Face-to-face classroom learning produced greater brain synchronisation in children than a zoom-based online session.

Abstract

The shift to online learning during the pandemic posed significant challenges for children, including reduce attention.1 Reports have also suggested that significantly more is learnt during face-to-face interactions, than via screens, such as language acquisition.2 The ability to learn and react to stimuli can be assessed with electroencephalography (EEG). Zivan et al focused on two groups of preschool children aged 4–5 years who took part in a six-week experiment that compared face-to-face and screen-based storytelling. The screen group showed more slow theta band waves on the EEG and fewer fast beta and gamma band waves than the face-to-face group when they rested after the intervention.3 Those patterns have been associated with attention difficulties and load. We looked at how face-to-face interaction facilitates learning and how can it be measured. This study compared learning outcomes and neurobiological correlates during face-to-face and online learning, using behavioural EEG data collected in a unique manner. Hyperscanning shows the neural synchrony between two or more individuals that results in brain synchronisation matrices. We hypothesised that the face-to-face interaction would produce better learning outcomes, and higher neural synchrony in the alpha and beta band waves, than online learning. We recruited a science teacher and six typically developing adolsecents, five female, with an average age of 11 ± 1.2 years and no developmental or neurological deficits. Data from the students were simultaneously recorded in the two learning conditions while connected to EEG devices (Figure 1). The parents and children provided written consent and assent, respectively, and the children could leave the study at any time. The institutional review board of the Technion, Haifa, Israel approved the study and the school provided permission. The learning sessions focused on two different, but comparable, scientific themes and included 15-min of explicit tutoring and a short video explaining the topic. The teacher and students also conducted a short experiment while they were connected to the EEG device. During the face-to-face condition the children sat in the laboratory, in front of the teacher, so they could see each other. During the online Zoom call, in gallery mode, the participants were all in separate rooms but they could see each other (Figure 1). Five written comprehension questions were administered at the end of each session. Means and standard deviations were calculated for the comprehension levels following each condition. No statistical analysis was possible due to the sample size. The EEG data were processed using multidimensional matrices containing the neural synchrony values between every pair of frontal electrodes associated with joint attention abilities. These were namely AF3, AF4, F4, F3, F8, and F7. The 6X6 synchronisation matrices for the online condition were subtracted from the face-to-face condition for alpha (8–12 Hz) and beta bands (12–30 Hz). They were averaged across time for all subjects and separately for each condition, namely within the children and between the children and teacher. The processing pipeline has previously been described.4 The results demonstrated higher comprehension levels for the face-to-face than online learning condition (X = 93.33 ± 10.32% vs. X = 76.66 ± 36.69%). More electrodes were highly synchronised between the teacher and pupils and between the pupils during the face-to-face versus online conditions in the alpha and beta frequency bands (72.2% vs. 75.1%). Children showed higher comprehension levels during the face-to-face condition, as hypothesised. There were lower standard deviation values and higher brain-to-brain synchronisation between the teacher and the pupils and within the pupils. Dialogue and peer interaction during interactive classroom learning have been reported for various domains, including language acquisition.2 Classroom dialogue can include interactions between small groups of students and whole class discussions orchestrated by the teacher.5 The class design in our study involved discussions in both the face-to-face condition and an online Zoom call. However, our study highlighted individual differences in the ability to be actively engaged in class while learning online, with regard to EEG and behavioural learning outcomes. It has been suggested that face-to-face learning, mediated by joint attention, is more beneficial than online learning. Zivan et al found this when they compared online and face-to-face storytelling.3 Although the information received during face-to-face learning may be cognitively demanding, the learning outcomes are greater. Online learning may also lack the joint attention and synchronisation that are created between class participants during face-to-face learning, due to external or internal disruptions. In our study, the interactive nature of face-to-face learning, with joint attention and eye contact between the participants, resulted in greater synchronisation and outcomes. Providing neuroscientific evidence for the differences in brain synchrony between these two learning conditions can deepen our understanding of the effect they have on children's outcomes. None. None.