Associations of bullying perpetration and peer victimization subtypes with preadolescent's suicidality, non-suicidal self-injury, neurocognition, and brain development

Abstract

Background: Although both peer victimization and bullying perpetration negatively impact preadolescents' development, the underlying neurobiological mechanism of this adverse relationship remains unclear. Besides, the specific psycho-cognitive patterns of different bullying subtypes also need further exploration, warranting large-scale studies on both general bullying and specific bullying subtypes.

Methods: We adopted a retrospective methodology by utilizing the data from the Adolescent Brain and Cognitive Development^SM Study (ABCD Study®) cohort collected between July 2018 and January 2021. Participants were preadolescents aged from 10 to 13 years. The main purpose of our study is to examine the associations of general and specific peer victimization/bullying perpetration with preadolescents' (1) suicidality and non-suicidal self-injury; (2) executive function and memory, including attention inhibition, processing speed, emotion working memory, and episodic memory; (3) brain structure abnormalities; and (4) brain network disturbances. Age, sex, race/ethnicity, body mass index (BMI), socioeconomic status (SES), and data acquisition site were included as covariates.

Results: A total of 5819 participants aged from 10 to 13 years were included in this study. Higher risks of suicide ideation, suicide attempt, and non-suicidal self-injury were found to be associated with both bullying perpetration/peer victimization and their subtypes (i.e., overt, relational, and reputational). Meanwhile, poor episodic memory was shown to be associated with general victimization. As for perpetration, across all four tasks, significant positive associations of relational perpetration with executive function and episodic memory consistently manifested, yet opposite patterns were shown in overt perpetration. Notably, distinct psycho-cognitive patterns were shown among different subtypes. Additionally, victimization was associated with structural brain abnormalities in the bilateral paracentral and posterior cingulate cortex. Furthermore, victimization was associated with brain network disturbances between default mode network and dorsal attention network, between default mode network and fronto-parietal network, and ventral attention network related connectivities, including default mode network, dorsal attention network, cingulo-opercular network, cingulo-parietal network, and sensorimotor hand network. Perpetration was also associated with brain network disturbances between the attention network and the sensorimotor hand network.

Conclusions: Our findings offered new evidence for the literature landscape by emphasizing the associations of bullying experiences with preadolescents' clinical characteristics and cognitive functions, while distinctive psycho-cognitive patterns were shown among different subtypes. Additionally, there is evidence that these associations are related to neurocognitive brain networks involved in attention control and episodic retrieval. Given our findings, future interventions targeting ameliorating the deleterious effect of bullying experiences on preadolescents should consider their subtypes and utilize an ecosystemic approach involving all responsible parties.

Keywords: Brain network; Bullying; NSSI; Subtype; Suicide.

Fig. 1

Associations of bullying subtypes with suicidality/NSSI and neurocognition in preadolescents. Adjusted for age, sex, race/ethnicity, site, BMI, marital status, income, and parental highest education. (1) A–C Odds radio was reported for binary suicidality/NSSI variables; (2) D–G coefficient was reported for continuous cognitive variables. vic, victimization; perp, perpetration

Fig. 2

Association of brain structure with victimization. After adjusting for age, sex, race/ethnicity, site, handedness, BMI, SES, and ICV, victimization was found to be positively associated with total cortical area of bilateral posterior dorsal part of the cingulate gyrus, while negatively associated with mean cortical thickness of bilateral paracentral lobule and sulcus

Fig. 3

Associations of resting-state functional connectivity with victimization and perpetration. A. After adjusting for age, sex, race/ethnicity, site, handedness, and mean motion, significant associations between victimization and DMN-FPN, DMN-DAN, VAN-DMN, VAN-DAN, VAN-CPN, VAN-CON, and VAN-SMH were found. B. As for perpetration, significant associations were found in SMH-CON, SMH-VAN, SMH-CPN (DMN, default mode network; FPN, fronto-parietal network; SMH, sensorimotor hand network; VAN, ventral attention network; DAN, dorsal attention network; CPN, cingulo-parietal network; CON, cingulo-opercular network)