the Bed-REST condition). Therefore, non-parametric testing was applied, and the Wilcoxon test for dependent samples was used to calculate the difference between the Floatation-REST and Bed-REST conditions, as well as the post- vs. pre-intervention effects. Median and interquartile ranges are indicated. The difference score between the post- (t2) and pre-measurement (t1) was used in the calculations (DIFF: t2-t1) for variables collected before and after the REST interventions.
The association between indices of ASC and emotional state was calculated using Spearman correlations. To determine whether the influence of the two conditions on state variables was mediated by ASC (X = state variables, Y = conditions, M = ASC; Fig. 1), we performed a mediation analysis with the PROCESS Macro (Version 3.4.1) by Hayes49. The influence of trait-related variables on state-related variables was determined using regression analysis with bootstrapping procedures that allows for the use of data that is not normally distributed50. False Discovery Rate (FDR51) calculations were used to correct the significance level for multiple comparisons.