English (pdf)
Article in xml format
Article references
Automatic translation
Send this article by e-mail
Cited by SciELO 
Similars in
SciELO 
Permalink
In today’s world full of conflicts, empathic capacity is becoming increasingly important. Empathy has long been viewed as a desirable characteristic and is generally associated with several interpersonal benefits, although recent research also points to the risk of some personal consequences (MacDonald & Price, 2019).
Empathy has been extensively studied over the last century, however there isn’t a consensus over its definition (Engelen & Röttger-Rössler, 2012). According to Davis (1983), it’s the ability to put ourselves in someone else’s shoes and thus understand their perspective and emotional states. Research suggests empathy as a multidimensional construct, involving at least two dimensions (see Sesso et al., 2021 for review): affective empathy, that involves experiencing and reacting to other’s emotional state (Bray et al., 2021; Davis, 1983; Stefan & Avram, 2018); and cognitive empathy, that implies intellectually understanding other’s perspective by imagine the reasoning behind that feeling or thought (MacDonald & Price, 2019; Stefan & Avram, 2018; Tone & Tully, 2014; Zaki & Ochsner, 2016). Although related, these two dimensions are distinct and can function independently (Tone & Tully, 2014).
Basic affective empathy components emerge noticeably young and can be observed as early as the neonatal period, involving both basic emotion-understanding skills (e.g., emotion recognition) and emotional reactivity processes (e.g., emotion contagion) (Bray et al., 2021; Tone & Tully, 2014). Evidence suggest that babies have a natural tendency to respond to other’s emotional signals, such that by earing same-age infants crying they resonate with that expression of distress, become themselves distressed and start crying too (Decety et al., 2018; Diego & Jones, 2007; Hutman & Dapretto, 2009; Tone & Tully, 2014; Zahn-Waxler & Van Hulle, 2012). The ability to reason about other’s intentions improves during the first few years of life, which facilitates a more sophisticated understanding of why other’s show signs of distress (Tone & Tully, 2014). In this line, cognitive empathy develops later than affective empathy, involving the same constructs of theory of mind and empathic accuracy (Bray et al., 2021; Decety et al., 2018).
Empathy and internalizing behaviour problems
Empathy is usually associated with positive outcomes, although under some circumstances it might be associated with maladaptive ones (Tone & Tully, 2014). This might happen when: (1) the empathic reaction to the other’s distress are extremely aversive do the self; (2) the individual take the other’s perspective in an excessive degree; (3) when individual ruminates about his role in the other’s distress; or (4) the individual has negative thoughts or feelings about other’s distress (Gambin & Sharp, 2016; Tone & Tully, 2014). This empathic overarousal might result in feelings of depression, anxiety, guilt and withdrawal from interpersonal situations (Gambin & Sharp, 2016; MacDonald & Price, 2019; Telle & Pfister, 2016; Tone & Tully, 2014). A systematic review by Schreiter et al. (2013), using adult with depressive symptoms, concluded that, depressive symptoms were associated with empathic distress and with impairments in cognitive empathy.
In this way, the link between extreme empathy and the vulnerability to develop internalizing symptoms, maybe understood through the lens of affective or cognitive empathy. Several studies have demonstrated that the affective dimension of empathy, which underlies personal distress, represents a risk factor for the development of different internalizing problems (Blair, 2005; Gambin & Sharp, 2016, 2018; Gawronski & Privette, 1997; Schreiter et al., 2013; Shu et al., 2017; Silton & Fogel, 2010; Thoma et al., 2011; Tone & Tully, 2014; Zahn-Waxler et al., 1991). In adults and youths, some research also points to the link between the personal distress and anxiety, depression and guilt (e.g., O’Connor et al., 2002, 2012; Schreiter et al., 2013; Thoma et al., 2011; Zahn-Waxler & Van Hulle, 2012). Literature also indicates that depressive and anxious patients, are more at risk to experience higher levels of distress and affective empathy, and lower levels of cognitive empathy (Alvi et al., 2020; Schieman & Turner, 2001; Schreiter et al., 2013; Washburn et al., 2016). According to O’Connor et al. (2002), individuals with depression may suffer from biased cognitions as they see themselves harmful to other’s and may react to other’s distress by imagining their own reaction to a past emotional pain (Gambin & Sharp, 2016; Schieman & Turner, 2001). Another study showed that people with anxiety are less accurately on tasks of cognitive empathy probably because of excessive theory of mind or over-mentalizing (Alvi et al., 2020; Washburn et al., 2016). Cognitive empathy, which might be related to co-rumination - the tendency to be distressed by discussing and rehashing other’s problems, without being able to comfort them, was also suggested as a risk for developing internalizing behaviours in other study (Cherewick et al., 2022).
Empathy and externalizing behaviour problems
Some studies focus on the key role of empathy in reducing some externalizing problems, like aggressive behaviour (Cherewick et al., 2022; Lo Cricchio et al., 2022). In this way, people who can understand and adopt other’s perspectives and have compassion and vicarious share the same emotions as them, are more likely to avoid mutual aggression or any type of harm (Eisenberg, 2000; Gambin & Sharp, 2016; Gini et al., 2005). Empathy might also reduce externalizing problems as it fosters the ability to comprehend the real motivation behind others’ behaviours and reduces the risk of wrongly classifying them as aggressive and, if they have that connotation, it also enhances the ability to tolerate them (Farrington, 1998). On the other way around, one study found that individuals who show a higher level of aggressive behaviour, also expressed lower empathic concern (Bush et al., 2000). Several other studies found that conduct disorder, which underlines aggressive behaviour, is related to lower levels of empathy (Lovett & Sheffield, 2007; Miller & Eisenberg, 1998; Schwenck et al., 2012; Sterzer et al., 2007). Also, according to Gambin and Sharp (2016), individuals with externalizing disorders display an opposite cognitive style, meaning that instead of having self-debasing distortions or self-focus, they are characterized by self-serving and blaming others, which decreases empathic distress and empathic responses. Despite these considerations, other studies did not find differences in empathy levels between violent and non-violent youths (Lindsey et al., 2001). Such weak associations between empathy and aggression were also documented in the Vachon et al. (2014) meta-analyses. However, increasingly literature demonstrates that empathy may also enhance interactions in the social world, promoting prosocial behaviour (Brazil et al., 2023).
Empathy and prosocial behaviour
Prosociality has been defined as a series of voluntary behaviours that are meant to benefit others and promote positive social relationships (Eisenberg & Miller, 1987; Nantel-Vivier et al., 2014). This trait can be characterized by the tendency to care, share, comfort and help other people (Flouri & Sarmadi, 2016). Prosocial competence emerges within the first years of life, due to the function of their growing cognitive and behavioural competence (Eisenberg & Miller, 1987; Nantel-Vivier et al., 2014). During preschool and school years, they become increasingly capable of having prosocial behaviours based on empathic feelings and thoughts (Eisenberg & Miller, 1987). Around this age, children also have increasing language skills, which prompts as a potential predictor for later concern of others (Tone & Tully, 2014). As children increases their understanding of emotions and causes, their ability to generate empathic responses and behaviours also expands (Tone & Tully, 2014).
From early on children show concern for other’s welfare and demonstrate signs of empathy and prosociality (Decety et al., 2016). Children of 1 and 2 years-old express comforting behaviours towards someone who is in distress and can even give up their favourite object as an empathetic action (Davidov et al., 2013; Knafo et al., 2008). Also, children of 12 months can help others by informing, like pointing toward an object that the experimenter is searching for (Dunfield et al., 2011). According to Decety et al. (2016), toddlers from 14-18 months old show signs of helping behaviours by fetching desired objects that are out of reach to the experimenter and helping to complete chores. Around 2 years old, children are sensitive to other’s distress and their prosocial interventions now take a variety of forms like sharing, helping and comforting (Decety et al., 2016). Finally, around 5-6 years old, prosocial behaviour is significantly correlated with the emotional state of the victim but not of their own, suggesting that empathic concern is what is influencing prosociality, instead of personal distress (Williams et al., 2014). As Decety et al. (2016) states, this early emergence of prosociality reflects a biological predisposition to act upon empathic motivations.
However, a past review found that this association between empathy and prosocial behaviours wasn’t significant (Underwood & More, 1982). Brazil et al. (2023) defend that these small effects may be attributed to the differential relations between the different forms of empathy and prosocial behaviour. Vossen et al. (2015) found a strongest association between cognitive empathy and prosociality, than with affective empathy. Decety and Yolder (2016) found no relation between affective empathy and prosocial behaviours but did find for the cognitive dimension. Other theorists argue that empathy mediates prosocial actions that have altruistic motivations, even though most studies use measures that are unlikely to be motivated by altruistic considerations and reflect scripted social behaviour or compliance to the peers demands (Eisenberg & Miller, 1987).
The literature around empathy has become increasingly extended over the years, although there isn’t a consensus over its associations with prosociality and internalizing or externalizing problems. In this line, the present study aims to examine the association between empathy, prosocial behaviours, internalizing and externalizing problems.
Method
Participants
The sample included 199 children (50.8% girls and 49.2% boys) between 10 and 15 years (M=12.05; SD=0.98). They were distributed by school years, 31.7% were in 5th degree (52.4% girls and 47.6% boys; age M=11.08, SD=0.41), 40.7% in the 6th degree (45.7% girls and 54.3% boys; age M=12.05, SD=0.67) and 27.6% in the 7th degree (56.4% girls and 43.6% boys; age M=13.16, SD=0.54). Most children had siblings (82.9%).
Mothers’ age ranged between 29 and 57 years (M=41.99; SD=5.36) and fathers between 29 and 65 years (M=44.61; SD=6.65). Mothers’ education level varied between 4 and 19 years (M=14.05; SD=3.52) and fathers between 4 and 21 years (M=13.71; SD=3.88). Most parents lived together (54.4% were married; 4.7% were cohabiting and 23.3% of the families were separated or divorced, 17.6% were in another situation) and worked full-time (96.2% mothers; 98.3% fathers).
Most of the children said they didn’t have or had professional psychological support (65.3%), mostly on private context 68.8%), 28.1% said that they have/had (mostly on private context 68.8%), and 6.5% preferred not to answer. From those who said that don’t have/had that support 19.7% said they wished to have it and 12% preferred not to answer, most of the children said that they didn’t want it. Most children said not having sleeping problems (49.2%), 34.7% reported having sleep problems and 16.1% preferred not to answer. Most children said they were not bullying victims (57.8%), 27.6% were bullying victims and 14.6% preferred not to answer.
Most children (51.3%) report spending 1 to 3 hours/per day seeing digital content (e.g., Netflix, Youtube), 16.6% report spending less time and 19.1% more time. Regarding online social networks 80.9% report spending less than 3 hours/per day (22.6% report not spending anytime, 48.7% less than 1 hour and 32.2% 1 to 3 hours/day), 8.5% report spending more than 6 hours/per day. The same was true for videogames with most of the children (63.8%) report spending less than 3 hours/per day (12.6% report not spending anytime, 33.2% less than 1 hour and 30.7% 1 to 3 hours/day), 18.1% report spending more than 6 hours/per day.
Instruments
Empathy.
Affective and cognitive empathy scores were obtained using the Questionnaire to Assess Affective and Cognitive Empathy in children (QACE, children version, Zoll & Enz, 2010). Children completed the self-reported questionnaire using a 5-point scale. The 22-item instrument assesses two dimensions: Affective Empathy (12 items), that relates to the process whereby emotions are experienced due to the perception of internal states in other (either emotions or thoughts and attitudes); and Cognitive Empathy (10 items), that includes different cognitive processes, from simple associative over learning mechanisms to explicitly taking over other’s perspective. The Global score is the average over all 22 items The instrument had acceptable psychometric qualities in the Portuguese version (α=.85 for the affective and α=.71 for the cognitive dimension; Veiga & Santos, 2011), as well as in the present study (α=.77 for the affective, α=.72 for the cognitive dimensions and global score α=.83).
Social behaviours.
Strengths and Difficulties Questionnaire (SDQ, Goodman, 1997) was used to measure prosocial, externalizing and internalizing behaviours. It is a brief behavioural screening that allow us to identify emotional and behavioural problems in children and adolescents. It has different versions, and, in this study, it was used the youth self-reported measure for 11-16 years-old (Goodman et al., 2010). Children and adolescents report on 25 items that are organized in 5 difficulties subscales, each one of them with 5 items: emotional symptoms, conduct problems, peer relationship problems, hyperactivity/inattention and prosocial behaviour. Each item is scored on a 3-point scale (0=Not True; 1=Somewhat True; and 2=Certainly True), with higher scores indicating more problems for all subscales, except for the prosocial one. The scoring can be done by two different methods. We can consider that there is the prosocial behaviour scale and generate the ‘total difficulties score’ with the other’s original scales (by summing the emotional symptoms, hyperactivity/inattention, conduct problems and peer relationship problems). In this study we considered the alternative score to low-risk populations, diving SDQ into ‘internalizing problems’ (by summing emotional symptoms and peer relationship problems), ‘externalizing problems’ (by summing conduct problems and hyperactivity/inattention) and prosocial scale (Goodman et al., 2010). In the Goodman et al. (2010) study, factor analyses generally supported this second scoring, with good convergent and discriminant validity on the internalizing and externalizing subscales. In this study there has an α=.58 for the prosocial behaviour, the ‘internalizing problems’ dimension had an α=.60 and the ‘externalizing’ dimension showed a α=.66.
Results
Children in our sample described themselves as empathic, with girls presenting significantly higher values compared to boys (see Table 1). There was also a significant negative correlation with child’s age (AE r=-.27, p<.01; CE r=-.15, p<.05; GE r=-.24, p<.001). Father’s education level was positive correlated with child’s empathy (AE r=.18, p<.05; CE r=.19, p<.05; GE r=.21, p<.05). No other significant differences or associations were found.
Children described themselves as having significantly more prosocial behaviour than behavioural problems ENT#091;t(2,198)=-29.71, p<.001 and t(2,198)=-19.15, p<.001 comparing prosocial behaviour with internalizing and externalizing problems respectivelyENT#093; and more externalizing than internalizing problems ENT#091;t(2,198)=8.18, p<.001ENT#093; (see Table 2). Girls presented significantly higher values on internalizing and prosocial behaviours compared to boys (see Table 2). There was also a significant negative correlation between externalizing behaviours and child’s age (r=-.17, p<.05). Fathers’ educational level was significantly correlated with child’s prosocial behaviour (r=.17, p<.05). No other significant differences or associations were found.
Three hierarchical regression analysis were performed, one for each type of behaviour (externalizing, internalizing and prosocial). Child’s sex and age as well as fathers’ education were entered on the first block and child’s empathy (both cognitive and affective) on the second block.
Although child’s sex was a significant predictor for both internalizing and prosocial behaviours, when empathy was added, only child’s affective empathy was significant (R 2 =.11, ΔR 2 =.07, ΔF=5.83, p<.01; with β=.36; t=3.31, p<.001) for internalizing behaviours and only child’s cognitive empathy was significant (R 2 =.20, ΔR 2 =.10, ΔF=9.14, p<.001; with β=.24; t=2.71, p<.01) for prosocial behaviour.
For child’s externalizing behaviours, there was a significant relation with child’s age (β=-.17; t=-2.00, p<.05), however, when empathy was added none of the variables reached significance.
Discussion
As expected, gender influenced our results, showing that girls were more prosocial and empathic, which is in line with the previous literature that shows that boys usually exhibit lower levels of this variables, possibly due to social perception and gender stereotype (Belacchi & Farina, 2012; Eisenberg & Fabes, 1998; Schwenck et al., 2014). Also, our study was in line with the theoretical field (Gambin & Sharp, 2016; Nantel-Viver et al., 2014), showing that girls have higher levels of internalizing problems, compared to boys. Zahn-Waxler and her collaborators (1991) proposed a model that explains that girls with highly capacity of empathy may be at particular risk for internalizing problems due to gendered patterns of socialization. Literature suggests that it is attributed importance to the girl’s ability to recognize, identify and respond to other’s emotional state, and, for those who already have excessive empathy this may put them at risk for empathic stress that leads to internalizing problems (MacDonald & Prince, 2019).
Regarding age, we got a surprising result that showed that empathy decreased with child’s age. In the study of Schwenck et al. (2014), it was found that older children and adolescents show higher cognitive empathy skills, compared to preschool years, particularly in the components of emotion recognition and affective perspective taking. In this way, it was expected that our results were in line with this framework and further studies will be necessary to explain our different result. Another result regarding age that was not that surprising, was that externalizing problems decreased in older children. In most of the studies, normative children exhibit lower levels of externalizing problems over time, as they have a better understanding of emotions and how to handle them (Broidy et al., 2003; Campbell et al., 2006; Fanti & Henrich, 2010; Nagin & Tremblay, 1999; Shaw et al., 2003).
Our findings demonstrated that children with higher affective empathy had more internalizing problems, which is broadly consistent with an extant literature (Gambin & Sharp, 2016, 2018; Schreiter et al., 2013; Shu et al., 2017; Tone & Tully, 2014). Specifically, this result indicates that a heightened identification with the other’s emotional experience, in some cases feeling those emotions through personal distress, is associated with internalizing symptoms such as depression, anxiety and stress (Bray et al., 2021; MacDonald & Prince, 2019). Some studies point to the positive association between empathic distress and depressive symptoms in adults (Schreiter et al., 2013). Other studies found an association between empathic distress and anxiety (Shu et al., 2017; Tibi-Elhanany & Shamay-Tsoory, 2011), and one found an association between affective sharing and anxiety (Tibi-Elhanany & Shamay-Tsoory, 2011). Some studies with children found that this excessive empathy and empathic distress, were related to depressive symptoms (Olweus & Endresen, 1998; Robins & Hinkley, 1989).
One theory developed by Bray et al. (2021) is that there are four components of empathy: cognitive empathy and three of affective empathy - affective sharing, empathic concern and empathic distress. Accordingly, cognitive empathy and affective sharing are empathy processes, while empathic distress and empathic concern are considered emotional reactions to other’s experiences (Davis, 1983). Some studies have argued that an individual may respond more with empathic distress, rather than empathic concern, if there is a gap off their ability for emotion regulation and self-other distinction (Lamm et al., 2007). Some studies have shown that emotion regulation is one of the core processes involved in empathy and is highly implicated in some internalizing problems, like depression and anxiety (Amstadter, 2008; Decety, 2010; Ehring et al., 2010). In this way, the most reasonable explanation is that individuals who excessively empathize with the negative emotions of other’s, may then feel difficulties in regulating and modulating their emotional expression, responding with empathic distress that then induce internalizing problems (MacDonald & Prince, 2019).
Alternatively, it was shown that there was an underlying tendency - emotional reactivity - between affective empathy and internalizing symptoms in children, such as anxiety, negative mood and interpersonal problems (Wei et al., 2005). Emotional reactivity is defined as the degree to which someone responds to environment stimuli with hypersensibility, lability or emotion flooding (Bray et al., 2021). In this way, children who are more sensitive and reactive to the emotional states of others may report higher levels of affective empathy and higher levels of anxiety and depression (Bray et al., 2021).
According to Zahn-Waxler and Hulle (2012), a high capacity of empathy is not detrimental on its own but can increase the levels of internalizing problems when it’s combined with other intra or inter-individual factors. Even though this study did not assess them, other authors found that rumination and worry can increase the link between both self-reported empathy (Greenberg et al., 2018) and internalizing problems (Watters & Wojciak, 2020).
Affective empathy increases brain activity and is involved in emotional processing, understanding and simulating other’s actions, so it facilitates mirroring of the other’s states to a greater extent than cognitive empathy (Gabin & Sharp, 2016). In this way, it makes sense that on our study cognitive empathy wasn’t related to internalizing problems, like the affective dimension was. This lack of findings is in line with some of the previous literature, that did not find associations between these two variables as well (Bray et al., 2021; Derntl et al., 2012; Gambin & Sharp, 2016; Lee, 2009; Schneider et al., 2012; Thoma et al., 2011). Our lack of a significant association between cognitive empathy and internalizing symptoms may be explained through measurement limitations or developmental factors. Taking this into account, cognitive empathy is developed throughout middle childhood (Devine & Hughes, 2016), so more complex cognitive skills that drive the association with internalizing symptoms, may not be yet present in children with an average of 12 years-old, like they were in our study.
Another consensual idea that has been accepted in the history of psychology is that empathy is an important determinant of prosocial behaviour (Aronfreed, 1970; Batson & Coke, 1981; Eisenberg & Miller, 1987; Feshbach, 1978; Hoffman, 1984). Regarding this, our study found that only child’s cognitive empathy was significant for prosocial behaviour. In this way, our findings indicate that cognitive and affective empathy appear to be independent constructs and relate differently to the prosocial aspects (MacDonald & Prince, 2019).
As it was exposed before, children who have excessive affective empathy and can’t clearly differentiate their own internal states from those around them, may feel personal distress and become very unlikely to help or to have prosocial behaviours (Hoffman, 1984). However, the capacity to accurately understand other’s emotions and differentiate between their emotional states and the other’s - which is a core component of cognitive empathy - has been cited as a precursor to prosociality (Eisenberg & Miller, 1987).
Emotion understanding is a very complex cognitive construct and allows children to catch other’s points of view, emotions, desires, beliefs, and intentions (Belacchi & Farina, 2012). These capacities seem to be fundamental to give an adequate response during interactions with others, particularly during times of distress (Belacchi & Farina, 2012). Other authors have suggested that other cognitive components are crucial for these positive social behaviours, like the ability to mentally represent emotion-eliciting events or people’s internal states (Barnett & Thompson, 1985; Belacchi & Farina, 2012).
Children with advanced perspective-taking ability may have an advantage in discerning when someone needs help when that need is subtle and must be inferred, compared to someone who can only understand when the need of the other is really obvious and salient (Peterson, 1982). In this way, children who can understand these subtle cues and can infer the other’s feelings, might be more aware off the need to offer assistance to the other’s and, therefore, come across as more prosocial (Barnett & Thompson, 1985). This perspective-taking ability is one of the highest cognitive implications and is associated with greater mind-reading skills, allowing children to comprehend hidden emotions (Belacchi & Farina, 2012). The increase of this cognitive ability that allows them to also comprehend different causes of emotions, promotes more empathic responses to the stimuli even when the facial expression is not coherent (Belacchi & Farina, 2012). Additionally, it facilitates to consider the implications of the personal characteristics of other’s and both the situation itself, implying a clear distinction between their own’s emotions and those of other’s (Belacchi & Farina, 2012). In this way, the comprehension of other’s emotions and the ability to differentiate between ours and those of others, are the core ingredient of cognitive empathy that then elicits prosociality.
Like every study, some limitations should be considered when interpreting our findings. Firstly, all measures were self-report questionnaires, which may limit comparisons with other studies that used task-based or observational measures. The value of self-report questionnaires relies on the assumption that the participants are aware of their feelings and can decipher and report them accurately (Eisenberg & Fabes, 1990). However, children may have difficulties in comprehending questions or to be able to differentiate among closely related emotional states (like feeling empathy or becoming stressed by the other’s) (Eisenberg & Fabes, 1990). In this line, future research should have a range of measures to assess empathy, including observational measures, and look for multiple informants (like parents, teachers and the self).
Future studies should include emotional regulation, including cognitive aspects (such as emotion recognition and perspective-taking) and affective aspects (such as excessive worry and the ability to manage personal distress). One of the best methods for assessing emotional regulation may be through measures of physiological reactions to concrete situations that induce personal stress. Finally, even though affective, and cognitive empathy are the most used dimensions of empathy, future studies could consider other empathy dimensions such as behaviour ones or the Bray’s et al. (2021) perspective accounting with the four components.
