Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Aug 1.
Published in final edited form as: Cogn Emot. 2022 Mar 23;36(5):805–820. doi: 10.1080/02699931.2022.2054779

Amenders and Avoiders: An Examination of Guilt and Shame for Toddlers and Their Older Siblings

Amy M Kolak 1, Brenda L Volling 2
PMCID: PMC9458630  NIHMSID: NIHMS1794826  PMID: 35319341

Abstract

Guilt- and shame-prone responding were examined in a sample of 146 18-month-old toddlers and their older siblings (M = 49.5 months, SD = 10.4) during mishap tasks which were used to differentiate both toddlers and their older siblings into Amenders (low avoidance) and Avoiders (high avoidance). Toddlers and older siblings classified as Amenders expressed more concern and were less distressed by the mishap than Avoiders. Children were divided into four groups: Amender-Amender (older sibling-toddler), Amender-Avoider, Avoider-Avoider, and Avoider-Amender to examine differences in sibling interaction and moral development. Older siblings in the Avoider-Avoider group were significantly more aggressive and less empathic toward toddlers than older siblings in the Avoider-Amender group. Toddlers in the Amender-Amender, Amender-Avoider, and Avoider-Amender groups showed significant gains in moral regulation from 18 to 24 months whereas toddlers in the Avoider-Avoider group did not. In contrast, while older siblings were generally high on moral regulation when toddlers were 18 months, this was not the case for older siblings in the Avoider-Avoider group, who had lower moral regulation scores that significantly increased over time. Findings are discussed with respect to the significance of sibling socialization for toddlers’ developing moral sensibility.

Keywords: guilt, shame, toddlers, preschoolers, siblings, prosocial behavior

Shame and guilt reflect distinct, albeit interrelated, self-conscious emotional responses to ones’ transgressions or failures. Individuals often experience and display shame or guilt when their behavior does not measure up to social expectations (Barrett & Nelson-Goens, 1997; Tangney, 1998; Tangney et al., 2009). These emotions stem from divergent thoughts about the self, such that shame involves a negative global evaluation of the self (i.e., “I am a bad person.”), whereas guilt reflects one’s negative evaluation of a specific behavior (i.e., “I did a bad thing.”). These differences in self-evaluations have ramifications for emotional and behavioral reactions (Lewis, 1971). Tangney and colleagues (2009) suggested that shame elicits feelings of smallness or worthlessness and accordingly, young children are inclined to “hide’ by averting their gaze away from others when they believe they have misbehaved (Barrett et al., 1993) or by showing body or facial collapse (Lewis et al., 1992). Conversely, guilt may be expressed through feelings of regret or concern that motivate reparative actions, even in children as young as 17 months (Barrett, 2005). The purpose of the present study was to examine the role that shame and guilt among 18-month-old toddlers and their older preschool-aged siblings played in sibling socialization and early morality. To do so, this study addressed the following three aims which were to 1) examine toddlers’ and older siblings’ guilt and shame responses to a mishap task, classifying them as either Amenders or Avoiders, 2) assess whether sibling dyads comprised of two Amenders, two Avoiders, or an Amender and an Avoider had qualitatively different sibling interactions (i.e., more empathy, less conflict), and 3) explore how such differences might give rise to different trajectories of moral development for toddlers and older siblings.

Guilt- and Shame-Prone Responding in Early Childhood

Mishap tasks, where others’ prized objects are rigged to break or get damaged, are widely-used laboratory paradigms employed to assess young children’s guilt- and shame-prone responses (Barrett et al., 1993; Cole et al., 1992; Kochanska et al., 1995; Kochanska et al., 2002). During mishap tasks, children are given a toy that has been rigged to break when left alone to play with it, leading children to believe they may be responsible for the damage. Children express a range of emotional (e.g., avoidance, tension, distress) and behavioral responses (e.g., concern, reparation, comments) to these manipulated transgressions. Based on young children’s avoidance (i.e., looking away) of the experimenter, Barrett and colleagues (1993) divided 2- to 3-year-old toddlers into Avoiders, those who were high on avoidance and Amenders, those who were low on avoidance. Amenders were more likely to tell the experimenter about the mishap and try to repair the toy, consistent with a guilt orientation than Avoiders, who displayed more discomfort consistent with a pattern of shame. Similarly, Drummond et al. (2017) recently classified 18-month toddlers as low-avoidant (i.e., guilt-prone Amenders) and high-avoidant (i.e., shame-prone Avoiders) and found that Amenders tried to repair the broken toy sooner and more frequently than Avoiders, whereas Avoiders used more gaze and bodily avoidance than Amenders. Collectively, studies have shown that very young children can be grouped as Amenders or Avoiders based on their affect and behavior during standard mishap tasks, which reflect patterns of guilt and shame, respectively (Lagattuta & Thomspon, 2007).

To address the first aim and create groups of Amenders and Avoiders based on their avoidance, 18-month-old toddlers and their older siblings were observed in separate mishap tasks to examine guilt- and shame-prone responding and were then classified as either Amenders or Avoiders, with Amenders low on avoidance and Avoiders high on avoidance. In line with previous research, we hypothesized that guilt-prone Amenders would express more concern for the broken object and make more reparative attempts than shame-prone Avoiders, who in addition to displaying more gaze avoidance, should display more affective distress and bodily tensions after the mishap.

Toddlers and Older Siblings as Amenders and Avoiders

Having information on both toddlers and older siblings also allowed us to examine developmental (sibling) differences in shame- and guilt-prone responding in the mishap task. Longitudinal research conducted when children were 22, 33, and 45 months of age documented several developmental differences in children’s distress responses following mishaps with children displaying significantly more bodily tension following the mishap at 45 months than when they were 22 and 33 months (Kochanska et al., 2002). At 22 months, children engaged in more gaze avoidance, experienced more distress, expressed more negative affect and less positive affect than they did at 33 or 45 months (Kochanska et al., 2002). Similarly, Walter and LaFreniere’s (2007) cross-sectional study of 2-, 3-, and 4- year-olds found that 2-year-olds displayed significantly more avoidance following the mishap than 4-year-olds. Therefore, we hypothesized that toddlers would use more gaze avoidance and be more distressed than their older siblings, with older siblings using more reparation than their toddler siblings. If this were the case, older siblings may serve as role models for toddlers and act as socializers in the development of the self-conscious emotions of guilt and shame for young children. Dunn and Munn (1986a) found that older siblings’ cooperative behavior with their 18-month-old siblings predicted toddlers’ conciliatory behavior (i.e., actions intended to deescalate conflict) and cooperation during sibling interactions six months later. Thus, toddlers may be learning prosocial and reparative behaviors from their older siblings. To that end, a recent study reported significant correlations between parental reports of toddlers’ and older siblings’ guilt (Scrimgeour et al., 2017). If toddlers are modeling their older siblings or older siblings are teaching toddlers about rule-compatible behavior (van Berkel et al., 2020), we would expect some correspondence between toddlers’ and older siblings’ guilt- and shame-prone responses during the mishap tasks. The second aim, then, was to examine the association between the Avoider and Amender status of toddlers and their older siblings, hypothesizing a greater likelihood of toddlers being classified as Amenders if their older siblings were also Amenders, and Avoiders if the older siblings were Avoiders.

Do Sibling Interactions Differ for Amenders and Avoiders?

To further examine the potential role of sibling socialization, children were divided into four sibling groups: (a) Amender (older sibling)-Amender (toddler), (b) Amender-Avoider, (c) Avoider-Avoider, and (d) Avoider-Amender. In this way, we could examine whether patterns of sibling interaction differed across these dyads that might indicate differences in sibling socialization. Toddlers develop prosocial behaviors, empathic concern, and emotional understanding in the context of cooperative sibling interaction, suggesting that the development of self-conscious emotions, such as guilt, may be fostered during cooperative sibling interactions (Dunn, 1988; Dunn & Munn, 1986a). Conversely, conflict between siblings was negatively associated with toddlers’ conciliatory behavior from 18 to 24 months, which may indicate that toddlers in high conflict sibling relationships are more prone to shame than guilt (Dunn & Munn, 1986b). Older siblings often direct and manage sibling interactions more than younger siblings (Brody et al., 1985), so toddlers may very well benefit when their older siblings who are Amenders are directing more cooperative and helpful behaviors toward them than when they have older siblings who are Avoiders. Though this top-down, unidirectional approach to sibling socialization, where older siblings influence younger siblings, has dominated the field, recent research tested a reciprocal, bidirectional model of sibling influence and found, as expected, that older siblings’ empathy predicted the younger siblings’ empathic concern across an 18-month period in early childhood. But they also reported that the younger siblings’ empathy predicted the empathic concern of the older siblings (Jambon et al., 2019). Because sibling relationships are dyadic, dividing siblings into dyads based on each sibling’s Amender or Avoider classification allowed us to simultaneously consider both older and younger siblings jointly in our analyses (e.g., Amender-Amender, Avoider-Avoider) and then determine if sibling interaction and children’s moral development differed across these dyadic pairings.

The sibling relationship is an important aspect of toddlers’ and preschoolers’ social worlds because it provides young children with opportunities to interact regularly with other social partners besides their parents. Sibling relationships during early childhood may be especially critical for the development of moral emotions and reparative behaviors, as prior research has found that early sibling interactions played a role in young children’s social-emotional understanding and cooperative behavior (Dunn, 1988; Dunn & Munn, 1986a). In addition, early childhood is a period witnessing significant growth in conscience development (Kochanska & Aksan, 2004), self-regulation (Perry, 2019), and social understanding (Wellman & Liu, 2004). Dunn and Munn (1986a) also found that between 18 and 24 months of age, helping, sharing, and cooperative behaviors during sibling interaction increased for both toddlers and their older siblings, and individual differences in children’s social understanding were positively associated with cooperative sibling interactions (Dunn et al., 1996). Thus, the burgeoning sibling relationship may have implications for the early development of self-conscious emotions, such as guilt and shame.

Given previous research, we hypothesized that sibling interactions would include more teaching, empathy, and positive involvement and less aggression and conflict/rivalry when toddlers and older siblings were both classified as Amenders than when both were Avoiders, and that sibling interactions would improve and be more positive and less negative from 18 to 24 months when both siblings were Amenders. Because older siblings often direct, manage, and lead sibling interactions (Brody et al., 1985), whether the older sibling is an Amender or Avoider may have a stronger effect on determining sibling relationship quality. Thus, one might expect that older siblings as Amenders would be more positive and less negative in their interactions with the toddlers, whereas older siblings as Avoiders may instigate more sibling conflict with their toddler siblings. It is less clear how siblings would interact with one another in Amender-Avoider and Avoider-Amender dyads; however, in light of Jambon and colleagues’ (2019) study supporting the reciprocal influence that siblings have on each other, toddlers’ concern and reparation or lack thereof may also elicit positive or negative behavior from their older siblings.

Self-Conscious Emotions and the Development of Early Moral Behavior

The final aim of this study was to examine whether there was evidence of growth in children’s moral sense, especially for toddlers between the ages of 18 and 24 months, when siblings were either Amenders or Avoiders. The period between 18 and 24 months is a significant time frame for the development of be self-regulation and the emergence of conscience and moral sensibility (Brownell & Kopp, 2007; Thompson & Newton, 2010). Guilt and shame involve different self-evaluative processes and behavioral responses to mishaps, and these self-conscious emotions have been related to children’s interpersonal relationships and emotional adjustment (Tangney, 1995). For instance, in a review of research on the correlates of self-conscious emotions, Muris and Meesters (2014) concluded that guilt, considered a reflection of adaptive functioning, was inversely related to children’s aggression, and shame was positively associated with depression, anger, and aggression for 5- to 18-year-old children. Less is known about developmental outcomes for guilt- and shame-prone children during toddlerhood and preschool. Even so, Walter and LaFreniere (2007) found that avoidance during mishaps among 2- to 4-year-old children was inversely related to teachers’ reports of social competence, prosocial behavior, tolerance, and joyfulness, and negatively associated with parental reports of moral regulation, an indicator of conscience development. Finally, Drummond et al. (2017) found that toddlers who exhibited more guilt (i.e., less avoidance, making reparations) following a mishap were more likely to engage in prosocial behavior, specifically empathic helping directed toward an experimenter who feigned emotional distress, than shame-prone children. Thus, a guilt-prone orientation (i.e., being an Amender) may motivate prosocial behavior whereas a shame-prone orientation (i.e., being an Avoider) may inhibit this type of behavior even in young children.

In the current investigation, we considered whether children in Amender/Avoider sibling dyads would differ with respect to the development of conscience, specifically affective discomfort (e.g., guilt/remorse) and moral regulation (e.g., reparation; Kochanska et al., 1994). According to Kochanska and colleagues (1994), affective discomfort taps arousal and distress as well as guilt following a transgression, whereas moral regulation assesses children’s abilities to refrain from wrongdoing and partake in desirable acts, such as reparation. Because older siblings are often socializers of their younger siblings’ social competence and self-regulation (Sawyer et al., 2002; Tucker et al., 1999; van Berkel et al., 2020), toddlers may be particularly sensitive to sibling interactions with an older sibling who is an Amender or Avoider given that moral emotions and prosocial behavior begin to emerge during the second year of life (Barrett, 2005; Lagattuta & Thompson, 2007). Therefore, we hypothesized that toddlers with Amender older siblings (i.e., those in Amender-Amender and Amender-Avoider dyads) would experience significant gains in conscience development from 18 to 24 months compared to toddlers with Avoider older siblings (i.e., those in Avoider-Avoider and Avoider-Amender dyads). Given the reciprocal nature of sibling interactions, it is also possible that older siblings may show differential gains in conscience development based on whether toddlers are Avoiders or Amenders, but because of the sheer lack of empirical work on siblings and self-conscious emotions, we do not advance any directional hypotheses at this time.

The Current Study

The current study examined the affective and behavioral responses of 18-month-old toddlers and their older siblings with three aims: (1) to identify guilt- (Amenders) and shame-prone (Avoiders) children using a laboratory mishap task and (2) examine the association between Amender and Avoider classifications across toddlers and older siblings, as well as differences in sibling interactions (i.e., rivalry, empathy), in line with a sibling socialization hypothesis, and finally, (3) to determine whether there was differential growth in toddlers’ and older siblings’ conscience (affective discomfort, moral regulation) from 18 to 24 months depending on whether they were Amenders or Avoiders.

Method

Participants

A total of 146 families (mothers, fathers, and 2 children) from a Midwestern city and the surrounding suburbs participated. The sibling pairs were comprised of 18-month-old toddlers (66 girls, 80 boys) and their older siblings (85 girls, 61 boys) who ranged in age from approximately 2½ to 7 years (M = 4.13; SD = 0.87). The sample was predominately European American with 86% of parents identifying as White, 5% African-American, 4% Asian-American, 3% Hispanic, and 2% Other. Parents were college-educated; over 86% of mothers and almost 79% of fathers reported having a Bachelor’s degree or higher. With respect to household income, 1% of the sample earned less than 20,000 a year, 25% earned between 20,000 and 59,999, 38% earned between 60,000 and 99,999 and the remaining 36% reported earning over 100,000.

Study Design

Children were part of a larger longitudinal investigation of changes in child and family functioning following the birth of the second child. The 241 families consisting of mother, father, and older sibling were recruited for Phase I of the investigation during the third trimester of the mother’s pregnancy with the second child and assessed at 1, 4, 8, and 12 months after this child’s birth. Phase II was conducted when second-born children were 18, 24, and 36 months, and data for the present report are from the 18- and 24-month timepoints. The mishap paradigm was only conducted at 18 months, but for both siblings. Mothers and fathers reported on children’s sibling interactions and moral development at 18 and 24 months.

Families who participated in the 18-month timepoint did not differ on demographic characteristics from the 241 families initially recruited for Phase I. Families who declined participation at 18 months often cited time constraints, with others unable to participate because they had moved or no longer met the eligibility requirements (i.e., parents had separated or divorced). Of the 146 families who participated at 18 months, 15 toddlers and 3 older siblings did not complete the mishap task, leaving 128 siblings for the paired analyses. The gender constellation of the 128 sibling dyads (older sibling-toddler) included 29 girl-girl, 47 girl-boy, 28 boy-boy, and 24 boy-girl dyads.

Procedure and Measures

Children were visited in their homes and participated in laboratory visits when toddlers were 18 and 24 months, and both parents completed questionnaires assessing children’s sibling interactions and conscience development. The mishap task was conducted separately with toddlers and older siblings in different rooms during the laboratory visit. Both playrooms were equipped with one-way mirrors and video recording equipment. The parent stayed in the room with the toddler and the experimenter for the duration of the assessment while the older sibling was escorted to a different room by another experimenter. The following assessments were completed for toddlers: 1) self-awareness task, 2) effortful control battery, and 3) mishap task, with some slight variation for older siblings: 1) effortful control battery, 2) emotion understanding, and 3) mishap task. Upon completion, the older sibling was reunited with the toddler and the parent, and both siblings participated in one final task together, the fishing game, to assess sibling sharing. All laboratory procedures were video recorded and later coded; however, the mishap task was the only observational data utilized for this report.

Mishap task.

The mishap task, as described by Barrett (1995), was the source of information on children’s guilt and shame and formed the basis for creating the Amender and Avoider groups. The experimenter (E) brought in a teddy bear that she introduced as her “special” bear and went on to enthusiastically show the child some of the bear’s unique features (i.e., shoelaces and zippered vest). After emphasizing how special the teddy bear was and the fun ways the child could play with it, E said she had to go get something in the other room. As she left the room, she told the child that they could play with the bear while she was gone, but “to please take good care of my favorite bear.” The bear’s arm had been altered such that it would fall off when the child played with it. One minute after the arm fell off, E returned to the room and asked “How’d it go?” At 2 minutes, E commented about the arm coming off (e.g., “It looks like my teddy bear’s arm fell off.”) and asked the child, matter-of-factly, “How do you think that happened?” At 3 minutes, E said with an empathic/concerned tone, “That’s too bad my teddy bear’s arm came off.” The mishap procedure ended a minute later (4 minutes after the arm fell off) when E reassured the child that they did not break the bear (e.g., “Lots of kids play with these toys and sometimes they break.”) and indicated that it could be fixed (e.g., “We can sew it up and it will be as good as new.”). Once the child had been assured that the bear could be fixed, E got out some other toys for the child and let the child play for 5 – 10 minutes to alleviate any discomfort that the mishap task may have caused.

The same mishap procedure was used for toddlers. The parent, who was in the room with the toddler, was asked to limit their interactions with the child throughout the assessment and specific instructions were given about what they should do during the mishap procedure. If the toddler showed the broken arm to the parent, parents were told to simply respond, matter-of-factly, with “Oh, it looks like the arm came off” and then resume the activity (i.e., reading a magazine, completing questionnaires) they had been doing.

Coding of guilt and shame behaviors.

Toddlers’ and older siblings’ responses to the 4-minute mishap procedure conducted at the 18- month timepoint were coded from video recordings using a coding scheme adapted from Kochanska and her colleagues (Kochanska et al., 1995; Kochanska et al., 2002). Facial tensions, handling of the broken object, bringing object to E (after E returns to room), attempting to repair the object, and commenting about the mishap were coded every 5 seconds as absent (0) or present (1). Total scores were created for each of these variables by summing the number of times this behavior occurred across the 48 intervals (36 intervals for bringing object to E). Bodily tensions (e.g., turning back to or moving away from E, covering face with hands, squirming or twisting unnaturally, hunching shoulders, rubbing hands or fingers together or playing with object in a manner that indicates tension) were also coded every 5 seconds using a 3-point scale where 0 = no signs of bodily tension, 1 = 1 sign of bodily tension, or 2 = 2 or more signs of bodily tension or 1 extreme sign of bodily tension and scores were then summed across all intervals to create a total score for bodily tensions. Gaze avoidance, which was coded when the child looked away, downward, or askance, was also coded during every 5-second interval and was coded as 0 = no gaze avoidance; 1 = brief glances away, downward, or askance, or if child averted gaze for less than 5 seconds; or 2 = long, if child averted gaze for more than 5 seconds (Kochanska et al., 1995; 2002). Scores were summed across all intervals to create a total gaze avoidance score and this score was used to distinguish between guilt-prone (Amenders) and shame-prone (Avoiders) toddlers and older siblings (Barrett et al., 1993).

Children’s overall distress and affect were also coded for each of the four 1-minute episodes (Kochanska et al., 2002). Children’s overall distress was coded on a scale from 1 to 4 where 1 = child is unaffected by the mishap, 2 = child briefly/mildly affected by the mishap, 3 = child is affected by the mishap, as evidenced by stilling, unease, and/or concern and 4 = child is strongly affected, may freeze or cry and be very uncomfortable or uneasy throughout the episode. Ratings were summed across the four episodes to create one score reflecting overall distress. Affect was coded as either negative, neutral, or positive for each of the four episodes and if it was negative or positive it was then coded as weak (and received a score of 1) or strong (and received a score of 2). A negative affect score was created by summing the weak and strong negative codes whereas the positive affect score was created by summing the weak and strong positive codes (scores could range from 0 – 8). Nineteen percent of the sample was coded by two coders (n = 53) for reliability purposes; Kappas ranged from .81 to .99 (M = .91).

Sibling relationship quality at 18 and 24 months.

Mothers and fathers completed questions about children’s sibling interactions at 18 and 24 months. Due to the age differences between toddlers and older siblings, age-appropriate measures were used to assess positive and negative behavior children exhibited toward their sibling.

For toddlers, mothers and fathers completed the 18-item Sibling Relationships in Early Childhood Questionnaire (SREC; Volling & Elins, 1998). Items were rated from 1 = never to 5 = always and composited to create 3 subscales: positive involvement (8 items, α = .74 and .75 for mothers and .78 and .79 for fathers at 18 and 24 months, respectively), conflict (7 items, α = .73 and .71 for mothers and .80 and .76 for fathers), and avoidance (3 items, α = .53 and .70 for mothers and .66 and .74 for fathers). Given the low reliabilities for avoidance at 18 months, avoidance was dropped. Mothers’ and fathers’ subscales were significantly correlated for positive involvement (18 months: r = .37, p < .001; 24 months: r = .37, p < .001) and conflict (18 months: r = .46, p < .001; 24 months: r = .40, p < .001), and averaged at each timepoint to create robust composites.

For older siblings, parents completed the Sibling Inventory of Behavior (SIB; Hetherington & Clingempeel, 1992; Schaefer & Edgerton, 1981). The 32 items were rated from 1 = never to 5 = always and combined to create 6 subscales: rivalry (7 items, α = .74 and .73 for mothers and .75 and .70 for fathers at 18 and 24 months), aggression (5 items, α = .61 and .79 for mothers and .81 and .73 for fathers), avoidance (5 items, α = .49 and .54 for mothers and .60 and .58 for fathers), involvement (6 items, α = .72 and .78 for mothers and .84 and .80 for fathers), empathy (5 items, α = .82 and .80 for mothers and .85 and .79 for fathers), and teaching (4 items, α = .80 and .73 for mothers and .76 and .77 for fathers). Given the low reliabilities (ranging from .49 to .60) for the avoidance subscales, avoidance was dropped from further analyses. At 18 and 24 months, mothers’ and fathers’ subscales were significantly correlated with r coefficients ranging from .28 to .49 (all p’s < .001) and averaged.

Moral development at 18 and 24 months.

For toddlers and older siblings, mothers and fathers completed 88 items from the My Child measure (Kochanska et al., 1994). Items, rated from 1 = extremely untrue, not at all characteristic to 7 = extremely true, very characteristic, were combined to create 8 subscales that mapped onto 2 composites, affective discomfort and moral regulation. Kochanska et al. (1994) define affective discomfort as the “emotional consequences of wrongdoing and affective responses to others” (p. 862), and was comprised of the scales, guilt/remorse, apology, concern over good feelings of the parent, and empathy (α = .68 to .80), whereas moral regulation “encompasses the child’s concerns about wrongdoing itself rather than the interpersonal consequences” (pp. 862–863), and was comprised of the scales, confession, reparation, internalized conduct, and concerns about others’ wrongdoing (α = .70 to .92), At 18 and 24 months, mothers’ and fathers’ reports for both siblings were significantly correlated, with r values ranging from .28 to .48 (all p’s < .001). To reduce single-reporter bias, mothers’ and fathers’ scores were averaged within each timepoint for each sibling.

Results

Distinguishing Guilt- and Shame-Prone Responding

Table 1 presents the descriptive statistics and correlations for toddlers’ and older siblings’ responses to the mishap. To address the first aim, we created Amender and Avoider groups using median splits based on the frequency of gaze avoidance observed during the mishap; toddlers’ median = 15.00 and older siblings’ median = 8.00; a procedure used repeatedly in prior research on young children’s guilt and shame (Barrett et al., 1993; Drummond et al., 2017).

Table 1.

Descriptive Statistics and Correlations for Toddlers’ and Older Siblings’ Responses to the Mishap at the 18-Month Timepoint

1. 2. 3. 4. 5. 6. 7. 8. 9.
1. Gaze Avoidance .11 .59*** .46*** .53** −.37*** −.56*** −.21* −.36*** −.29***
2. Bodily Tension .58*** −.06 .58*** .36*** −.22** −.63*** −.19* −.30*** −.25**
3. Overall Response .40*** .58*** .02 .48*** −.30*** −.47*** −.10 −.10 .04
4. Negative Affect .26** .35*** .45*** −.02 −.43*** −.29*** −.18* −.16 −.12
5. Positive Affect −.18* −.17 −.18* −.36*** −.01 .25** .10 .11 .07
6. Handles Object −.38*** −.31*** −.06 −.22* .03 −.16 .23** .23** .30***
7. Brings/Points to Object −.22* −.11 .05 −.18* .09 .31*** .16 .24** .43***
8. Repairs Object −.15 −.03 −.22* −.02 −.01 .25** .22* .08 .38***
9. Comments about Mishap −.06 −.08 .32*** −.04 .09 .41*** .26** .17* −.03
Toddlers
 Mean 16.00 18.23 8.80 1.44 .86 19.51 1.06 .89 4.98
 SD 7.92 11.62 2.13 1.72 1.23 13.01 1.46 1.42 5.07
Older Siblings
 Mean 10.49 17.20 9.59 .86 1.31 33.52 3.14 6.67 11.23
 SD 8.85 13.73 2.15 1.15 1.75 11.97 2.82 5.13 6.23
t(127) 5.57*** 1.00 −2.95** 3.13** −2.41*** −8.34*** −7.94*** −12.47*** −8.66***
Open in a new tab
*

p < .05

**

p < .01

***

p < .001.

Note. Correlations among toddlers’ responses are below the diagonal. Correlations among older siblings’ responses are above the diagonal. Correlations across toddlers’ and older siblings’ responses are bolded and underscored on the diagonal. Paired t-tests were conducted to examine mean differences between toddlers’ and older siblings’ responses to the mishap.

Toddlers.

Low avoidance Amenders (n = 68; 35 girls and 33 boys) and high avoidance Avoiders (n = 63, 24 girls and 39 boys) groups were not related to toddler gender, χ2(1, N = 131) = 2.36, p = .12. A single-factor MANOVA with group as the between-subjects factor using the other observed behaviors during the mishap task revealed group differences in line with guilt- and shame-prone responding. A significant multivariate effect was found for group, Pillai’s Trace = .25, F(8, 122) = 5.20, p < .001, η2 = .25. Results from univariate ANOVAs are presented in Table 2 and show that toddlers in the Amenders group displayed significantly fewer bodily tensions, were less distressed overall by the mishap, exhibited less negative affect, handled the bear more frequently, and brought the bear to E more often than toddlers in the Avoiders group. These findings replicate previous research with 17-month-old toddlers (see Barrett et al., 1993), in which two distinct patterns of guilt (low avoidance) and shame (high avoidance) were found during the mishap.

Table 2.

Univariate ANOVA Results for Toddlers’ and Older Siblings’ Emotional and Behavioral Responses to the Mishap for Low Avoidance and High Avoidance Groups

Toddlers Older Siblings
Low Avoidance
(n = 68)		 High Avoidance
(n = 63)		 Low Avoidance
(n = 73)		 High Avoidance
(n = 70)
M SD M SD F(1, 129) η2 M SD M SD F(1, 141) η2
Bodily Tension 14.04 10.18 24.06 10.67 30.23*** .19 10.32 10.85 23.74 13.42 43.65*** .24
Overall Response 8.34 1.93 9.38 2.21 8.31** .06 8.67 1.82 10.49 2.15 30.76*** .18
Negative Affect 1.16 1.43 1.78 1.96 4.26* .03 .32 .68 1.43 1.26 43.44*** .24
Positive Affect 1.03 1.34 .65 1.06 ns .02 1.88 2.05 .77 1.69 15.49*** .10
Handles 23.49 13.91 14.78 10.28 16.41*** .11 39.42 8.34 27.26 11.75 51.92*** .27
Brings/Points 1.29 1.71 .79 1.05 3.97* .03 3.85 3.21 2.53 2.86 7.86** .05
Repairs 1.10 1.60 .70 1.47 ns .02 8.70 5.89 4.97 4.41 18.09*** .11
Comments 5.50 5.41 4.35 4.56 ns .01 12.73 6.33 9.16 5.48 12.82*** .08
Guilt/Shame .24 .44 −.27 .42 45.81*** .26 .41 .40 −.42 .48 129.35*** .48
Open in a new tab
*

p < .05

**

p < .01

***

p < .001.

Note. Means are bolded in the table to highlight the larger value within each pair of significant findings. The analyses for older siblings controlled for older siblings’ age at the 18-month timepoint laboratory visit. Guilt/shame is a composite variable that was created for each sibling by averaging across standardized behavioral indicators of guilt (i.e., positive affect, handles, brings/points, repairs, and comments) and reverse-coded, standardized indicators of shame (i.e., gaze avoidance, bodily tension, overall response, and negative affect). High scores on this composite reflect more guilt-proneness whereas low scores reflect more shame-proneness.

Older siblings.

Similarly, low avoidance Amenders (n = 73; 44 girls and 29 boys) and high avoidance Avoiders (n = 70, 40 girls and 30 boys) groups were unrelated to gender of the older sibling, χ2(1, N = 143) = .15, p = .70, and did not differ by age, t(141) = .23, p = .82. A single-factor MANCOVA, with group as the between-subjects factor, and older siblings’ age as a covariate, was conducted on the remaining behaviors observed during the mishap task, with a significant multivariate effect for group, Pillai’s Trace = .45, F(8, 133) = 13.65, p < .001, η2 = .45. Univariate ANOVA results shown in Table 2 indicate that older siblings in the Amenders group displayed significantly fewer bodily tensions, were less distressed overall by the mishap, exhibited more positive affect and less negative affect, attempted to repair the bear more frequently, brought the bear to E more often, made more comments to E about the mishap, and handled the bear more frequently following the mishap than older siblings in the Avoiders group, again replicating patterns of guilt- and shame-responding consistent with prior research (Barrett et al., 1993; Drummond et al., 2017).

As an additional means to validate differences across Amender and Avoider classifications, overall composites of guilt/shame were created for each sibling by averaging across standardized behavioral indicators of guilt (i.e., positive affect, handles, brings/points, repairs, and comments) and reverse-coded, standardized indicators of shame (i.e., gaze avoidance, bodily tension, overall response, and negative affect) in the mishap task. High scores on this composite reflect more guilt-proneness whereas low scores reflect more shame-proneness. As shown in Table 2, univariate ANOVAS indicated that toddlers and older siblings in the low avoidance (i.e., guilt-prone) groups had significantly higher scores on the guilt/shame composite than those in the high avoidance (i.e., shame-prone) groups. In the remainder of this paper, we will refer to toddlers and older siblings in the low avoidance groups as Amenders and those in the high avoidance groups as Avoiders (Barrett et al., 1993).

Older Siblings as Socializers of Toddlers’ Self-Conscious Emotions

To address the second aim which was to assess sibling socialization, we first conducted t-tests to examine developmental (sibling) differences in toddlers’ and older siblings’ responses to the mishap task, expecting that older siblings would engage in more of the behaviors, including reparation attempts, than toddlers. As reported in Table 1, paired samples t-tests showed that older siblings displayed significantly less gaze avoidance and negative affect, more positive affect, and appeared to be more distressed by the mishap than toddlers, and they were also significantly more likely to handle and try to repair the broken object as well as bring, point out, or comment about the broken object to the experimenter than toddlers. Thus, older siblings were doing more of these behaviors than toddlers and could be acting as models for their toddler-aged siblings.

We next examined the association between Amender and Avoider classifications across toddlers and older siblings to see if there was a significant association, which one might expect if older siblings are acting as socializers (i.e., Amender toddlers should be more likely to have Amender older siblings than Avoider older siblings). A chi-square analysis was not significant, χ2(1, N = 128) = 1.94, p = .16. We proceeded to create four sibling groups based on their Amender and Avoider status (older sibling-toddler) in an effort to test the reciprocal model of sibling influence whereby one can examine the effects of the younger as well as the older siblings: (1) Amender-Amender (n = 39), (2) Amender-Avoider (n = 28), (3) Avoider-Avoider (n = 33), and (4) Avoider-Amender (n = 28) .There was no association between the Amender/Avoider groups and gender constellation of the sibling dyad, χ2(9, N = 128) = 12.81, p = .17, and a one-way ANOVA revealed no differences for older siblings’ age, F(3, 124) = .05, p = .98.

Does it Matter if your Sibling is an Amender or an Avoider?

To further examine the possibility of sibling socialization, we turned to parent reports of sibling interaction. Table 3 presents correlations and descriptive information for these variables for toddlers and older siblings. Preliminary analyses were conducted to examine whether the age of the older sibling or the gender constellation (e.g., sister-sister, sister-brother) of the sibling dyad were related to parent reports of sibling interactions. Correlations revealed several significant relations between older sibling age and sibling behavior and one-way ANOVAS showed differences in sibling interactions based on the gender constellation (see Table S1 and Table S2 in Supplemental materials for a summary of these analyses). As such, both older sibling age and gender constellation were included as covariates in subsequent analyses.

Table 3.

Descriptive Statistics and Correlations for Sibling Interaction and Moral Development Within and Across the 18- and 24-Month Timepoints

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.
1. T PI .59 *** −.20* −.08 −.17 .59*** .51*** .61*** .40*** .35*** .23* .17
2. T Con .03 .58 *** .42*** .60*** −.27** −.28*** −.19* .01 −.14 −.07 −.15
3. OS Riv −.05 .50*** .63*** .58*** −.16 −.07 .14 .07 −.10 .06 −.08
4. OS Agg −.03 .71*** .65*** .65 *** −.29*** −.32*** −.17 −.02 −.11 −.18 −.22*
5. OS Inv .57*** −.12 −.12 −.19* .69*** .59*** .60*** .26** .27** .25** .26**
6. OS Emp .51*** −.17* −.17* −.32*** .66*** .67 *** .69*** .24* .23* .35*** .25**
7. OS Teach .49*** −.19* −.04 −.25** .64*** .70*** .64 *** .27** .28** .35*** .30***
8. T AD .39*** .03 .10 .00 .18* .24** .29*** .63 *** .74*** .37*** .26**
9. T MR .41*** −.00 −.04 −.03 .25** .26*** .28*** .77*** .73 *** .26** .33***
10. OS AD .27*** −.09 −.06 −.24** .22** .42*** .42*** .42*** .34*** .82 *** .64***
11. OS MR .25** −.10 −.06 −.25** .29*** .46*** .45*** .25** .37*** .63*** .82 ***
18 Months
 Mean 3.86 2.26 2.48 2.59 3.53 3.87 3.22 3.71 3.59 4.64 4.52
 SD .39 .49 .49 .42 .48 .52 .60 .54 .58 .56 .51
24 Months
 Mean 3.77 2.51 2.73 2.75 3.58 3.86 3.40 3.85 3.78 4.70 4.54
 SD .40 .45 .47 .49 .45 .48 .54 .61 .58 .53 .54
Open in a new tab
*

p < .05

**

p < .01

***

p < .001.

Note. Correlations among variables at the 18-month timepoint are below the diagonal. Correlations among variables at the 24-month timepoint are above the diagonal. Correlations across timepoints are bolded and underscored on the diagonal. T = toddlers, OS = older siblings, PI = positive involvement, Con = conflict, Riv = rivalry, Agg = aggression, Inv = involvement, Emp = empathy, Teach = teaching, AD = affective discomfort, MR= moral regulation.

A series of 4 (group: Avoider/Amender) x 2 (time: 18 and 24) repeated measures ANCOVAs (controlling for age of the older sibling and gender constellation), with Amender/Avoider group as a between-subjects factor and time as a repeated factor, were conducted with sibling behaviors as dependent variables. Post-hoc pairwise comparisons with the Dunn-Sidak correction for multiple comparisons were used.

Older siblings’ behavior.

There was a significant main effect of Amender/Avoider group for older siblings’ rivalry, aggression and empathy (see Table 4). Older siblings in the Avoider-Avoider dyad were significantly more aggressive and less empathic toward toddlers than older siblings in the Avoider-Amender dyads; however, pairwise comparisons did not show significant differences in rivalry across Amender/Avoider groups. No significant effects were found for time or the time x group interactions.

Table 4.

Means and Standard Errors (in Parentheses) for Sibling Interaction Variables for Amender/Avoider Dyads (Older Sibling-Toddler)

Amender-Amender Amender-Avoider Avoider-Avoider Avoider-Amender F(3, 101) η2
Toddlers
 Positive Involvement  3.83 (.06)  3.76 (.07)  3.78 (.07)  3.97 (.07)  1.97 .06
 Conflict  2.35 (.07)  2.46 (.08)  2.44 (.08)  2.36 (.09)  .71 .01
Older Siblings
 Rivalry  2.51 (.08)  2.69 (.08)  2.68 (.08)  2.41 (.09)  2.72* .08
 Aggression  2.60 (.08)  2.73 (.08)  2.82 (.08)a  2.50 (.09)b  2.83* .08
 Involvement  3.54 (.08)  3.57 (.08)  3.48 (.08)  3.72 (.09)  1.37 .04
 Empathy  3.89 (.08)  3.91 (.08)  3.67 (.07)a  4.04 (.09)b  3.28* .09
 Teaching  3.34 (.08)  3.35 (.09)  3.13 (.09)  3.45 (.10)  2.17† .06
Open in a new tab
†

p < .10

*

p < .05.

Note. The first term in the name of the Amender/Avoider dyad refers to the older sibling’s classification whereas the second term in each dyad name refers to the toddler’s classification (e.g., an Amender-Avoider dyad is comprised of an older sibling Amender and a toddler Avoider). All means with different superscripts for a given variable are significantly different at a p ≤ .05.

Toddler siblings’ behavior.

There was a significant time effect, F(1, 101) = 3.84, p = .05 η2 = .04, indicating toddlers’ conflict behaviors directed toward older siblings increased from 18 (M = 2.28, SE = .05) to 24 months (M = 2.53, SE = .04), but there was no main effect of group nor time x group interaction.

Growth in Children’s Conscience from 18 to 24 months

The final aim was to examine how children’s conscience development, as reflected in their affective discomfort and moral regulation, changed over time. Preliminary analyses revealed associations between children’s conscience and the older siblings’ age (see Table S1) and gender constellation of the sibling dyad (see Table S2) so these were added as covariates in subsequent analyses. Two 4 (group) x 2 (sibling) x 2 (time: 18 and 24 months) repeated measures ANCOVAs with group as the between-subjects factor and sibling and time as repeated factors, controlling for older siblings’ age and gender constellation, were conducted with affective discomfort and moral regulation as the dependent variables. Dunn-Sidak tests which adjusted for multiple comparisons were used for the post-hoc pairwise comparisons.

Affective discomfort.

Significant main effects of sibling, F(1, 96) = 3.83, p = .05 η2 = .04, and time, F(1, 96) = 8.82, p < .01, η2 = .08, were found for affective discomfort, which were qualified by a significant sibling x time interaction, F(1, 96) = 4.27, p = .04 η2 = .04. Pairwise comparisons (all p’s < .01) revealed that older siblings, M18 months = 4.63, SE = .06, M24 months = 4.68, SE = .05, displayed higher levels of affective discomfort than toddlers, M18 months = 3.70, SE = .06, M24 months = 3.85, SE = .06) at both timepoints. There was also a significant increase in toddlers’ affective discomfort from 18 to 24 months, in general, but no increase for older siblings. No significant main effect for Amender/Avoider group nor interactions were found.

Moral regulation.

There was a significant sibling x time x Amender/Avoider group interaction for parents’ reports of moral regulation, F(3, 96) = 3.07, p < .05, η2 = .09. This 3-way interaction is depicted in Figure 1 and the means for toddlers’ and older siblings’ moral regulation can be found in Table 5. Pairwise comparisons (all p’s < .05) showed that older siblings had higher scores on moral regulation than toddlers at 18 and 24 months across all groups. There were no mean differences across groups for either toddlers or older siblings at 18 and 24 months, but there were significant changes in moral regulation across 18 and 24 months by groups and these can be found in Table 5. Toddlers in Amender-Amender, Amender-Avoider, and Avoider-Amender groups showed significant gains in moral regulation from 18 to 24 months whereas toddlers in the Avoider-Avoider group did not. Only older siblings in the Avoider-Avoider group showed significant gains in moral regulation over the same 6-month period.

Figure 1.

Figure 1

Open in a new tab

Changes in Toddlers’ and Older Siblings’ Moral Regulation from the 18- to 24-Month Timepoint across Amender/Avoider Dyads (Older Sibling-Toddler)

Table 5.

Means and Standard Errors (in Parentheses) for Toddlers’ and Older Regulation for each Amender/Avoider Dyad (Older Sibling-Toddler)

18 Months 24 Months
Older Siblings
 Amender-Amender 4.62 (.09) 4.62 (.10)
 Amender-Avoider 4.60 (.10) 4.65 (.12)
 Avoider-Avoider 4.32 (.10)a 4.44 (.11)b
 Avoider-Amender 4.58 (.11) 4.48 (.12)
Toddlers
 Amender-Amender 3.63 (.10)a 3.90 (.11)b
 Amender-Avoider 3.57 (.12)a 3.77 (.12)b
 Avoider-Avoider 3.54 (.11) 3.58 (.11)
 Avoider-Amender 3.60 (.13)a 3.83 (.13)b
Open in a new tab

Note. The first term in the name of the Amender/Avoider dyad refers to the older sibling’s classification whereas the second term in each dyad name refers to the toddler’s classification (e.g., an Amender-Avoider dyad is comprised of an older sibling Amender and a toddler Avoider). All means between toddlers and older siblings at a given timepoint for each Amender/Avoider group are significantly different at p < .001. Means significantly different across 18 and 24 months, p < .05, for each sibling within each group are denoted with different superscripts.

Discussion

The primary goal of the current report was to advance an understanding of the role that siblings have on young children’s self-conscious emotions, specifically guilt and shame, and whether there were any effects on children’s conscience development. We were particularly interested in examining siblings because older siblings may be socializing agents of younger siblings’ self-conscious emotions and conscience development, so it may make a difference if toddlers were interacting with an older sibling who was an Amender versus an Avoider. Further, we were interested in examining whether there were reciprocal effects for both siblings by also examining whether it mattered if the toddler was an Amender or Avoider. We replicated earlier research finding a pattern of emotions and behavior (i.e., low gaze avoidance and distress, yet high concern) that described Amenders and one that described Avoiders (i.e., high gaze avoidance and distress) (Barrett et al., 1993; Drummond et al., 2017). Although no direct association was found between toddlers’ and older siblings’ Amender/Avoider classifications, sibling interactions differed across the Amender/Avoider dyads suggesting that the sibling relationship might serve as a socializing context. Finally, there was evidence that toddlers and older siblings showed differential growth in conscience, particularly moral regulation, across a 6-month period from 18 to 24 months, depending on which Amender/Avoider dyad they belonged.

Are There Identifiable Patterns of Guilt (Amenders) and Shame (Avoiders) for Young Children?

We were able to replicate findings showing two patterns of behavioral responses indicative of early guilt- and shame-prone responding in the mishap paradigm for toddlers and their older siblings. For both toddlers and their older siblings, Amenders were more likely to bring or show the broken object to the experimenter than their Avoider counterparts and older siblings classified as Amenders were also more likely to comment on the damaged toy bear and make reparation attempts to fix it than older siblings classified as Avoiders. The fact that the 18-month-old toddlers’ comments and reparation attempts did not differ based on their Amender or Avoider status may be due to these young toddlers’ limited language and cognitive skills at 18 months, which may have restricted their ability to verbally comment about the toy or understand how to “repair” it. Even so, these findings are consistent with prior work suggesting that rudimentary forms of guilt- (Amenders) and shame-proneness (Avoiders) can already be observed in children as young as 18 months of age in a laboratory mishap task; an age coinciding with significant growth in self-conscious emotions during the second year of life (Lagattuta & Thompson, 2007).

Amenders and Avoiders Together: Contributions to Sibling Interactions

We also found that older siblings reacted differently to the mishap than toddlers; they were more distressed and more likely to make reparation attempts than toddlers. Such differences indicate that older siblings may very well act as role models or socializers for their toddler siblings, and in turn, contribute to growth in younger siblings’ moral and conscience development. Because we found no initial association between the toddlers’ and older siblings’ Amender and Avoider classifications, it appeared at first glance that there was no support for a sibling socialization hypothesis. Yet, when we proceeded to dig further into whether there were differences in sibling interactions, which may serve as the training ground for developing self-conscious emotions and early conscience, there were indeed differences across dyads based on their Avoider/Amender status. In contrast to expectations, toddlers whose older siblings were Amenders were not the recipients of more empathic sibling behavior compared to dyads in which both siblings were Avoiders. Older siblings in the Avoider-Avoider dyads, however, were significantly less empathic toward toddlers than those in the Avoider-Amender dyads. Toddlers, surprisingly, were the recipients of more empathy from their siblings when they were Amenders but their older siblings Avoiders. Perhaps guilt-prone toddlers demonstrate similar behaviors reflecting concern and reparation during sibling interaction and this behavior may elicit more concern and comfort from an Avoidant older sibling, enabling these sibling pairs to get along better than when both siblings are Avoiders. Thus, even toddlers, especially those demonstrating early signs of guilt, may contribute to positive sibling interactions, a finding consistent with recent research highlighting the bidirectionality of empathic concern within sibling pairs (Jambon et al., 2019).

There were also differences in the darker side of sibling interactions. Here, older siblings in the Avoider-Avoider dyads were significantly more aggressive toward toddlers than older siblings in the Avoider-Amender dyads. Specifically, toddlers, who were Avoiders themselves, were the recipients of more sibling aggression from an older sibling who was also an Avoider. That aggressive sibling behavior may result in dyads in which both children are prone to shame seems consistent with the research linking shame to anger and aggression (Muris & Meesters, 2014). Aggressive sibling interactions create hostile environments that may not be conducive to the development of toddlers’ guilt and the accompanying reparations in response to transgressions. Having an older sibling who is aggressive and hostile may lead to feelings of low self-worth and internalizing behavior problems (Dunn et al., 1996; Fagan & Najman, 2003), including a sense of shame. Further, aggressive older siblings who avoid responsibility for transgressions are likely not modeling appropriate behaviors that might give rise to more cooperative and reparative behaviors, including apologizing for wrongdoing, making amends, and experiencing guilt. Sibling relationships, however, involve two partners engaged in reciprocal exchanges and hence, both children are contributing to the quality of the relationship (Jambon et al., 2019; Pike & Oliver, 2017). Even though older siblings often are the teachers, leaders, and managers of sibling interactions, the direction of influence is not unilateral from older to younger, but mutually reciprocal (Sawyer et al., 2002; Tucker et al., 1999). Toddlers may very well be contributing to how sibling interactions unfold over time, and in turn, be producers of their own development and this may explain why Amender toddlers who might be more likely to display concern and reparation during sibling interactions are the recipients of less aggressive behavior from their Avoider older siblings. More research is needed to examine the reciprocal influence of older and younger siblings in creating different relationship dynamics that can affect young children’s sociomoral development and particularly, guilt- and shame-prone responding.

Self-Conscious Emotions of Guilt and Shame and the Development of Morality

In the current study, we also examined how being an Amender or Avoider was related to children’s conscience development. Conscience consists of both affective and behavioral components that signify young children’s internalization of parental values and standards that lay the foundation for moral affect and behavior (Thompson & Newton, 2010). Because toddlerhood is an age period marked by significant advances in social cognition, self-awareness, and perspective-taking skills (Brownell & Kopp, 2007; Kagan, 1981), we were primarily interested in whether there would be gains in toddlers’ conscience development from 18 to 24 months, when they had an older sibling who was an Amender. Older siblings may serve as both a model and socializer of moral and empathic behavior for toddlers; thus, we hypothesized that toddlers with older siblings who were Amenders would show more gains in moral development, as evidenced by increases in affective discomfort and moral regulation, from 18 to 24 months than toddlers with Avoider older siblings. This hypothesis was confirmed when examining moral regulation, but not affective discomfort, as toddlers in the Amender-Amender and Amender-Avoider groups showed an increase in moral regulation from 18 to 24 months. Toddlers in the Avoider-Amender group also showed gains in moral regulation from 18 to 24 months, whereas toddlers in the Avoider-Avoider dyads did not. These findings suggest, akin to those for sibling relationship quality, that the Avoider-Avoider combination placed toddlers on a less accelerated path of growth in moral regulation, which may have been a consequence of the less empathic and more aggressive interactions directed at them by their older siblings, who, themselves, were low in moral regulation at the 18-month timepoint. Why toddlers in the Avoider-Amender dyads also showed growth in moral regulation is more difficult to explain; but recall that these Amender toddlers were also the recipients of more empathic sibling interactions by their older siblings, indicating that positive sibling interactions with an empathic older sibling during early childhood may very well support toddlers’ developing moral sense and expression of self-conscious emotions.

Another interesting, but unexpected, finding given the age of the older siblings in this study was the differences in growth found for older siblings’ moral regulation. Recall that only older siblings in the Avoider-Avoider dyads showed significant gains in moral regulation across this period, reaching comparable scores at 24 months to older siblings in the other dyads. Because these findings emerged even after controlling for the age of older siblings, these differential trajectories for older siblings may reflect a delay in moral regulation at the initial timepoint and “catch-up” growth by the later timepoint. This delay in moral regulation for older siblings might also explain why toddlers in the Avoider-Avoider dyads did not show similar gains in moral regulation from 18 to 24 months compared to toddlers in the other dyads. These different trajectories of moral regulation for both siblings in the Avoider-Avoider sibling relationships may be due, in part, to the greater aggression and lower empathy, characteristic of sibling interaction in these dyads. Aggressive sibling interactions with a lack of empathy, combined with limited access to good exemplars of moral behavior modeled by an older sibling, may undermine the development of guilt-prone, reparative responses to transgression in young toddlers. Future research is needed to examine further how the sibling relationship contributes to moral development in early childhood.

Limitations and Future Directions

Although this study has many strengths, including the unique focus on the sibling context, it is not without its limitations. First, participating families were predominantly European American, middle class, and college educated mother-father families in the US with at least two children. Future research would benefit greatly from the recruitment of more diverse samples, in terms of race/ethnicity, socioeconomic status, and family structure, to study self-conscious emotions. Second, even though guilt- and shame-prone responding were observed in a widely-used, laboratory-based mishap task, we relied on parents’ reports of children’s sibling interactions and conscience development. We composited across mothers’ and fathers’ reports to reduce single reporter bias; however, we must acknowledge that shared method variance may have contributed to some of the findings. Third, despite research that has demonstrated that variables based on median splits are as robust as continuous variables and may actually be preferred because they are straightforward, parsimonious, and easy to understand (Iacobucci et al., 2015), we need to acknowledge that medians are sample-specific and some individual variability may be lost by dichotomizing a continuous variable. Thus, even though similar methods have been used in prior studies, creating groups based on median splits may reduce the generalizability of our findings. Moreover, no prior study has considered the development of guilt- and shame-prone responding in the context of the sibling relationship, so further research is needed to replicate these results. Finally, whereas parent reports of sibling interactions and moral development were available at both 18 and 24 months, the laboratory-based mishap task was only conducted at the 18-month timepoint so we were unable to examine the reciprocal and bidirectional nature of relations between sibling interactions and the development of guilt and shame over time, nor whether children initially classified as Amenders or Avoiders at 18 months remained Amenders and Avoiders at 24 months.

Despite these limitations, the sibling design of the current study provides a more nuanced understanding of children’s developing self-conscious emotions within the family. Sibling interactions were associated with toddlers’ and older siblings’ self-conscious emotions of guilt and shame, uncovering the potential benefits of having an empathic older sibling in promoting a guilt orientation to transgressions, and the possible risks of being the recipient of sibling aggression/rivalry in developing a shame orientation. We know the sibling context can contribute to the development of toddlers’ early morality (Dunn & Munn, 1986a). Additional research is certainly needed to both replicate the current findings and to extend these results longitudinally to disentangle the direction of effects. Until then, we end by noting that toddlers spend a considerable number of hours each day engaged in both cooperative play and conflict with an older sibling (Dunn, 1983; Perlman & Ross, 1997), yet siblings are rarely studied as socializers of early moral development. Findings from the current research provide some of the first evidence linking the quality of sibling interaction with both the expression of self-conscious emotions such as shame and guilt, and the development of moral regulation across 18 to 24 months for young toddlers. Future research should continue to investigate the sibling relationship as a developmental context for the emergence of early conscience and self-conscious emotions.

Supplementary Material

Supp 1

Acknowledgments

This research was supported by grants from the John E. Fetzer Institute (Project 2228) and the Eunice Kenney Shriver National Institute of Child Health and Human Development (R01HD054573, K02HD047423) to Brenda L. Volling. We are extremely grateful to the parents and children of the Family Transitions and Toddler Development Study for their participation.

Footnotes

We have no known conflict of interest to disclose.

References

  1. Barrett KC (1995). A functionalist approach to shame and guilt. In Tangney JP & Fischer KW (Eds.), Self-conscious emotions: The psychology of shame, guilt, embarrassment, and pride (pp. 25–63). New York, NY: Guilford Press. [Google Scholar]
  2. Barrett KC (2005). The origins of social emotions and self-regulation in toddlerhood: New evidence. Cognition and Emotion, 19(7), 953–979. https://doi-org.nuncio.cofc.edu/10.1080/02699930500172515 [Google Scholar]
  3. Barrett KC, & Nelson-Goens GC (1997). Emotion communication and the development of the social emotions. In Barrett KC (Eds.), The communication of emotion: Current research from diverse perspectives (pp. 69–88). San Francisco, CA: Jossey-Bass. [DOI] [PubMed] [Google Scholar]
  4. Barrett KC, Zahn-Waxler C, & Cole PM (1993). Avoiders vs. amenders: Implications for the investigation of guilt and shame during toddlerhood? Cognition and Emotion, 7(6), 481–505. 10.1080/02699939308409201 [DOI] [Google Scholar]
  5. Brody GH, Stoneman Z, MacKinnon CE, & MacKinnon R (1985). Role relationships and behavior between preschool-aged and school-aged sibling pairs. Developmental Psychology, 21(1), 124–129. 10.1037/0012-1649.21.1.124 [DOI] [Google Scholar]
  6. Brownell CA, & Kopp CB (2007). Transitions in toddler socioemotional development: Behavior, understanding, relationships. In Brownell CA & Kopp CB (Eds.), Socioemotional development in the toddler years: Transitions and transformations. (pp. 1–39). Guilford Press. [Google Scholar]
  7. Cole PM, Barrett KC, & Zahn-Waxler C (1992). Emotion displays in two-year-olds during mishaps. Child Development, 63(2), 314–324. 10.2307/1131481 [DOI] [PubMed] [Google Scholar]
  8. Drummond JK, Hammond SI, Satlof-Bedrick E, Waugh WE, & Brownell CA (2017). Helping the one you hurt: Toddlers’ rudimentary guilt, shame, and prosocial behavior after harming another. Child Development, 88(4), 1382–1397. 10.1111/cdev.12653 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Dunn J (1983). Sibling relationships in early childhood. Child Development, 54(4), 787–811. https://doi-org.nuncio.cofc.edu/10.2307/1129886 [Google Scholar]
  10. Dunn J (1988). The beginnings of social understanding. Cambridge, MA: Harvard University Press. [Google Scholar]
  11. Dunn J & Munn P. (1986a). Siblings and the development of prosocial behaviour. International Journal of Behavioral Development, 9(3), 265–284. [Google Scholar]
  12. Dunn J, & Munn P (1986b). Sibling quarrels and maternal intervention: Individual differences in understanding and aggression. Child Psychology & Psychiatry & Allied Disciplines, 27(5), 583–595. https://doi-org.nuncio.cofc.edu/10.1111/j.1469-7610.1986.tb00184.x [DOI] [PubMed] [Google Scholar]
  13. Fagan AA, & Najman JM (2003). Association between early childhood aggression and internalizing behavior for sibling pairs. Journal of the American Academy of Child & Adolescent Psychiatry, 42(9), 1093–1100. https://doi-org.nuncio.cofc.edu/10.1097/01.CHI.0000070240.24125.EA [DOI] [PubMed] [Google Scholar]
  14. Grusec JE, Chaparro MP, Johnston M, & Sherman A (2014). The development of moral behavior from a socialization perspective. In Killen M & Smetana JG (Eds.), Handbook of moral development, 2nd ed. (pp. 113–134). Psychology Press. https://doi-org.nuncio.cofc.edu/10.4324/9780203581957.ch6 [Google Scholar]
  15. Hetherington EM, & Clingempeel WG (1992). Coping with marital transitions: A family systems perspective. Monographs of the Society for Research in Child Development, 57(2–3), 1–242. 10.2307/1166050 [DOI] [Google Scholar]
  16. Iacobucci D, Posavac SS, Kardes FR, Schneider MJ, & Popovich DL (2015). Toward a more nuanced understanding of the statistical properties of a median split. Journal of Consumer Psychology, 25(4), 652–665. https://doi-org.nuncio.cofc.edu/10.1016/j.jcps.2014.12.002 [Google Scholar]
  17. Jambon M, Madigan S, Plamondon A, Daniel E, & Jenkins JM (2019). The development of empathic concern in siblings: A reciprocal influence model. Child Development, 90(5), 1598–1613. https://doi-org.nuncio.cofc.edu/10.1111/cdev.13015 [DOI] [PubMed] [Google Scholar]
  18. Kagan J (1981). The second year: The emergence of self-awareness. Cambridge, MA: Harvard University Press. [Google Scholar]
  19. Kochanska G, & Aksan N (2004). Conscience in childhood: Past, present, and future. Merrill-Palmer Quarterly, 50(3), 299–310. https://doi.org.nuncio.cofc.edu/10.1353/mpq.2004.0020 [Google Scholar]
  20. Kochanska G, Casey RJ, & Fukumoto A (1995). Toddlers’ sensitivity to standard violations. Child Development, 66(3), 643–656. 10.2307/1131940 [DOI] [Google Scholar]
  21. Kochanska G, DeVet K, Goldman M, Murray K, & Putnam SP (1994). Maternal reports of conscience development and temperament in young children. Child Development, 65(3), 852–868. https://doi-org.nuncio.cofc.edu/10.2307/1131423 [PubMed] [Google Scholar]
  22. Kochanska G, Gross JN, Lin M, & Nichols KE (2002). Guilt in young children: Development, determinants, and relations with a broader system of standards. Child Development, 73(2), 461–482. 10.1111/1467-8624.00418 [DOI] [PubMed] [Google Scholar]
  23. Lagattuta KH, & Thompson RA (2007). The development of self-conscious emotions: Cognitive processes and social influences. In Tracy JL, Robins RW, & Tangney JP (Eds.), The self-conscious emotions: Theory and research (pp. 91–113). New York, NY: Guilford Press. [Google Scholar]
  24. Lewis HB (1971). Shame and guilt in neurosis. New York, NY: International Universities Press. [Google Scholar]
  25. Lewis M, Alessandri SM, & Sullivan MW (1992). Differences in shame and pride as a function of children’s gender and task difficulty. Child Development, 63(3), 630–638. 10.2307/1131351 [DOI] [PubMed] [Google Scholar]
  26. Muris P & Meesters C (2014). Small or big in the eyes of the other: On the developmental psychopathology of self-conscious emotions as shame, guilt, and pride. Clinical Child and Family Psychology Review, 17, 19–40. 10.1007/s10567-013-0137-z [DOI] [PubMed] [Google Scholar]
  27. Perlman M, & Ross HS (1997). The benefits of parent intervention in children’s disputes: An examination of concurrent changes in children’s fighting styles. Child Development, 68(4), 690–700. 10.2307/1132119 [DOI] [Google Scholar]
  28. Perry NE (2019). Recognizing early childhood as a critical time for developing and supporting self-regulation. Metacognition and Learning, 14(3), 327–334. https://doi-org.nuncio.cofc.edu/10.1007/s11409-019-09213-8 [Google Scholar]
  29. Pike A, & Oliver BR (2017). Child behavior and sibling relationship quality: A cross-lagged analysis. Journal of Family Psychology, 31(2), 250–255. https://doi-org.nuncio.cofc.edu/10.1037/fam0000248 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Rushton JP, Brainerd CJ, & Pressley M (1983). Behavioral development and construct validity: The principle of aggregation. Psychological Bulletin, 94(1), 18–38. 10.1037/0033-2909.94.1.18 [DOI] [Google Scholar]
  31. Sawyer KS, Denham S, DeMulder E, Blair K, Auerbach-Major S, & Levitas J (2002). The contribution of older siblings’ reactions to emotions to preschoolers’ emotional and social competence. Marriage & Family Review, 34(3–4), 183–212. https://doi-org.nuncio.cofc.edu/10.1300/J002v34n03_01 [Google Scholar]
  32. Schaefer E & Edgerton M (1981). The sibling inventory of behavior. University of North Carolina: Chapel Hill, NC. [Google Scholar]
  33. Scrimgeour MB, Mariotti EC, & Blandon AY (2017). Children’s physiological regulation and sibling conflict as correlates of children’s conscience development. Social Development, 26(2), 329–348. 10.1111/sode.12194 [DOI] [Google Scholar]
  34. Tangney JP (1995). Shame and guilt in interpersonal relationships. In Tangney JP & Fischer KW, J. P. (Eds.), Self-conscious emotions: The psychology of shame, guilt, embarrassment, and pride (pp. 114–139). New York, NY: Guilford Press. [Google Scholar]
  35. Tangney JP (1998). How does guilt differ from shame? In Bybee J, Bybee J (Eds.), Guilt and children (pp. 1–17). San Diego, CA: Academic Press. 10.1016/B978-012148610-5/50002-3 [DOI] [Google Scholar]
  36. Tangney JP, Youman K, & Stuewig J (2009). Proneness to shame and proneness to guilt. In Leary MR & Hoyle RH (Eds.), Handbook of individual differences in social behavior (pp. 192–209). New York, NY: Guilford Press. [Google Scholar]
  37. Thompson RA, & Newton EK (2010). Emotion in early conscience. In Arsenio WF & Lemerise EA (Eds.), Emotions, aggression, and morality in children: Bridging development and psychopathology. (pp. 13–31). American Psychological Association. https://doi-org.nuncio.cofc.edu/10.1037/12129-001 [Google Scholar]
  38. Tucker CJ, Updegraff KA, McHale SM, & Crouter AC (1999). Older siblings as socializers of younger siblings’ empathy. The Journal of Early Adolescence, 19(2), 176–198. https://doi-org.nuncio.cofc.edu/10.1177/0272431699019002003 [Google Scholar]
  39. van Berkel SR, Song J, Gonzalez R, Olson SL, & Volling BL (2020). Don’t touch: Developmental trajectories of toddlers’ behavioral regulation related to older siblings’ behaviors and parental discipline. Social Development. 10.1111/sode.12440 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Volling BL, & Elins JL (1998). Family relationships and children’s emotional adjustment as correlates of maternal and paternal differential treatment: A replication with toddler and preschool siblings. Child Development, 69(6), 1640–1656. 10.2307/1132137 [DOI] [PubMed] [Google Scholar]
  41. Walter JL, & LaFreniere PJ (2007). Preschoolers’ avoidance in a mishap paradigm: Implications for emotional adjustment, guilt, and shame. North American Journal of Psychology, 9(1), 111–130. [Google Scholar]
  42. Wellman HM, & Liu D (2004). Scaling of theory-of-mind tasks. Child development, 75(2), 523–541. 10.1111/j.1467-8624.2004.00691.x [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp 1
NIHMS1794826-supplement-Supp_1.docx (38.2KB, docx)

RESOURCES