A randomized controlled trial on the digital socio-emotional competence training Zirkus Empathico for preschoolers

Abstract

In this randomized controlled trial (RCT), the digital socio-emotional competence training Zirkus Empathico was tested in 74 Central European children (5.1 (0.9) years; 34 females) within a longitudinal design (three time points: T1 = pre-training; T2 = immediately following 6-week training, T3 = 3-month follow-up). The pre-registered primary outcome was empathy, secondary outcomes included emotion recognition, prosocial behavior, and behavioral problem reduction; furthermore, children's neural sensitivity to facial expressions quantified with event-related potentials. Compared to controls (N = 38), Zirkus Empathico participants (N = 36) showed increases in empathy (d = 0.28 [-0.17, 0.76]), emotion recognition (d = 0.57 [0.01, 1.06]), prosocial behavior (d = 0.51 [0.05, 0.99]) and reduced behavioral problems (d = 0.54 [0.08, 1.03]). They also showed larger P3 amplitudes to happy vs. angry and neutral facial expressions post-training. Thus, Zirkus Empathico may be a promising digital training for social competence in preschoolers.

Fig. 1. Parent and child ratings of socio-emotional competence (SEC) reports.

Orange: Zirkus Empathico group. Gray: Control training. Error bars indicate standard errors (SE); center line represents median; upper and lower box limits represent quartile 1 and quartile 3; whiskers represent minimum and maximum values; d z-standardized represents the standardized change scores (difference between post- and pre-training values). Note: Standardization was achieved by subtracting each value from the variables mean and dividing them by the variable’s standard deviation (x – x̄)/SD). Standardization was only carried out for visualization purposes; statistical analyses were carried out wih the raw d values. GEM Griffith Empathy Measure, EMK Inventory to survey of emotional competences for 3- to 6-year-olds, EM Empathy, ER Emotion recognition, SDQ Strength and Difficulties Questionnaire, PB Prosocial behavior, BP Behavioral problems. P = parent rating; CH = child assessment. Asterisk (*) indicates significant difference with p-value < 0.05 calculated within separate ANCOVAs for each outcome.

Fig. 2. Distribution information on parent and child ratings of socio-emotional competence (SEC) reports.

Measures were separated in parent and child ratings of a empathy, b emotion recognition and c prosocial behavior. Orange: Zirkus Empathico group. Gray: Control training. Small boxplots: Error bars indicate standard errors (SE); center line represents median; upper and lower box limits represent quartile 1 and quartile 3. Small line plots: Error bars indicate standard error (SE); center line represents mean. GEM Griffith Empathy Measure, EMK Inventory to survey of emotional competences for 3- to 6-year-olds, EM Empathy, ER Emotion recognition, SDQ Strength and Difficulties Questionnaire, PB Prosocial behavior, BP Behavioral problems. P = parent rating, CH = child assessment.

Fig. 3. ERP trajectories and morphologies for the control and Zirkus Empathico group.

Averaged P1, N170, and P3 waveforms for happy (green), angry (blue), and neutral (gray) facial expressions (ROI P1/P3 electrodes: PO3, PO7, PO9, O1, O2, Oz, PO4, PO8, PO10; ROI N170 electrodes: P7, TP7, CP5 and P8, TP8, CP6. Shadowed areas indicate the time windows used to identify participants’ individual peaks and mean amplitudes. Topographies summaries the averaged P3 activity (300–500 ms) displaying amplitude difference scores for facial expression contrasts (from the left): Controls happy–neutral, angry–neutral, and happy-angry; Zirkus Empathico group happy–neutral, angry–neutral, and happy-angry.

Fig. 4. Flow Diagram of the trial.

ITT intention-to-treat, PP per protocol, CC complete cases.

Fig. 5. Training elements of Zirkus Empathico.

Left: Exemplary module sequence of identifying other people’s facial expressions. Middle: Overview of different training modules targeting own and others’ emotion recognition from situational cues; empathy and prosocial actions. Right: Interactive manikin to visualize own and other’s emotional states with sliders for valence and arousal. Consent to publish all images was obtained from the individuals who are displayed here as part of the Zirkus Empathico training.

Fig. 6. Electrode montage with channel locations used as regions of interest (ROIs).

Dark gray: Channels used for the P1 and P3 component. Blue: Channels used for the left and right ROI of the N170 component.