Developmental consequences of the digitally-mediated landscape: implications for empathic processing at the neural and behavioural levels-of-analysis during the early adolescent pubertal transition
Item statusRestricted Access
Embargo end date12/07/2024
Banga, Claire Ann
BACKGROUND: Empathy is fundamental to the adaptive development of emotional and prosocial functioning in childhood and adolescence (Eisenberg, 2018); whereas, a number of forms of psychopathology and maladaptive developmental outcomes are associated with marked deficits in empathic processing (Farrell & Vaillancourt, 2021). Cross temporal meta-analytic evidence indicates a concerning trend of average levels of empathy decreasing in younger generational cohorts (Konrath et al., 2011), while concurrently, mental health difficulties such as depression, anxiety, suicidality, and loneliness are more prevalent in younger generations (Twenge et al., 2018). These generational patterns map onto the historical invention, subsequent adoption, and eventual market saturation of digital technologies including the internet, smartphones, and social media. Such trends have given rise to the rewiring hypothesis which contends that the substantial amounts of time spent engaged with digital technologies may be resulting in neuroplastic changes in developing brains that may be impeding adaptive emotional functioning (Greenfield, 2014; Montag & Diefenbach, 2018). AIMS: The current project addresses the primary research question (RQ) of whether the vast range of digital technology-based activities that young people engage in on a daily basis – termed ‘screen-time’ (ST) – may be impacting their empathic development. The theoretical framework of the Perception-Action Model of Empathy (Preston & de Waal, 2002) was employed as the basis for operationalizing the study of empathy. The first set of RQs ask whether higher levels of ST predict deficits in cognitive (RQ1.1) and affective (RQ1.2) empathy. The second set of RQs evaluate the predictive validity of subjective versus objective measures of ST (RQ2.1) and the statistical convergence between behavioural, neurophysiological, and global-report measures of empathy (RQ2.2). SAMPLE: The pubertal transition is a particularly important period in neural development as substantive changes in neurobiological organization occur that create a vulnerability to maladaptive affective outcomes (Guyer et al., 2016). Thus, the current study drew from a sample of typically developing preadolescent children (11 to 14 years of age) which marks an age group both beginning and in the midst of puberty. The project consisted of a design of three successive studies which drew from the same sample recruited from Scottish secondary schools. METHOD: Study 1 (n = 263) consisted of a smartphone-based study which combined ecological momentary assessment (EMA) with the collection of passive smartphone sensing data including objective time estimates of smartphone screen-time (SMT). The study used eMoodie (emoodie.com), an application specifically designed for developmental research, which was developed and validated in the context of this project. The design allowed for the momentary assessment of ST by young people, in their everyday environments, in a way that could help to operationally define and measure increasingly complex patterns of use such as media multitasking (MMT). Using multi-level modelling, comparisons between EMA surveys, a variety of global measures of SMT, MMT, and smartphone addiction, were made relative to objective data on sensed SMT. It was hypothesized that subjective measures would show poor convergent validity with objective data (H2.1). Study 2 (n = 165) was a behavioural study of empathy which employed a video-based empathic accuracy task as a performance-based measure of cognitive empathy (i.e., perspective taking; Zaki & Ochsner, 2011) with a stimulus set which was also developed and validated in the context of this project. It was hypothesized that empathic accuracy performance would be inversely related to ST (H1.1), and it would show poor convergence with child self-reported empathy scores (H2.2). Study 3 (n = 39) was a laboratory-based neurophysiological study of pain empathy using event-related potentials (ERPs) based on a well-validated task which is thought to directly measure affective empathy based on the temporal resolution of the processes being measured (Coll, 2018; Fan & Han, 2008). Importantly, participants were grouped based on high versus low ST levels to compare differences in empathic processing at the neural level-of-analysis. It was hypothesized that there would be deficits in pain empathic processing in the high but not the low ST group (H1.2). Finally, by combining the results across the three studies, statistical models evaluating the relationship between ST and empathy were conducted, whilst controlling for the child’s sex and parental attachment style. RESULTS: Findings from Study 1 established the feasibility and acceptability of the use of eMoodie as a research tool with high overall compliance (85%) in the weeklong EMA study (Domoff, Banga, et al., 2021). EMA measures of SMT were the best predictor of sensed SMT, while global estimates bore no statistical relationship to actual time usage (H2.1). Further analyses sought to examine group differences in SMT based on self-reported smartphone addiction which detected a bias in reporting whereby those with normative usage overreported time spent, whereas those with problematic usage underreported SST. The nature of these systematic errors would theoretically lead to a pattern of results in the evidence-base which fail to detect a true effect (i.e., Type II error) which may help to explain inconsistent results in media-effects research (e.g., Ivie et al., 2020). Results of Study 2 and 3 which compare the statistical convergence across the multiple methodologies employed to study empathy in this project, indicated that empathic accuracy task scores and ERP measures tapped different sources of variance. This supports a multidimensional model of empathy comprising both cognitive and affective dimensions. When compared to global questionnaires, no child self-report empathy measures yielded any statistical relationship with these more objective methodologies (H2.1-2.2), with the only predictive variable being parental reports of the child’s perspective-taking abilities. In Study 3, ERP measures showed group differences based on ST consumption in both early and late components of pain empathy whereby the high ST group showed deficits in empathic processing. Therefore, the current findings provide neurophysiological evidence of the rewiring hypothesis (H1.2). Statistical modelling showed that ST predicted empathy across both cognitive and affective dimensions after controlling for important covariates (H1.1-1.2). CONCLUSIONS: Taken together, this project’s findings establish a detrimental effect of ST on the development of empathy both at the behavioural and neural levels-of-analysis in a preadolescent sample. The implications are that high levels of ST may lead to deficits in social skills facilitated by empathy such as prosociality. From a developmental psychopathology perspective, high levels of ST may increase the probability of maladaptive developmental trajectories linked to psychopathology. Moreover, the results of the current project indicate substantial measurement error and poor validity in the dominant single time-point self-report methods used in media-effects and psychological research on empathy.