There are a variety of studies demonstrating the negative impact of maltreatment experiences in a peer context in childhood and adolescence on psychosocial adjustment and particularly mental health (for a review see McDougall and Vaillancourt, 2015). These experiences, also called peer victimization experiences, include different kinds of maltreatment experiences that occur in interactions with peers, e.g., overt forms like physical or verbal violence, or relational maltreatment experiences associated with rejection or exclusion from a social group (De Los Reyes and Prinstein, 2004; Siegel et al., 2009; Sansen et al., 2015). Thus, peer victimization can be distinguished from child maltreatment, where the violence is perpetrated by adults or caregivers, including forms of emotional, physical, or sexual maltreatment (World Health Organization, 1999). Similar to the negative effects of child maltreatment (for a review see Carr et al., 2020), peer victimization is associated with problems in several areas, such as academic achievement or social adjustment (e.g., Schwartz et al., 2005; Juvonen et al., 2011; Takizawa et al., 2014). In addition to their influence on social and economic outcome variables, experiences of peer victimization seem to significantly increase the risk of experiencing various mental disorders like depression, anxiety disorders, PTSD, or substance abuse in childhood and adulthood (e.g., Stapinski et al., 2014; Hébert et al., 2016; Earnshaw et al., 2017). Moreover, a longitudinal study showed an association between frequent victimization and suicide attempts and suicide for girls later in life (Klomek et al., 2009). Thus, the experience of peer victimization in childhood and adolescence has long-term consequences that have a particular impact on mental health even decades after the exposure. Following on from this, studies indicate that experiences of adverse experiences in childhood and adolescence are related to an altered stress response, or structural and functional brain changes which in turn may have an impact on mental health in adulthood (Brendgen et al., 2017; Aults et al., 2019; Quinlan et al., 2020). In addition to physiological factors, altered attentional processes, or attentional biases, have been discussed in the context of traumatic childhood experiences and psychopathology in later life (Fani et al., 2011; Günther et al., 2015; Kelly et al., 2015; Iffland et al., 2019).

Attentional biases refer to the altered attentional focus on stimuli, are influenced by the valence or relevance of a stimulus, and are shaped by individual factors, such as emotional states or psychopathological symptoms (Koster et al., 2004, 2005; Bar-Haim et al., 2007; Cisler and Koster, 2010; Hankin et al., 2010; Peckham et al., 2010). Since they can influence perception and interpretation, and thus cognition and behavior, attentional biases are therefore considered in theories of the development and maintenance of mental disorders (for a review see Cisler and Koster, 2010). Investigating attentional biases in more detail, three different forms of attentional bias can be distinguished (i.e., facilitated attention, difficulties of disengagement, and attentional avoidance; Koster et al., 2004; Cisler and Koster, 2010). Facilitated attention is reflected in the way that emotional stimuli attract attention and thus attention is shifted to these stimuli more quickly. Difficulties in disengagement refer to the extent to which a stimulus attracts attention. This is accompanied by the difficulty in shifting attention from one stimulus to another stimulus. Attentional avoidance is manifested by the avoidance of shifting attention to potentially threatening stimuli and instead directing it to stimuli that are not threatening (Koster et al., 2004; Cisler and Koster, 2010).

Attentional biases have robustly been shown in individuals with depression or various forms of anxiety disorders (for a detailed overview see Bar-Haim et al., 2007; Cisler and Koster, 2010; Peckham et al., 2010). Furthermore, attentional biases have been repeatedly reported in victims of abuse and neglect with and without psychopathology (e.g., Pine et al., 2005; Fani et al., 2011; Romens and Pollak, 2012; Günther et al., 2015; Kelly et al., 2015). Previous research has suggested that experiences of maltreatment appear to influence attentional processes mainly in response to threatening stimuli (Gibb et al., 2009; Iffland and Neuner, 2020). In these studies, abused children had a higher tendency to attend to threatening stimuli, had problems shifting their attention away from cues of anger, and were faster in recognizing anger with less information (for a detailed overview see Jaffee, 2017). For example, using an emotional Stroop task and a dot-probe task, Iffland and Neuner (2022) found that emotional abuse was a significant predictor of attentional biases toward negatively associated neutral faces. In the dot-probe task, this was reflected in facilitated attention to these facial stimuli (Iffland and Neuner, 2022). Other studies found attention avoidance of threatening stimuli associated with child maltreatment (Pine et al., 2005; Kelly et al., 2015). With respect to studies of attentional biases in depression, there is evidence that altered attention allocation in the context of maltreatment does not only refer to a potential threat (Romens and Pollak, 2012; Günther et al., 2015). Günther et al. (2015) examined the connection between child maltreatment experiences and attentional processes using a dot-probe task in adults with a diagnosis of major depression. The authors found that experiences of child maltreatment were associated with altered attention allocation to sad facial expressions. This result was independent of symptom severity. Sustained attention toward sad faces was shown to be a stronger mood-congruent bias in depressed individuals with a history of child maltreatment (Günther et al., 2015). However, there have also been studies that found no evidence of attentional biases to emotional stimuli in general or to negative stimuli when analyzing reaction times within the dot-probe task in maltreated individuals (Fani et al., 2011; Hoepfel et al., 2022). Thus, although the results are not entirely conclusive, there is substantial evidence of attentional bias in the context of child maltreatment experiences.

Similar to child maltreatment perpetrated by adults or caregivers, relational peer victimization was associated with altered attention processes. Particularly, peer abused children showed less interference when confronted with victim-related words in an emotional Stroop task (Rosen et al., 2007). In another study examining adult psychiatric patients and healthy controls, Iffland et al. (2019) reported attentional biases in individuals who experienced relational peer victimization. Independent of the presence of mental illness, peer victimized individuals showed attentional avoidance in response to emotional words. Notably, avoidance was found not only in response to threatening stimuli but to emotional stimuli in general. In addition, Iffland and Neuner (2022) identified attentional biases for neutral faces previously conditioned with negative stimuli in individuals with experiences of relational peer victimization. Specifically, retrospective reports of relational peer victimization made an incremental contribution to the prediction of attentional biases beyond child maltreatment. Yet, they found no evidence of attentional avoidance, but rather an attentional bias toward threatening stimuli (Iffland and Neuner, 2022). Hence, regarding the impact of experiences of peer victimization on attention allocation, findings are not entirely conclusive as they have differed concerning the type of attentional biases and the valence of the stimuli for which biases occur.

To extend existing knowledge regarding the association between relational peer victimization and attentional processes we conducted an online facial dot-probe task (MacLeod et al., 1986) with emotional faces. The present study sought to investigate the influence of relational peer victimization experiences in childhood and adolescence on attentional processes and biases when using stimuli that are relevant in social interactions (emotional facial expressions). Attentional processes were examined in relation to positive and negative stimuli, with negative stimuli distinguished between negative, non-threatening stimuli (sad faces) and potentially threatening stimuli associated with victimization experiences (angry and disgusted faces). Social threat and exclusion are communicated not only through angry facial expressions but also through disgusted facial expressions, as there is an interpersonal aspect of disgust that is elicited by undesirable individuals to protect the social order (Rozin et al., 2008; Tybur et al., 2013). This more nuanced stimulus selection including potentially threatening and non-threatening negative emotions was used to provide a more accurate analysis of attentional processes that extends the findings from previous dot-probe studies in peer-victimized adults (Iffland et al., 2019; Iffland and Neuner, 2022). Drawing from the findings of previous research (Iffland and Neuner, 2022), relational peer victimization was expected to make a significant contribution to the prediction of attentional biases beyond the influence of child maltreatment experiences. We assumed that this would be particularly the case for emotions that are relevant in the context of peer victimization (i.e., anger, disgust). Based on previous results (Iffland and Neuner, 2022) we expected attentional biases to be evident in heightened attention to potentially threatening stimuli.

A detailed description of the attentional bias index scores for the different emotions, the mean values, standard deviations, and one-sample t-tests of the absolute index scores for the presentation of one emotion each are shown in Table 3. The results of the t-tests showed that the orienting scores for disgusted and happy faces differed significantly from zero and were negative, indicating attentional avoidance of happy and disgusted facial expressions.

The bivariate Pearson correlation coefficients between maltreatment experiences and the different index scores for each trial type can be found in Table 4. The correlations between the FBS sum score and the different index scores for each emotion were not significant (all FDR corrected p's > 0.05). The analyses showed a positive correlation between the emotional abuse score and the disengaging score for sad-neutral trials as well as a positive correlation between the emotional neglect score and the disengaging score for sad-neutral trials.

Given the ambiguous findings on attentional biases in the context of peer victimization, the present work provided new insights into the relationship between relational peer victimization and attentional biases beyond the influence of child maltreatment experiences. In this context, the present study was designed to provide differentiated accounts of attentional biases in the context of maltreatment and peer victimization experiences, thus extending previous research. Consistent with our hypothesis we found altered attentional processes in individuals reporting higher levels of victimization experiences in the present sample. However, this influence was found in sad faces and not, as previously hypothesized, in emotions that were expected to be relevant as threatening stimuli in the context of peer victimization. Furthermore, altered attention processes were found in individuals reporting experiences of emotional maltreatment when confronted with angry facial expressions.

In the present study, the results indicated evidence for facilitated attention to sad facial expressions in individuals with higher levels of relational peer victimization experiences beyond the influence of experiences of child maltreatment. This effect was seen even when controlling for symptoms of psychopathology. These findings were consistent with the results of Günther et al. (2015), who also found facilitated attentional orienting to sad faces in depressed individuals with experiences of child maltreatment when controlling for depressive symptoms. Previous research suggested that attentional biases not only manifest in biased attention regarding potentially threatening stimuli but could also be influenced by a person's mood or are mood-congruent (Koster et al., 2005; Hankin et al., 2010; Romens and Pollak, 2012; Günther et al., 2015). Similarly, previous research revealed the existence of attentional biases in currently depressed or at-risk children and adolescents (e.g., Joormann et al., 2007; Hankin et al., 2010). Accordingly, the presentation of faces in the current study may have triggered negative emotions associated with social interactions, which may have facilitated processing of sad stimuli. In line with this argument, previous research has emphasized the relevance of sadness in the context of victimization and maltreatment. Victims of bullying tend to be insecure and fearful and they are more likely to have a negative view of themselves and rate themselves as stupid or flawed (Olweus, 1994). In a study by Mahady Wilton et al. (2000) the authors observed the behavior of elementary school children and found signs of sadness significantly more often in victims of bullying than in perpetrators, which could be related to the perceived failures in achieving one's goals in social situations. The authors note that sadness signals to the perpetrator that his goal of causing suffering is being met and thus becomes reinforcing, increasing the likelihood of becoming a victim (Mahady Wilton et al., 2000). In addition, previous studies found increased self-reported sadness among victims of bullying (Camodeca and Goossens, 2005; Glew et al., 2005). However, heightened attention for sadness cues is not exclusively associated with experiences of peer victimization. Romens and Pollak (2012) used a mood induction before a dot-probe task with depression-relevant cues and found that children with experiences of physical abuse showed heightened attention for these cues after the induction of sadness. This result is in line with various other studies showing altered responses on a behavioral and neural level in studies using sad faces in participants with various forms of traumatic childhood experiences (for a review see Saarinen et al., 2021). Hence, findings of an altered reaction to sad facial expressions in the wake of peer victimization in the present study may also apply to adverse childhood experiences in general. In conjunction with evidence of mood-congruent bias in depression and at-risk depression (e.g., Joormann et al., 2007; Hankin et al., 2010), the present findings may be indicative of biased information processing in peer victimized individuals that may be relevant in putting individuals at risk for the development of psychopathology. Following Rosen et al. (2007), victims may implicitly associate themselves with victimization which decisively influences cognitions, behavior, and emotions in future social situations.

There was no influence of peer victimization on participant scores when angry or disgusted faces were presented. Further, there was no significant influence of peer victimization on reaction times for happy faces although the results of the one sample t-tests suggest that participants generally showed significant avoidance of happy and disgusted faces. Therefore, the findings for the overall sample are not reflected in the analyses for peer victimization. The present results contradict the findings of Iffland et al. (2019) and Iffland and Neuner (2022) who found a significant contribution of peer victimization experiences for attentional biases for potentially threatening stimuli and positive emotional stimuli in their studies. Using a social conditioning task or social evaluative words in these studies, the participants presumably established a reference to themselves in terms of the potentially threatening nature of the stimuli, which may have significantly influenced attentional processes. The simple presentation of emotional faces used in the present study may not activate the social victim schema in a way that leads to higher vigilance for threatening stimuli or emotional stimuli in general. This may lead to the finding that emotions, which were expected to be relevant in the context of peer victimization, were not associated with attentional biases here. We suspect that the mere presentation of emotional faces is more of a projection screen for one's emotional state. Since the results of Iffland et al. (2019) and Iffland and Neuner (2022) also point in different directions concerning the type of attentional biases, it can be assumed that stimulus choice is likely to be crucial for the presence and nature of attentional biases.

Furthermore, analyses showed altered attentional processes that were related to higher levels of emotional childhood maltreatment. These processes were particularly evident for angry faces and suggest that experiences of emotional abuse led to increased attention toward angry faces. The results confirmed the assumption that attention processing of angry faces as potentially threatening stimuli is influenced by adverse childhood experiences (Gibb et al., 2009; Kelly et al., 2015; Iffland and Neuner, 2020). In addition, our findings indicated attentional avoidance of happy facial expressions in individuals reporting emotional abuse experiences, which may be indicative of dysfunctional emotion regulation. In support of this hypothesis, Burns et al. (2010) showed that experiences of emotional abuse were associated with difficulties in emotion regulation. In contrast to experiences of emotional abuse, our results showed that emotional neglect was associated with avoidance of angry faces. These differentiated findings for different subtypes of maltreatment experiences are in line with the results of Iffland and Neuner (2020) who used a face in the crowd task in their study to highlight that attentional processes differ between the two forms of emotional maltreatment. They reported a faster detection of negative faces in victims of emotional abuse, whereas slower recognition of negative and neutral faces was more likely in victims of emotional neglect. However, in contrast to the findings of Iffland and Neuner (2020), our findings regarding emotional neglect are less indicative of a general avoidance of emotional faces than of more differentiated processes, possibly involving the avoidance of highly salient stimuli (here angry faces). This is supported by the finding that emotional neglect was associated with an attentional shift toward happy faces. These findings could thus be the result of emotion regulation strategies, which in the context of emotional neglect could be associated with avoidance of threatening stimuli and a shift toward positive stimuli. However, because the final regression model for the orienting score for happy-neutral trials did not reach significance, these conclusions must be viewed with caution. Against our expectations, there was no relationship between maltreatment experiences and reaction times for disgust-neutral trials. Horstmann (2003) showed that disgusted facial expressions are more likely to be interpreted as expressions of emotional experience, whereas anger is more likely to be perceived as having an informative character at the interpersonal level. Future studies should therefore consider the use of disapproval faces as stimuli for social rejection (Burklund et al., 2007). Our results indicate differentiated attentional processes that are influenced by various forms of maltreatment, but particularly in individuals with emotional or relational maltreatment experiences.

In line with previous results, our study showed that experiences of relational peer victimization and emotional child maltreatment in childhood and adolescence influence attentional processes in adulthood. Higher levels of peer victimization were associated with facilitated attention to sad facial expressions in our sample. The results are thus indicative of mood-congruent attentional biases in individuals who have experienced relational peer violence. In addition, altered attentional processes for angry faces were present in participants with higher levels of emotional child maltreatment experiences. Adverse childhood experiences, particularly experiences of emotional maltreatment and relational peer victimization, can thus be considered relevant to the development of cognitive schemata that continue to be activated in adulthood, and therefore can potentially influence new experiences, feelings, thoughts in social situations, and thus presumably mental health.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Ethics Committee of Bielefeld University. Written informed consent was not provided because of the design as an online study. The participants gave their informed consent by clicking on a box.

KB contributed to the conception and design of the work, the acquisition, analysis, interpretation of the data, drafted, revised, approved the manuscript, and ensures the accuracy and integrity of any part of the work. BI was the chief investigator for this study, contributed to the conception of the study, supervised data analyses, participated in the interpretation of the data, and critically revised the manuscript for important intellectual content. All authors contributed to the article and approved the submitted version.