E4 wristband

Nocturnal sleep was assessed using E4 wristbands (Empatica, U.S.) worn by the participants, which were research devices for real-time physiological monitoring [27 (link)]. Total sleep time and efficiency were evaluated by applying the Cole-Kripke algorithm after extracting the values of movements during sleep using an accelerometer sensor [28 (link)].

The changes in the child’s behaviour and anxiety levels. The primary outcome measures will be assessed by

Physiologically measurable parameters by using a medical grade wearable device (E4^® wrist bands—Empatica Inc., 1 Broadway, Cambridge, MA 02142, United States—ISO 13485 Cert. No. 9124.EPTC). These parameters are

Electrodermal activity (EDA) also known as Galvanic skin response (GSR),

Blood volume pulse (BVP),

Heart rate (HR),

Surface skin temperature.

Dental anxiety scores at baseline before the (acclimatisation session) and after treatment (second session) using the Modified Child Dental Anxiety Scale-face version (MCDASf).¹¹

Children’s behaviour score will be recorded using the Frankl Behaviour Rating Scale (FBRS).¹²

Prior to the experiment, participants were asked to provide their informed consent and complete a questionnaire that was designed to collect demographic information. Subsequently, the researchers escorted participants to the smart home environment, where they were oriented to the locations of labeled sensors, appliances, and furniture. Additionally, participants were requested to wear Empatica E4 wristbands to generate benchmark heart rate data.
During the experiment, facilitators observed participants' activity performance through cameras in the room next to the apartment unit while delivering activity instructions to participants via a phone. Activity instructions were delivered in sequence by phases and scenarios to participant and no further instructions were provided unless participants asked for specific assistance. One minute transition time was allotted between each scenario, in order to reset all activated sensors. An observer documented activity information, including start and end timestamps, using an annotation sheet. Sensor readings were collected, formatted, and stored in the cloud database via our smart home data ecosystem. Only one participant was tested at a time, and each experiment session lasted ~3 h.

The Empatica E4 sampled EDA at 4 Hz, and skin conductance (SC) was measured in micro-Siemens (μS). Due to variation in the length of EDA readings across participants, analysis windows were standardized by removing the first minute of data. The subsequent 8 min were used in the analysis.
Continuous Decomposition Analysis (CDA) was performed on EDA data in Ledalab (Benedek and Kaernbach, 2010 (link)) to decompose SC data into tonic (skin conductance level; SCL) and phasic (skin conductance response; SCR) components (Dawson et al., 2007 ). Event markers were imported from Empatica E4 wristbands to determine the central 8 min SC analysis window for each exposure session. For the purposes of this study, two measures representing tonic and phasic activity were used; the mean tonic activity within a treatment window, and the number of significant SCR events, respectively.

We perform multiple types of analyses to evaluate our hypothesis about the effectiveness of safe actuation in regulating brain states. These include physiological signal processing and evaluating cognitive performance levels. To evaluate cognitive performance state, we follow the methodology presented in^{101 (link),102 (link)} to model latent performance state⁶ . Using a systematic approach, we relate the cognitive performance state to the subjects’ correct/incorrect responses and reaction times in the n-back experiments. We next estimate the latent performance state. Moreover, we employ physiological measurements and use a state-space representation to model the internal arousal state. To estimate the internal arousal state, we analyze the skin conductance signal collected via Empatica E4 wristbands. By applying a deconvolution algorithm and inferring underlying neural impulses, we establish a marked point process Bayesian filter to estimate hidden cognitive arousal state^{83 (link),91 (link)}. For the purpose of statistical analysis, we conducted a one-way analysis of variance (ANOVA) to test for differences in behavioral data, physiological signals, and all derived metrics.

We used two wearable Empatica E4 wristbands and a portable muse headband for EEG recording. Using the Empatica E4 wristbands, we collected electrodermal activity (EDA) (or skin conductance) that tracks the changes in skin conductivity using two metal electrodes, blood volume pulse (BVP), from which heart rate variability can be measured, using a photoplethysmography sensor, motion-based activity using a 3-axis accelerometer sensor, and skin temperature using infrared thermopile. Using the 2016 edition muse headband, we collected brain activity using four EEG sensors.