Disengagement is as revealing as prosocial action for young children’s responding to strangers in distress: How personal distress and empathic concern come into play

Abstract

In a paradigm of simulated stranger distress designed to elicit empathic arousal, this study examined multiple elements of responding in 61 preschoolers. Disengagement from stranger distress was underscored in addition to prosocial responding. All children encountered a female adult stranger feigning stomach ache followed by an infant manikin emitting cry sound in a bassinet. Behaviors were coded for other-oriented behaviors, personal distress, and disengagement. In contrast to the traditional supposition, behaviors indicative of personal distress covaried positively with empathic concern and negatively with disengagement. The findings of multiple regression analysis demonstrated how empathic concern and personal distress jointly related to disengaging behaviors in children’s response to stranger distress.

Keywords

Disengagement prosocial behavior personal distress empathic concern unfamiliar social partners

Young children’s prosocial response to social others in need has received ample attention for its association with socioemotional competence later in life (Denham et al., 2007; Eisenberg et al., 2006; Roth-Hanania et al., 2011). Much less visible in the literature, but potentially just as illuminating, is children’s disengaging behavior from others’ distress. Disengagement involves one’s response patterns that aid to detach the self from unpleasant experiences (usually negative emotions) by a range of behaviors including turning inward, shutting down, shifting attention from sources of arousal, or physically moving away, and so on (Campos et al., 1983; Denham et al., 2007; Rice et al., 2007). A lack of sensitive engagement in helping situations has been described as one aspect of callous–unemotional behavior that carries implications for aggression and antisocial behaviors (Frick & White, 2008; Viding & Kimonis, 2018; Waller, & Hyde, 2017; White, 2014). It remains unclear regarding what processes are involved in a child’s disengagement in a context that clearly reveals another person’s plight. Examining these processes may prove helpful in reducing insensitive responding and nurturing individual prosocial competence (Hepach & Warneken, 2018).

Children Are Not Always Very Helpful

A majority of researchers who study young children’s helping considers cooperation as part of human nature (e.g., Hepach, 2017; Warneken & Tomasello, 2006). Early in life, infants show sensitivity to others’ needs and emotions (Grossmann & Johnson, 2007) and social preference for those who have helped others over those who have harmed others (Hamlin & Wynn, 2011; Van de Vondervoort & Hamlin, 2018). Advancement in socio-cognitive understanding and emotional maturation during toddler years ushers in increases in prosocial behavior, including some simple forms of cooperation, sharing, comforting, and instrumental helping (Dunfield et al., 2011; Malti & Dys, 2018; Paulus, 2018; Svetlova et al., 2010). When children transition into late toddlerhood and preschool years, children help while taking into account others’ needs, the familiarity of social partners, and moral values of the helping action (Brownell et al., 2009; Hepach & Warneken, 2018; Malti & Dys, 2018; Paulus, 2019; Waugh & Brownell, 2017). Overall, existing theories and empirical findings suggest that there is an increase in the amount and complexity of prosocial behavior from infancy to early childhood. This increase, however, appears to taper off during late childhood (Hay & Cook, 2007; Malti & Dys, 2018). For example, Malti et al. (2016) found that, from late toddlerhood on, while some forms of prosocial behavior (e.g., cooperation) may still show increases with age, sharing behavior, however, did not show linear increases, and helping behavior actually displayed decreases over time. These developmental changes reflect increasingly discriminative, nuanced, and differentiated characteristics of prosocial behavior, which may likely be due to the combination of children’s dispositional factors (such as temperamental inhibition, proclivity for empathy, etc.) and advancing socio-cognitive evaluations of situational factors (such as social rewards, the familiarity of social partners, cost-effectiveness of action, etc.) (Malti & Dys, 2018).

While early emergence of prosocial behavior has been celebrated by many, a growing number of researchers have demonstrated findings that underscore children’s inconsistent helping and lack of responses to others in needy circumstances (e.g., Carpendale et al., 2015; Dunfield et al., 2011; Hammond & Brownell, 2018; Warneken & Tomasello, 2006; Waugh, & Brownell, 2017). In Dunfield et al.’s (2011) study on infants’ responding to others’ needs, for example, none of the infants (both 18- and 24-month-olds) spontaneously engaged in comforting behavior even after seeing the experiment hit her knee on the edge of a table and made her plight clearly known. This not-so-rosy aspect of responding when children were called to respond to distressed others has been noted by Paulus (2019): “In these studies, only a few children reacted immediately. Most of them required several cues before acting, and some did not help at all” (p. 3).

In the literature, some forms of deficits in attending to others’ distress cues, failures to recognize interpersonal emotions, and a lack of concern and sensitive responding in helping situations have been considered as early correlates of callous–unemotional behavior during childhood—an important predictor for the development of antisocial behavior (Waller & Hyde, 2017). These correlates are often interpreted as indicators for low levels of empathy, poor sense of guilt, and uncaring interpersonal style. It is unclear, however, whether young children’s failure to respond prosocially observed in either naturalistic or experimentally structured contexts suggests risks for developing callus–unemotional pattern or antisocial behavior. Could it be that, as Waugh and Brownell (2017) argue, not responding prosocially (or disengaging from the needy situations) is actually more prevalent and normal during early childhood than traditionally thought? So far, the literature has limited information on disengaging responding and the processes involved when children have opportunities to attend to and help needy others.

The present study sought to describe a less trumpeted, but potentially revealing, aspect of responding, namely, disengagement, in a context clearly showing distress in social others. This research also aimed to examine how other elements of prosocial responding related to disengaging behavior. Including both prosocial and disengaging responding may provide a more complete delineation of what young children do when they witness others in need. Moreover, illustrating how other processes are associated with children’s tendency to engage in or to disengage from a helping situation may contribute to a better understanding of early factors relevant to the development of prosocial and/or callus–unemotional behavior.

Response Emotions in Relation to Action in Helping Situations

Emotions in response to needy others have been considered as important processes underlying prosocial behavior (Drummond et al., 2016; Hoffman, 2000; Pletti et al., 2017). Prosocial motives and actions are purportedly supported by other-focused empathic emotions collectively referred to as empathic concern (Batson, 2011; Batson et al., 1987; Davis, 1994; Eisenberg & Miller, 1987; Hoffman, 1991). Emotions of empathic concern are organized around the well-being of others and may include feelings of concern, sympathy, tenderness, compassion, softheartedness, and so on. Another aspect of emotional responding to others’ distress can be self-focused and may involve feelings of alarm, aversion, frustration, anxiety, anger, helplessness, guilt, shame, and so on. This emotional aspect of responding has been described as personal distress (Batson, 2011; Davis, 1994), which is considered to be associated with a range of disengaging behaviors as a means to regulate one’s own unpleasant arousal (Campos et al., 1983; de Paul & Guibert, 2008; Denham et al., 2007; Eisenberg & Fabes, 1998; Silkenbeumer et al., 2016; Tully et al., 2015).

Notably, however, it is contended that the aversive arousal in a needy situation can also be relieved via prosocial helping because helping may likely ameliorate a sense of guilt, avoid criticism, and bring about reward from social others (Batson, 2011; Cialdini et al., 1997; Davis, 1994; Drummond et al., 2016; Hepach et al., 2012; Malti & Dys, 2018; Paulus, 2018, 2019). In this case, although the underlying motivations may be egoistic, the behavioral expressions are nonetheless seemingly prosocial. Further, findings with toddlers indicated that those who were more easily aroused and showed more negative affect tended to show more helping (Howes & Parver, 1987). Mothers who showed higher levels of arousal responded to infants’ negative cues with more empathic responses (Weisenfeld et al., 1984). Therefore, the purported positive mappings of “empathic concern with prosocial responding” and “personal distress with disengaging responding” may not fully capture the relations between empathy-related emotions and subsequent actions.

The Relation Between Empathic Concern and Personal Distress

The contrast drawn between empathic concern and personal distress has aided in our conceptualization of the qualitative distinctness of the other- and self-focused processes. Such a conceptual contrast, however, has encountered scrutiny. For example, Zahn-Waxler and Radke-Yarrow (1990) have cautioned that emphasis on such “a sharply drawn distinction” (p. 125) may lead to overly simplified dichotomous thinking about the two aspects of emotional responding that obscures their potentially important ontological relations. There are reasons to expect that the two emotions are not necessarily inversely related. First, they both involve emotional arousal in response to another’s distress; hence, they may similarly reflect a person’s general affect propensity and response patterns. In fact, positive correlations between empathy and distress have been reported (e.g., Batson et al., 1987; Davis, 1983; Fabes et al, 1990; Lin & McFatter, 2012; López-Pérez, & Ambrona, 2015). Moreover, an alternative, stemming from a developmental perspective, has postulated that empathy originates during toddlerhood on the basis of personal distress (Hoffman, 2000; Zahn-Waxler & Radke-Yarrow, 1990). Specifically, infants’ early reflexive crying in reaction to other infants’ crying (Sagi & Hoffman, 1976) has been regarded as a primitive antecedent of empathic arousal, which is experienced with no differentiation between the self and others. During toddlerhood, with socio-cognitive advancement in self-awareness, children gain a better understanding of the self/other distinction and become capable of processing others’ experiences with a clearer sense of ownership of the affect. In this way, children’s distress arousal becomes increasingly modulated into more mature forms of empathy. For this reason, some have argued that self-focused contagious arousal is a developmental antecedent of modulated forms of empathic concern (Hoffman, 1991; Zahn-Waxler & Radke-Yarrow, 1990). Other researchers also denote that egoistic distress does not necessarily clash with altruistic concern and they may covary in the same direction to jointly confer prosocial responding (e.g., Hirschberger, 2010; Hornstein, 1991). Hence, experiencing distress arousal, instead of being a predictor for a range of disengaging actions, may possibly also form a basis for engagement in an empathy-arousing situation.

Present Study

This study examined multiple elements of empathy-related responding to distress in strangers in preschool children. Instead of merely describing prosocial responding, this study focused on underscoring how preschoolers displayed disengagement from social others’ distress. We chose to examine preschool children, because during preschool years, prosocial behavior becomes more novel, selective, and nuanced, supported by their increasing sophistication in socio-cognitive evaluation, language processing, and emotion regulation (Brownell et al., 2009; Denham et al., 2007; Dunfield & Kuhlmeier, 2013; Hay & Cook, 2007; Malti & Dys, 2018; Paulus, 2019). However, information is sparse on disengaging behavior in preschool children.

We considered that a context with unfamiliar social partners would provide ample opportunities to observe the coexistence of empathy-related responding as well as unresponsiveness and disengagement. This was because an unfamiliar social context poses ambiguity and uncertainty, under which people tend to disengage for their own survival instead of investing their efforts in assisting others (Burger et al., 2001). In this study, we included an adult stranger and an infant stranger. Children tend to perceive infant strangers as weaker and in greater need for protection than adult strangers (Zebrowitz & Montepare, 2008), thus, may likely show more engaging responses to infants (Marsh et al., 2005). Including adult and infant strangers may show a fuller picture of empathy-related responding in preschoolers than only relying on one type of stranger.

The study addressed two research questions regarding (1) preschoolers’ disengaging behaviors in a context designed to elicit responding to distress in unfamiliar social partners, and (2) the association between the two response emotions—empathic concern and personal distress—and how they jointly related to disengagement. Based on the literature review, we predicted that, across the two stranger conditions, preschoolers would display not only other-oriented responding but also self-oriented distress and disengaging behaviors due to the multidimensional nature of empathy-related responding. Because the literature is limited on disengaging behavior of preschoolers in empathy-arousing contexts, hence, the examination of disengagement in this study was primarily exploratory.

For the relations between the multiple elements, we first considered the relation between empathic concern and personal distress because this relation would help predict how the two response emotions related to disengagement. With the discrepant views about the relation in the literature, we were advised by the notion that “a conceptual distinction does not imply ontological distinctiveness” (Campos et al., 2004, p. 379) and opted to depart from the oversimplified dichotomies sharply drawn between empathic concern or personal distress (Zahn-Waxler & Radke-Yarrow, 1990). Following the developmental alternative that views personal distress as an antecedent emotional substrate for empathic concern (as discussed previously), we regarded that two response emotions would both similarly reflect the child’s general attempt to adapt to situations encountered. Support for this decision could be exemplified by Tully et al.’s study (2015) that conceptualized empathy as a reflection of autonomic flexibility and personal distress inflexibility, each indexed by high and low heart rate variabilities, respectively. Their findings showed that the two response emotions were both associated with high facial distress and were both indicative of empathic engagement. To Tully et al.’s surprise, autonomic inflexibility characteristic of personal distress was linked to better emotion understanding embedded in children’s narratives—a form of engagement, suggesting the need to question the views that only emphasize the inverse relation between empathic concern and personal distress. Moreover, in a pilot study using only the infant stranger simulated condition described in the present research, we also found that the time preschoolers (N = 37, 17 females, M _age = 45.6 months, SD _age = 0.91) spent in behaviors indicative of empathic concern and personal distress revealed positive association (data available upon request). Thus, we predicted that empathic concern and personal distress would covary in the same direction in both the adult and infant conditions.

If empathic concern and personal distress covaried positively as predicted, then we hypothesized that they would both relate to response actions in the same direction. Specifically, the prediction was that empathic concern and personal distress would both positively relate to prosocial responding but negatively relate to disengagement. Thus, we predicted that personal distress would not necessarily positively relate to disengagement; in contrast, personal distress would even possibly negatively covary with disengagement.

Method

Participants

Sixty one typically developing preschoolers (23 females, M _age = 44 months, SD _age = .94) from a medium-sized Southern US city participated in the study. The majority of children were Whites (79%), 11% Blacks, and 10% Others. The research was approved by the Institutional Review Board at the authors’ University.

Setting and Cry Stimulus

The experiment was conducted in a 2.4 m × 4.0 m room equipped like a nursery with children’s books and toys on top of a table, chairs for preschoolers, a rocking chair, a chair for the caregiver, a bassinet, and an infant bottle in the bassinet. The cry stimulus was constructed using a 4-week-old male infant’s cry, recorded at home prior to a scheduled feeding. The digitization of the cry stimulus was conducted using a Kay Elemetrics CSL unit at 44.1 kHz with 16 bit sampling. The audio file of the cry was played via a JBL 4301 wireless speaker (hidden underneath the bassinet mattress), with peak amplitudes of approximately 82 dB (relative to 20 μN/m²) at 1 m from the source. The cry sample exhibited increasing intensity and lasted approximately 5 min.

Procedure

The caregiver sat in the corner of the room throughout the experiment but was instructed not to intervene or prompt any responding from the child. All children went through the Adult Stranger condition first due to the consideration that the second part of the cry stimulus would be so intense that would overwhelm some children causing the interaction with the adult stranger impossible.

The Adult Stranger condition

A female confederate came in holding a weighted lifelike infant manikin (19″ long) presenting herself as the mother of the infant. The confederate sat down in a rocking chair across the caregiver and started chatting with the caregiver. The conversation continued until the child warmed up and no longer showed behaviors indicative of anxiety toward the unfamiliar adult.

After building rapport, the stranger began feigning a sudden onset of gastrointestinal distress with her hands touching her stomach and painful expressions on her face, marking the beginning of the Adult Stranger condition. Unless the child responded, the confederate continued simulating the discomfort with constant groaning and verbalizing, “Oh, it hurts!” and indicated that the pain became worse. If the child responded, the confederate would say, “Thank you, but it still hurts!” The child’s responding were observed for as long as 5 min before the confederate stood up, placed the infant in the bassinet, and indicated that she had to go to her car for medicine. The confederate then left the room, marking the end of the observation. The observation terminated if the child displayed high levels of distress.

The Infant Stranger condition

About 15 s after the confederate left, an increasingly intense cry stimulus was played through the wireless speaker. The Infant Stranger condition lasted approximately 5 min to allow for observing the child’s spontaneous responding. If the child displayed high levels of distress in respond to the simulated infant crying, the condition ended instantly to prevent the child from experiencing more distress.

Behavioral Coding

The Oberver XT version 14 was the software program used for behavioral coding during video playback. Coding of behaviors was event-based with exact onset- and offset-times marked. Two preschoolers were excluded from data analysis due to unsatisfactory quality of the recorded video; thus, the sample consisted of 59 preschoolers. As summarized in Table 1, behaviors were coded for three major behavioral categories, including (1) other-oriented behavior, (2) personal distress, and (3) disengagement. Other-oriented behavior was defined as behavior that suggested a focus on the well-being another person (more details below). Behaviors indicative of personal distress were those that showed self-focused affective or physical processes as a result of witnessing another’s distress, including, for example, frowning, restlessness, covering ears, nervous laughing, whining, fussing, crying, asking the caregiver for comfort, asking to leave, and so on. Disengagement referred to any acts that indicated the child’s disinterest in the well-being of others or reluctance to engage with the person in distress. Examples of disengagement included, ignoring the distressed (e.g., continuing the previous activities), irrelevant speech, looking away from the distressed, moving away from the distressed, and so on. Co-occurrence of behavioral categories of other-oriented behavior and personal distress in time was possible (e.g., trying to help while showing distress emotions at the same time); however, other-oriented behavior and disengagement were mutually exclusive. The three behavioral categories were not exhaustive because there were moments when behaviors were too ambiguous to be coded as any one of the categories.

Table 1.

Coding Scheme for the Three Major Behavioral Categories and the Subcategories of the Other-Oriented Behaviors.

Behavioral category	Description	Examples
Other-oriented behaviors	Voluntary responses with a clear focus on the well-being of the distressed strangers
Concerned expressions	Responses showing sympathy or sadness on behalf of the distressed strangers	Concerned looing, mimicking facial expression of the distressed stranger
Cognitive inquiry	Behaviors reflecting the child’s reasoning about the distressed strangers	Asking about the cause of stranger distress, discussing with the parent about solutions
Approaching the distressed	Physically moving toward the distressed strangers	Walking over to the distressed strangers
Helping actions	Behaviors that focus on relieving distress of strangers	Patting the infant, offering comfort to the adult stranger
Personal distress	Self-focused distress reactions in response to the distressed	Burying face in hands, frowning, laughing nervously, seeking comfort, whining, fussing, crying
Disengagement	Responses reflecting the child’s indifference or reluctance to respond to the distressed strangers	Irrelevant speech, looking away from the distressed, covering ears, remaining uninfluenced, engaging in previous activity

The category of other-oriented behavior included four subcategories: (1) concerned expression, (2) cognitive inquiry, (3) physically approaching the distressed, and (4) helping actions. Concerned expression (indicative of empathic concern) was defined as a facial expression that showed obvious concern for the victim. Cognitive inquiry referred to any action that involved investigating, reasoning, or assessing the cause and/or situation of the distressed (e.g., visually searching for cause of the distress, asking how to help the distress, etc.). Approaching the distressed referred to physical movements that increased the proximity to the distressed. A helping action might involve offering comfort to the distressed to alleviate the distress of another person (e.g., patting the distressed gently, giving objects to the distressed in order to give comfort, verbally offering solutions, etc.). The four subcategories of other-oriented behavior were not mutually exclusive in time, that is, co-occurrence of two or more subcategories was possible. An example of such co-occurrence was when the child was showing concerned expression while inquiring about the cause of the stranger’s distress and approaching the distressed at the same time.

The duration of a behavior was indexed by the percent of time spent in the entire observation session. As an example, if preschoolers spent 33% of time in other-oriented behavior, it indicates that in 33% of the entire observation period preschoolers displayed at least one other-oriented behavior. When behaviors belonging to one category overlapped in time, the duration of the category was determined by the start time of the leading behavior and the end time of the ending behavior.

There were two primary coders of the videos, each responsible for randomly selected 50% of the videos. The primary coders received training from an experienced coder by coding 15 videos (randomly selected) under guidance. During the training phase, the two primary coders compared their coded data and discussed disagreements with the experienced coder until all the disagreements were resolved. There were two reliability coders, each of them was paired with one of the primary coders and was responsible for coding 20% of the primary coder’s videos. Interobserver reliability κ (Cohen's Kappa) values between the primary coders and the reliability coders ranged from .81 to .92.

Result

Responding Across Conditions

As predicted, preschoolers showed multiple elements of responding when witnessing the distress in both adult and infant strangers. Across the Adult and Infant conditions, preschoolers spent significantly greater amount of time in disengagement (percent of time, M = 41.13%, SD = 22.51%) than in other-oriented behavior (as a category, M = 33.21%, SD = 16.86%) and personal distress (M = 11.45%, SD = 13.52%) (all were significantly different from one another at p = .05 level). Gender differences were found in the time spent in the three behavioral categories. Females spent significantly more time in other-oriented behavior (M = 43.74%, SD = 16.12%) than males (M = 35.75%, SD = 11.68%), t(58) = 2.16, p = .02, but less time in disengagement (M = 31.48%, SD = 20.92%) than males (M = 41.375, SD = 19.44%), t(58) = −1.89, p = .03. There was no significant difference in behavior showing personal distress between females (M = 7.86, SD = 8.68) and males (M = 6.51%, SD = 11.99%).

For the subcategories of other-oriented behavior, across the two conditions, preschoolers spent the greatest amount of time in expressions indicative of empathic concern (M = 20.81%, SD = 13.56%), followed by cognitive inquiry (M = 8.99%, SD = 7.33%), helping actions (M = 14.89%, SD = 10.77%), and physically approaching the distressed (M = 1.00%, SD = 1.52%). All the results of comparisons between the four subcategories were significant at the p < .0001 level.

A Difference Score to Describe Behavioral Tendency Across Conditions

To describe preschoolers’ relative “tendency” for prosocial versus disengaging responding across the two conditions, we utilized a difference score tallied by subtracting the proportion of time spent in disengagement from that in other-oriented behavior. As a result, a positive difference score indicated a greater tendency to display prosocial responding, and a negative difference score disengagement. Preschoolers with positive difference scores were regarded as “prosocial responders” while those with negative scores were “disengagers.”

Based on the difference score, there were 20 prosocial responders and 39 disengagers. To compare how prosocial responders differed from disengagers, we used Wilcoxon rank-sum tests due to the small sample sizes and non-normal distributions of the variables observed by quantile plots (see Figure 1 for an example). The results indicated a significant difference between the two distributions of the time spent in personal distress, with the median for prosocial responders being significantly higher (median, Mdn = 20.32%, interquartile range: 5.31% to 26.31%) than the median for disengagers (Mdn = 4.28%, interquartile range: 1.68% to 11.40%), χ² value = 10.63, df = 1, p < .0001. For the subcategories of other-oriented behavior, prosocial responders spent significantly more time in empathic concern (Mdn = 32.82%, interquartile range: 17.75% to 40.25%) than disengagers (Mdn = 13.74%, interquartile range: 6.73% to 21.86%), χ² value = 16.67, df = 1, p < .0001. Prosocial responders also spent more time in cognitive inquiry (Mdn = 15.48%, interquartile range: 5.11% to 19.51%) than disengagers (Mdn = 4.46%, interquartile range: 2.36% to 10.24%), χ² value = 16.67, df = 1, p < .0001. Prosocial responders and disengagers did not differ in the times spent in approaching the distressed or helping actions.

Figure 1.

(a) Quantile Plot for the Duration (Proportion of Time Within the Entire Observation) of Personal Distress for Prosocial Responders Who Showed More Other-oriented Behaviors Than Disengaging Behaviors (n = 20). (b) Quantile Plot for the Duration (Proportion of Time Within the Entire Observation) of Personal Distress for Disengagers Who Showed More Disengaging Behaviors Than Other-oriented Behaviors (n = 39).

Summing up, times spent in empathic concern and cognitive inquiry constituted the main differences between prosocial responders and disengagers in other-oriented behavior. Notably, prosocial responders spent significantly more time in personal distress than disengagers.

Responding by Condition

There was no difference in the amount of time spent in other-oriented behavior between the two conditions. But the mean duration of behaviors indicative of personal distress in the adult condition (M = 15.22%, SD = 24.72%) was longer than the infant condition (M = 10.51%, SD = 17.77%), t (58) = 1.75, p = .04. Preschoolers did not differ in the times spent in disengagement between the two conditions. Within the category of other-oriented behavior, preschoolers spent more time in cognitive inquiry in the infant condition (M = 14.89%, SD = 10.77%) than in the adult condition (M = 4.85%, SD = 7.98%), t (60) = 6.83, p < .0001. Preschoolers also spent more time in physically approaching actions toward the infant (M = 2.64%, SD = 3.67%) than the adult (M = .91%, SD = 2.47%), t (60) = 3.27, p = .00. Time spent in helping actions was significantly longer toward the infant (M = 5.21%, SD = 12.27%) than the adult (M = .45%, SD = 1.42%), t (60) = 3.02, p = .00.

Bivariate associations

To explore the associations of other-oriented behavior and personal distress with disengagement as reported below, empathic concern was the only subcategory that yielded significant results. Thus, the description regarding joint associations of other- and self-oriented responding with disengagement focused on the subcategory of empathic concern.

As predicted, but in contrast to the traditional notion about the opposing qualitative natures of empathic concern and personal distress, empathic concern in response to both adult and infant strangers was positively correlated with personal distress, r = .30, p = .02, and r = .60, p < .001, respectively (Tables 2 and 3). Empathic concern was negatively correlated with disengagement (as would be expected) for both the adult and infant conditions, r = −.58, p < 0.0001, and r = −.44, p < 0.001, respectively. Also as predicted, but in contrast to the traditional notion of the positive association of personal distress with disengagement, personal distress in the two conditions was negatively correlated with disengagement, r = −.46, p < 0.0005, and r = −.36, p < 0.01, respectively. Thus, for both conditions, empathic concern and personal distress covaried in the same direction and both varied with disengagement in the opposite direction. However, we did not find the positive association of empathic concern with approaching or helping in both conditions as predicted.

Table 2.

Correlations Between Behaviors in the Adult Stranger Condition (N = 59).

	Other-oriented	Empathic concern	Cognitive inquiry	Approaching the distressed	Helping action	Personal distress
Other-oriented
Concerned expression	.92**
Cognitive inquiry	.25*	−.11
Approaching the distressed	.01	−.06	.00
Helping actions	.12	.03	.12	.31*
Personal distress	.23*	.30*	.02	−.15	−.12
Disengagement	−.72**	−.58***	−.06	−.11	−.29*	−.46**

*p < .05; **p < .005; ***p < .0005.

Table 3.

Correlations Between Behaviors in the Infant Stranger Condition (N = 59).

	Other-oriented	Empathic concern	Cognitive inquiry	Approaching the distressed	Helping action	Personal distress
Other-oriented
Concerned expression	.36*
Cognitive inquiry	.47**	.08
Approaching the distressed	.32*	−.17	.22
Helping actions	.24	.01	−.09	.02
Personal distress	.27**	.60***	.01	−.14	−.16
Disengagement	−.68**	−.44***	−.36*	−.08	.34*	−.36*

*p < .05; **p < .005; ***p < .0005.

Prediction of disengagement from personal distress and empathic concern

To address the question regarding how empathic concern and personal distress jointly related to disengagement, multivariate regression analysis was performed to test two models, one for the adult condition and the other the infant condition. Both models had disengagement entered as the criterion and personal distress as the main predictor and empathic concern as the regulatory/process factor. The rationale was based on the account that self-focused arousal is a developmental antecedent of a modulated form of empathic concern (Hepach et al., 2013; Hoffman, 1991; Zahn-Waxler & Radke-Yarrow, 1990), which often supports prosocial responding. Gender was controlled in both models due to the gender differences reported previously.

The results of the adult stranger model revealed an interaction effect between personal distress and empathic concern on disengagement from the distress of the adult stranger, b = .20, SE = .08, t = 2.60, p = .01, with empathic concern attenuating the strength of the negative association between personal distress and disengagement (Table 4). Figure 2 summarizes this interaction, showing the prediction of disengagement from three levels of personal distress (−1SD, mean, +1SD) at three levels of empathic concern (−1SD, mean, +1SD). Subsequent simple slope tests indicated that the negative relation between personal distress and disengagement varied with levels of empathic concern, with higher levels of empathic concern associated with weaker associations between personal distress and disengagement, suggesting a moderating role of empathic concern. The infant stranger model indicated that, when empathic concern was controlled, the relation between personal distress and disengagement became nonsignificant, b = −.14, SE = .15, t = −.97, p = .34, suggesting a mediating role of empathic concern, F (2, 58) = 7.38, p = .001. R ²(adj) = 0.2. The indirect effect is −.19, SE = .09, t = –2.27, confidence interval (CI): [−0.39, −0.03] (bias corrected at 95% CI yielded by a bootstrapping procedure with 5,000 resamples) (Figure 3). Of note, the data did not fit a pattern of mediation by reversing the roles of personal distress and empathic concern in the model. Taken together, both the two regression models revealed an overarching pattern indicating the negative association of personal distress with disengagement with empathic concern either moderating or mediating the association.

Table 4.

Results of Multiple Regression Analysis for the Prediction of Disengagement (N = 59).

	Predictors	Estimate b (SE)	t ratio	p level	CI	Model
Criterion	Intercept	−0.07 (0.10)	−0.73	.46	[−.27, .12]	F (4, 54) = 13.73
	Gender	−0.11 (0.10)	−1.07	.29	[−.31, .09]
	PD)	−0.44 (0.11)	−3.79	.0002	[−.66, −.22]
	EC	−0.49 (0.10)	−4.74	<.0001	[−.70, −.29]	R² = .46, p < .0001
	(PD) × (EC)	0.20 (0.08)	2.60	.01	[.05, .36]

Note. All variables were standardized. (PD) × (EC) = cross product of personal Distress and Empathic Concern. CI = confidence interval: [lower 95%, upper 95%].

Figure 2.

Prediction of the Duration (Proportion of Time Within the Entire Observation) of Disengagement From Three Levels of Personal Distress (−1SD, Mean, +1SD) at Three Levels of Empathic Concern (−1SD, Mean, +1SD) (N = 59) in Response to the Adult Stranger’s Distress.

Figure 3.

Results of Multiple Regression Analysis Regressing the Duration (Proportion of Time Within the Entire Observation) of Disengagement on the Durations of Personal Distress and Concerned Expression in Response to the Infant Stranger’s Distress (N = 59).

Discussion

This study described a largely overlooked, but potentially important, element in preschool children’s responding to social others’ distress, namely, disengagement. During preschool years, children exhibited empathic sensitivity to distressed strangers despite evidence of egoistic distress. They showed empathic concern, cognitively explored the causes of the distress, physically approached the strangers, and offered comfort and help. Such proficiency in regulating distress arousal and exhibiting modulated forms of empathic responding evidenced sophisticated socioemotional competence at the preschool ages (Dunfield & Kuhlmeier, 2013; Hepach & Warneken, 2018; Lamm et al., 2007). Although exhibiting remarkable prosocial competence, preschoolers sometimes also responded with a lack of interest, indifference, insensitivity, or withdrawal (sometimes in conspicuous manners). The findings demonstrated the multidimensional nature of empathy-related responding during preschool years.

Several findings were unique to preschoolers’ disengaging behavior observed in this study. First, across conditions, preschoolers spent more than 40% of time in disengaging behavior, which was significantly longer than the duration of other-oriented behavior (about 33% of time). Second, based on the difference score between the durations of other-oriented and disengaging behaviors, there were more disengagers (n = 39) than prosocial responders (n = 20). Third, compared to prosocial responders, disengagers spent less time in behaviors indicative of empathic concern and cognitive inquiry (as would be expected). Somewhat counterintuitively, however (when considering the traditional notion about disengagement as a means of regulating negative emotions as noted), disengagers spent less time in behaviors indicative of distress arousal. Fourth, across conditions, disengagement was negatively correlated with empathic concern (as would be expected), but negatively correlated with personal distress—a pattern opposite to the purported positive link between personal distress and disengaging behavior. Along with another unanticipated finding (although not regarding disengagement) that empathic concern and helping were not related, these results did not square with the putative positive associations of “empathic concern with prosocial responding” and “personal distress with disengaging responding.”

The findings pointed to an inconvenient truth about helping in preschoolers when the strangers’ predicament was plain to see. Perhaps, as Waugh and Brownell (2017) noted for very young children, failure to behave prosocially in preschoolers is more typical than expected—particularly during a time when children become more selective, discriminative, and calculated in helping situations (Carpendale et al., 2015; Malti & Dys, 2018). Numerous possible combinations of situational and dispositional factors might have hindered preschoolers from spontaneously responding to the distressed strangers in a prosocial manner. It was plausible, as would be expected, that preschoolers disengaged to distract the self from the source of distress as a means to regulate arousal (Campos et al., 1983; Denham et al., 2007). Some preschoolers might have disengaged because they were not sure what the problems were and how and when to intervene effectively—important elements that support prosocial responding (Dunfield & Kuhlmeier, 2013; Svetlova et al., 2010). The presence of a competent adult might have diminished children’s sense of responsibility for taking helping action—similar to the bystander effects described by Darley and Latané (1968) in adults and Plötner et al. (2015) in young children. Individual differences in temperamental dispositions, for example, social inhibition (Laible et al., 2014; Song et al., 2017; Young, Fox, & Zahn-Waxler, 1999), might have led to disengaging tendency. It was also possibly due to the context of unfamiliar social partners in the current study; preschoolers disengaged because of their growing preferences for helping familiar social partners and in-group members over strangers or out-group members (Hay & Cook, 2007; Malti et al., 2016). Relatedly, an unfamiliar social context might have made preschoolers hesitant to take helping actions even though they displayed expressions of concern. As Maner and Gailliot (2007) reported, the link between empathic concern and helping appeared more pronounced when the person in need was a familiar social partner rather than a stranger. Thus, some preschoolers’ empathic concern was not related to helping action—a circumstance captured by Zahn-Waxler and Radke-Yarrow’s (1990) statement “Feelings of concern and compassion do not require action, any more than prosocial behaviors automatically imply sympathic concern” (p. 125).

In addition to the possible reasons behind preschoolers’ disengagement noted above, this study provided a unique perspective to examine disengaging behavior via the lens of personal distress and its relation with empathic concern. Although preschoolers spent more time in personal distress in the adult stranger condition (possibly due to adult’s more conspicuous verbal descriptions and bodily expressions of discomfort that elicited more arousal, Zebrowitz et al., 2003), the durations of disengagement did not differ. Importantly, for both conditions, empathic concern and personal distress covaried in the same direction, providing support for a dearth of relevant descriptions that considered both as two seemingly divergent yet fundamentally inseparable affective responses, which in concert reflect one’s general proclivity for sensitive responding to others’ situations (Batson et al., 1987; Davis, 1983; Hirschberger, 2010; Hoffman, 1991; Hornstein, 1991; Zahn-Waxler & Radke-Yarrow, 1990). The positive association between empathic concern and personal distress provided a basis to interpret the somewhat counterintuitive finding of the inverse association between personal distress and disengagement—an association not often documented in the literature but has been reported (e.g., Howes & Parver, 1987; Lin & McFatter, 2012; Waugh & Brownell, 2017; Weisenfeld et al., 1984). The present study demonstrated that, although in different ways (an interaction and a mediated association between the two response emotions in the two conditions, respectively), disengagement reflected a response action jointly incurred by personal distress and empathic concern. The findings from both conditions indicated that disengaging from the distressed strangers was more likely to occur when empathic concern was low; and low levels of empathic concern were closely connected to low levels of personal arousal. It appeared that sufficient levels of aversive arousal might be needed to bring forth sufficient levels of empathic concern, which subsequently made disengaging responding less likely to happen.

Although preschoolers’ disengaging responding was evident and not necessarily surprising, the patterns of associations between response emotions and actions carried relevance in considering the risk factors of developing callous–unemotional qualities. Low levels of arousal may be due to deficits in perceptual and socio-cognitive processing of others’ distress, which in turn is associated with a lack of concern and sensitive responding—often reflected as low empathy, poor sense of guilt, and an uncaring interpersonal style (Waller & Hyde, 2017). Future research should address the question of to what extent should disengagement be considered as a normal part of responding as opposed to red flags for developing callous and unemotional traits when witnessing another person in distress.

Limitations and Future Directions

The present study has several limitations. First, the present study observed preschoolers’ disengaging and prosocial behaviors in a controlled laboratory setting. Caution is needed when generalizing the current findings to real-life situations. Factors such as the presence of onlookers (adults or children, familiar or unfamiliar social partners), available tools or resources, and so on, need to be examined in various settings to better inform children’s disengaging and prosocial responding. Second, the findings were based on regressional findings, rendering it inconclusive about the causal relations of the behaviors. Future research can benefit from utilizing sequential and/or time series analysis to better describe the temporal organizations of various response components and provide insights for the underlying dynamic processes leading to prosocial and disengaging behaviors. Third, there was a lack of information regarding cognitive competencies and dispositional qualities of preschoolers that potentially influence prosocial responding (Drummond et al., 2016; Waugh & Brownell, 2017; Young et al., 1999). Incorporating data informing perspective taking capabilities, metacognitive skills, and personality qualities can better inform correlates of children’s disengaging and prosocial responding.

Despite the limitations, this study extended prior research on prosocial behavior in young children by documenting how different aspects of empathy-related responding related to one another. Instead of focusing merely on actions that are perceived to be prosocial, the current study documented behaviors suggestive of distress arousal and disengagement. Notably, the findings highlighted the counterintuitive positive association between empathic concern and personal distress that has been conceptualized as qualitatively contradictory underlying different motivational goals. Further, the present study described how personal distress and empathic concern jointly related to disengaging behavior—a unique perspective that has implications for ameliorating insensitive responding and nurturing individual prosociality.

Conclusion

When responding to strangers in distress, self-oriented distress may not necessarily imply disengaging action; likewise, other-oriented concern may not automatically lead to helping. This study provided evidence that personal distress and empathic concern covaried in the same direction and related in concert to young children’s disengagement from distress in both adult and infant strangers. Simply focusing on prosocial responding delineates an incomplete picture of how children respond to others in distress. Disengagement merits rightful attention for its nature and function and its relations to other elements of empathy-related responding.

Footnotes

Acknowledgments

The authors are grateful to Nicholas Barker, Maddison Knott, and Patrick M. Chu for their assistance in various aspects of this research. The authors would also like to thank the Infancy Laboratory at the University of Connecticut for generously offering the infant cry stimulus for this research.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Instructional and Scientific Equipment Fund at the authors’ University.

ORCID iD

Hung-Chu Lin

References

Batson

C. D

. (2011). Altruism in humans. Oxford University Press.

Batson

C. D.

Fultz

Schoenrade

P. A.

(1987). Distress and empathy: Two qualitatively distinct vicarious emotions with different motivational consequences. Journal of Personality, 55, 19–39.

Brownell

C. A.

Svetlova

Nichols

(2009). To share or not to share: When do toddlers respond to another’s needs? Infancy, 14(1), 117–130.

Burger

J. M.

Soroka

Gonzago

Murphy

Somervell

(2001). The effect of fleeting attraction on compliance to requests. Personality and Social Psychology Bulletin, 27(12), 1578–1586.

Campos

J. J.

Barrett

K. C.

Lamb

M. E.

Goldsmith

H. H.

Stenberg

. (1983). Socio-emotional development. In Mussen

P. H.

(Ed.), Handbook of child psychology (4th ed., pp. 783–915). Wiley.

Campos

J. J.

Frankel

C. B.

Camras

(2004). On the nature of emotion regulation. Child Development, 75, 377–394.

Carpendale

J. I. M.

Kettner

V. A.

Audet

K. N.

(2015). On the nature of toddlers’ helping: Helping or interest in others’ activity? Social Development, 24, 357–366.

Cialdini

R. B.

Brown

S. L.

Lewis

B. P.

Luce

Neuberg

S. L.

(1997). Reinterpreting the empathy-altruism relationship: When one into one equals oneness. Journal of Personality and Social Psychology, 73, 481–494.

Darley

J. M.

Latané

(1968). “Bystander intervention in emergencies: Diffusion of responsibility”. Journal of Personality and Social Psychology, 8, 377–383.

10.

Davis

M. H

(1983). Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of Personality and Social Psychology, 44, 113–126.

11.

Davis

M. H

. (1994). Empathy: A social psychological approach. Westview Press.

12.

Denham

S. A.

Bassett

H. H.

Wyatt

. (2007). The socialization of emotional competence. In Grusec

J. E.

Hastings

P. D.

(Eds.), Handbook of socialization: Theory and research (pp. 614–637). Guilford Press.

13.

Drummond

J. D. K.

Hammond

S. I.

Satlof-Bedrick

Waugh

W. E.

Brownell

C. A.

(2016). Helping the one you hurt: Toddlers ‘rudimentary guilt, shame, and prosocial behavior after harming another. Child Development, 88(4), 1382–1397.

14.

Dunfield

Kuhlmeier

V. A.

(2013). Classifying prosocial behavior: Children’s responses to instrumental need, emotional distress, and material desire. Child Development, 84, 1766–1776.

15.

Dunfield

Kuhlmeier

V. A.

O’Connell

Kelley

(2011). Examining the diversity of prosocial behavior: helping, sharing, and comforting in infancy. Infancy, 16, 227–247.

16.

Eisenberg

Fabes

R. A

. (1998). Prosocial development. In Damon

(Series Ed.) Eisenberg

(Vol. Ed.), Handbook of child psychology, Vol. 3: Social, emotional, and personality development (5th ed., pp. 701–778). Wiley.

17.

Eisenberg

Fabes

R. A.

Spinrad

T. L

. (2006). Prosocial development. In Eisenberg

(Vol. Ed.) Damon

Lerner

R. M.

(Series Eds.), Handbook of child psychology, Vol. 3: Social, emotional, and personality development (6th ed., pp. 646–718). Wiley.

18.

Eisenberg

Miller

P. A.

(1987). The relation of empathy to prosocial and related behaviors. Psychological Bulletin, 94, 100–131.

19.

Fabes

R. A.

Eisenberg

Miller

(1990). Maternal correlates of children’s vicarious emotional responsiveness. Developmental Psychology, 26, 639–648.

20.

Frick

P. J.

White

S. F.

(2008). Research review: The importance of callous-unemotional traits for developmental models of aggressive and antisocial behavior. Journal of Child Psychology and Psychiatry, 49, 359–375.

21.

Grossmann

Johnson

M. H.

(2007). The development of the social brain in human infancy. European Journal of Neuroscience, 25(4), 909–919.

22.

Hamlin

J. K.

Wynn

(2011). Young infants prefer prosocial to antisocial others. Cognitive Development, 26(1), 30–39. https://dx-doi-org.web.bisu.edu.cn/10.1016/j

23.

Hammond

S. I.

Brownell

C. A.

(2018). Happily unhelpful: Infants’ everyday helping and its connections to early prosocial development. Frontiers in Psychology, 9, 1770.

24.

Hay

D. F.

Cook

K. V

. (2007). The transformation of prosocial behavior from infancy to childhood. In Brownell

C. A.

Kopp

C. B.

(Eds.), Socioemotional development in the toddler years: Transitions and transformations (pp. 100–131). Guilford Press.

25.

Hepach

(2017). Prosocial arousal in children. Child Development Perspectives, 11, 50–55.

26.

Hepach

Vaish

Tomasello

(2012). Young children are intrinsically motivated to see others helped. Psychological Science, 23, 967–972.

27.

Hepach

Vaish

Tomasello

(2013). A new look at children’s prosocial motivation. Infancy, 18, 67–90.

28.

Hepach

Warneken

. (2018). Early development can reveal the foundation of human prosociality. Current Opinion in Psychology, 20, iv–viii.

29.

Hirschberger

. (2010). Compassionate callousness: A terror management perspective on prosocial behavior. In Mikulincer

Shaver

P. R.

(Eds.), Prosocial motives, emotion, and behavior: The better angels of our nature (pp. 201–219). APA.

30.

Hoffman

M. L.

(1991). Is empathy altruistic? Psychological Inquiry, 2 (2) , 131–133.

31.

Hoffman

M. L

. (2000). Empathy and moral development: Implications for caring and justice. Cambridge University Press.

32.

Hornstein

H. A.

(1991). Empathic distress and altruism: Still inseparable. Psychological Inquiry, 2(2), 133–135.

33.

Howes

Parver

(1987). Toddlers’ responses to the distress of their peers. Journal of Applied Developmental Psychology, 8, 441–452.

34.

Laible

Carlo

Murphy

Augustine

Roesch

(2014). Predicting children’s prosocial and co-operative behavior from their temperamental profiles: A person-centered approach. Social Development, 23, 734–752.

35.

Lamm

Batson

C. D.

Decety

(2007). The neural substrate of human empathy: Effects of perspective-taking and cognitive appraisal. Journal of Cognitive Neuroscience, 19, 42–58.

36.

Lin

H.-C.

McFatter

(2012). Empathy and distress: Two distinct but related emotions in response to infant crying. Infant Behavior and Development, 35(4), 887–897.

37.

López-Pérez

Ambrona

(2015). The role of cognitive emotion regulation on the vicarious emotional response. Motivation and Emotion, 39, 299–308.

38.

Malti

Dys

S. P.

(2018). From being nice to being kind: development of prosocial behaviors. Current Opinion in Psychology, 20, 45–49.

39.

Malti

Zuffiano`

Cui

Ongley

S. F.

Peplak

Chaparro

M. P.

Buchmann

. (2016). Children’s sympathy, guilt, and moral reasoning in helping, cooperation, and sharing: A six-year longitudinal study. Child Development, 87, 1783–1795.

40.

Maner

J. K.

Gailliot

M. T.

(2007). Altruism and egoism: Prosocial motivations for helping depend on relationship context. European Journal of Social Psychology, 37(2), 347–358.

41.

Marsh

A. A.

Ambady

Kleck

R. E.

(2005). The effects of fear and anger facial expression on approach- and avoidance-related behaviors. Emotion, 5, 119–124.

42.

Paulus

(2018). The multidimensional nature of early prosocial behavior: A motivational perspective. Current Opinion in Psychology, 20, 111–116.

43.

Paulus

. (2019, in press). Is young children’s helping affected by helpee’s need? Preschoolers, but not infants selectively help need others. Psychological Research.

44.

Pletti

Scheel

Paulus

(2017). Intrinsic altruism or social motivation-what does pupil dilation tell us about children’s helping behavior? Frontiers in Psychology, 8, 2089.

45.

Plötner

Over

Carpenter

Tomasello

(2015). Young children show the bystander effect in helping situations. Psychological Science, 26(4), 499–506.

46.

Rice

J. A.

Levine

L. J.

Pizarro

D. A

. (2007). “Just stop thinking about it”: Effects of emotional disengagement on children’s memory for educational material. Emotion, 7, 812–823.

47.

Roth-Hanania

Davidov

Zahn-Waxler

(2011). Empathy development from 8 to 16 months: Early signs of concern for others. Infant Behavior & Development, 34, 447–458.

48.

Sagi

Hoffman

M. L.

(1976). Empathic distress in the newborn. Developmental Psychology, 12, 175–176.

49.

Silkenbeumer

Schiller

Holodynski

Kärtner

(2016). The role of co-regulation for the development of social-emotional competence. Journal of Self-Regulation and Regulation, 2, 11–26.

50.

Song

J.-H.

Colasante

Malti

(2017). Helping yourself help others: Linking children’s emotion regulation to prosocial behavior through sympathy and trust. Emotion, 18, 518–527.

51.

Svetlova

Nichols

Brownell

(2010). Toddlers’ prosocial behavior: From instrumental to empathic to altruistic helping. Child Development, 81, 1814–1827.

52.

Tully

E. C.

Donohue

M. R.

Garcia

S. E.

(2015). Children’s empathy responses and their understanding of mother’s emotions. Cognition and Emotion, 29, 118–129.

53.

Van de Vondervoort

J. W.

Hamlin

J. K.

(2018). The early emergence of sociomoral evaluation: Infants prefer prosocial others. Current Opinion in Psychology, 20, 77–81. https://dx-doi-org.web.bisu.edu.cn/10.1016/jcopsyc.2017.08.014.

54.

Viding

Kimonis

E. R

. (2018). Callous–unemotional traits. In Patrick

C. J.

(Ed.), Handbook of psychopathy (pp. 144–164). The Guilford Press.

55.

Waller

Hyde

L. W.

(2017). Callous–unemotional behaviors in early childhood: Measurement, meaning, and the influence of parenting. Child Development Perspectives, 11(2), 120–126.

56.

Warneken

Tomasello

(2006). Altruistic helping in human infants and young chimpanzees. Science, 311, 1301–1303. https://doi.org/10.1126/science.1121448

57.

Waugh

W. E.

Brownell

C. A.

(2017). “Help yourself!” What can toddlers’ helping failures tell us about the development of prosocial behavior? Infancy, 22(5), 665–680.

58.

Weisenfeld

A. R.

Whitman

P. B.

Malatesta

C. Z.

(1984). Individual differences among adult women in sensitivity to infants: Evidence in support of an empathy construct. Journal of Personality and Social Psychology, 46, 118–124.

59.

White

B. A.

(2014). Who cares when nobody is watching? Psychopathic traits and empathy in prosocial behaviors. Personality and Individual Differences, 56, 116–121.

60.

Young

Fox

Zahn-Waxler

(1999). The relation between temperament and empathy in 2-year-olds. Developmental Psychology, 35(5), 1189–1197.

61.

Zahn-Waxler

Radke-Yarrow

(1990). The origins of empathic concern. Motivation and Emotion, 2, 107–130.

62.

Zebrowitz

L. A.

Fellous

J. M.

Mignault

Andreoletti

(2003). Trait impressions as overgeneralized responses to adaptively significant facial qualities: Evidence from connectionist modeling. Personality and Social Psychology Review, 7, 194–215.

63.

Zebrowitz

L. A.

Montepare

J. M.

(2008). Social psychological face perception: Why appearance matters. Social Personality Psychology Compass, 2(3), 1497–1517.