Prolonged Infant Crying: Caregiving Quality and Child Physical Abuse Risk

Abstract

Research suggests that prolonged infant crying may increase risk for child physical abuse (CPA). However, few studies have examined behavioral responses to infant crying among parents at risk for CPA. The present study sought to fill this gap by using a simulated infant to examine how mothers and fathers with varying degrees of CPA risk respond to prolonged infant crying. Specifically, a sample of 184 general population caregivers (107 mothers and 77 fathers) participated in a task that involved attempting to soothe a simulated infant that cried continuously for 30 min. The simulated infant sessions were video-recorded, and quality of caregiving behavior was coded in 5-min segments. Participants rated their negative affect (feelings of upset, distress, irritability) at the outset of the data collection session, before beginning the simulated infant task, and after the simulated infant task concluded. It was predicted that high-risk caregivers, compared to low-risk caregivers, would display lower quality caregiving, higher levels of negative affect, and that CPA risk group differences would increase over time. Gender differences were explored in relation to the aforementioned hypotheses. Over the course of the 30-min simulated infant task, the quality of caregiving behavior diminished among both high- and low-risk caregivers. As expected, high-risk caregivers, compared to low-risk caregivers, exhibited lower quality caregiving behaviors and were more likely to discontinue the simulated infant task early. In addition, high-risk, compared to low-risk, caregivers reported higher levels of negative affect throughout the data collection session, with the highest level of negative affect reported by high-risk caregivers after completing the simulated infant task. Overall, the quality of caregiving exhibited by men and women did not significantly differ. The present findings highlight the importance of early intervention designed to support caregivers’ abilities to respond effectively to prolonged infant crying.

Keywords

child maltreatment physical child abuse parenting

Research suggests that infant crying increases during the first weeks of life, peaks during the second month, and declines until stabilizing in the fourth month postpartum (Barr, 1990). The longest crying bouts are commonly observed at the peak of the “crying curve” (i.e., during the second month postpartum; Barr, 2012). At their peak, crying bouts may be prolonged, resistant to soothing, and unrelated to hunger, wet diapers, or caregiving style (Barr, 2012). Prolonged, inconsolable bouts of crying are distressing events for most caregivers. To demonstrate this point, Barr et al. (2014) assessed how postpartum mothers responded to a 10-min audio recording of infant crying and found that mothers’ frustration increased as the crying persisted over time (Barr et al., 2014). Similarly, Fujiwara et al. (2011) investigated the relationship between three modes of infant distress (fussing, crying, and inconsolable crying) and caregivers’ self-reported frustration. The results indicated that caregiver frustration differed by mode of infant distress, such that long bouts of inconsolable crying (compared to fussing or crying) were most strongly related to caregiver frustration (Fujiwara et al., 2011). Relatedly, Fairbrother et al. (2014) compared cognitive, emotional, and behavioral reactions reported by mothers who listened to 10 min of infant crying (cry condition) to that of mothers who listened to 10 min of infant cooing (coo condition). Compared to women in the coo condition, women in the cry condition reported higher levels of frustration, negative emotions, urges to withdraw from the infant, and infant-directed harm thoughts (Fairbrother et al., 2014).

In light of these findings, it is perhaps unsurprising that prolonged inconsolable crying is considered a potential precipitant of aggression towards infants (Adamsbaum et al., 2010; Reijneveld et al., 2004). Given their fragile constitution, aggression towards infants can result in serious physical injuries (Leventhal et al., 2010), including abusive head trauma (AHT; Barr, 2012). A review of child death cases in the U.S. from 2005 to 2009 indicated that approximately one third of all child maltreatment-related deaths were due to AHT (Palusci & Covington, 2014). Infants who survive AHT face short- and long-term consequences such as vision problems, developmental delays, physical disabilities, and hearing loss (e.g., Eismann et al., 2020; Lind et al., 2016; Manfield et al., 2021; for reviews see also Barr, 2012; Narang & Clarke, 2014).

Given the serious consequences associated with aggression towards infants, it is especially important to understand how physically abusive caregivers and those who are at high-risk for aggression respond to prolonged inconsolable infant crying. In one of the early studies on this topic, Frodi and Lamb (1980) found that abusive parents, compared to non-abusive parents, reported feeling more annoyed and less sympathetic after watching a 2-min video of a crying infant. Similarly, Crouch et al. (2008) found that high CPA risk parents, compared to low CPA risk parents, rated a crying infant depicted in a 2-min video more negatively and reported higher levels of hostile feelings after watching the video.

Relatedly, Compier-de Block et al. (2015) examined handgrip force (using a dynamometer) following exposure to infant crying among a sample of maltreating and non-maltreating mothers. In this study, mothers were asked to listen to infant crying sounds and infant laughing sounds. The results indicated that maltreating mothers (compared to non-maltreating mothers) used excessive force more often following exposure to infant crying (vs. laughter sounds). Collectively, these studies suggest that, when exposed to infant crying, abusive caregivers and high-risk caregivers may experience more negative feelings (e.g., distress, irritation) and use excessive force more often than their non-abusive and low-risk counterparts (e.g., Compier-de Block et al., 2015; Crouch et al., 2008; Frodi & Lamb, 1980).

While informative, prior studies examining how abusive and high-risk caregivers react to infant crying have been limited in that they typically relied on brief (e.g., 2 min) segments of video or audio recordings of a crying child. Although such methodology may be well suited for research designed to examine how caregivers respond to various visual/acoustical characteristics of infant cries (e.g., Crowe & Zeskind, 1992), it fails to provide an opportunity to observe how caregivers respond behaviorally to prolonged bouts of infant crying.

Indeed, few studies have examined how high CPA risk caregivers respond behaviorally to crying infants. Although not specifically focused on parental responses to infant crying, Cerezo et al. (2008) examined CPA risk group differences in maternal responses to their infants’ behaviors during 5-min free play interactions at 3 months, 12 months, and 15 months. Results revealed that high-risk mothers tended to respond to difficult infant behaviors (e.g., crying, negative social approach, noncompliance) by engaging in affectionate but intrusive caregiving during the 3-month and 12-month interactions. In contrast, low-risk mothers tended to respond to difficult infant behaviors with affectionate and sensitive caregiving during the 3-month and 12-month interactions. Overall, the Cerezo et al. findings suggested that high-risk parents (compared to low-risk parents) exhibit lower quality caregiving behaviors in response to infant crying.

In order to advance research examining how caregivers behave in response to prolonged infant crying, it is important to consider the challenges inherent in this line of research. For example, infants do not cry on cue and they vary considerably (over time and across infants) in their cry characteristics. Thus, studying parental responses to infant crying during parent-child interactions is fraught with challenges. Although the use of audio/video recordings allow researchers to standardize infant cry stimuli, such methods are limited in that they do not provide parents with the opportunity to attempt to soothe the infant. To overcome these challenges, the present study used a simulated crying infant as a standardized crying infant stimulus that parents could attempt to soothe. Infant simulators were originally developed as teaching aides and have been used: 1) to train medical professionals on a variety of topics, and 2) to educate youth about the demands of parenting and the harm caused by shaking infants (Barnett & Hurst, 2003; Roberts & McCowan, 2004). Use of an infant simulator in research allows the researcher to control the onset/offset of infant crying and to manipulate the responsiveness of the infant to the caregiver’s efforts to soothe it (McKay et al., 2019; Rutherford, 2019; Rutherford et al., 2013; Voorthuis et al., 2013).

Present Study

The present study was designed to examine how caregivers with varying levels of CPA risk respond to a simulated infant programed to cry continuously for 30 min. It was predicted that the overall quality of caregiving behaviors exhibited by high-risk caregivers would be lower than that of low-risk caregivers. Moreover, it was expected that the quality of caregiving would diminish over time and the hypothesized CPA risk group differences would increase over the course of the 30-min simulated infant task. It was also predicted that caregivers would experience increases in negative affect over time as they attempted to soothe the inconsolable simulated infant, with the highest levels of negative affect reported by high-risk caregivers. Given that fathers are often the alleged perpetrator in cases of abusive head trauma in infants (Scribano et al., 2013), research incorporating both mothers and fathers with varying degrees of CPA risk is needed to advance research in this area. Whenever possible, gender-specific effects were explored in relation to the aforementioned hypotheses.

Method

Participants

Participants were recruited from the community surrounding a Midwestern university in the United States from 2016 to 2018. Participants had to be at least 18 years or older and have at least one child less than 18 years of age living in the home. A total of 233 general population parents were recruited for inclusion in this study. Of the total sample, 28 participants were excluded for faking good, 7 for random responding, 7 participants were missing caregiving quality data due to equipment malfunction, and 7 were missing 10% or more data on other measures. Participants in the final sample (N = 184) had a mean age of 36.6 years (SD = 11.9) and were 58.2% female. With respect to race, 47.0% were African American, 45.9% were Caucasian, 4.9% were multi-racial, 1.6% were Asian, and 0.6% were Native American. In terms of ethnicity, 8.7% were Hispanic. As for marital status, 59.8% of participants were single, separated, or divorced, and the remaining 40.2% of participants were married or cohabitating. On average, participants had 2.0 children living in their homes (SD = 1.1) and the average age of their youngest child was 6.7 years (SD = 5.4 years). The sample was diverse in terms of educational attainment: 35.9% reported 12 or fewer years of education, 39.2% had completed an Associate’s degree or some college, and 24.9% had completed a Bachelor’s degree or higher. With respect to annual household income, 34.4% of the sample reported an annual household income of $10,000 or less, 31.1% between $10,000 and $29,000, and 34.4% reported an annual household income greater than $30,000.

Procedures

Procedures for this study were reviewed and approved by the institutional review board at the authors’ institution. To recruit participants for the preliminary screening sessions, informational flyers were distributed through local businesses and service agencies (e.g., daycares, restaurants, grocery stores, churches, social service agencies). Participants who met the eligibility criteria for the present study (i.e., 18 years of age or older, with at least one child living in the home) were invited to complete the procedures described below.

Upon arrival for the simulated infant data collection session, parents were asked to read and sign consent forms that provided them with information regarding the procedures, risks, benefits, and voluntary nature of their participation. Upon providing consent, participants rated their state affect, completed a series of performance-based tasks, and then rated their state affect again. Data were collected individually in a private room that was set up to resemble a nursery.

Next participants completed the simulated infant task. At the outset of the task, participants were led to believe that the simulated infant (RealCare Baby-3) could respond to facial expressions, vocal expressions, and physical contact the same way as a normal infant. The research assistant then demonstrated how to soothe the simulated crying infant by doing things such as briefly talking to, smiling at, touching, rocking, and then feeding the simulated infant. During the demonstration, the research assistant wore a caregiver microchip on their wrist (hidden under a bandage) that, when in contact with the simulated infant, caused the infant to stop crying in response to certain caregiving behaviors (e.g., baby bottle touched to the simulator’s mouth). After the demonstration, participants were handed the infant simulator and instructed that it was their turn to care for the infant simulator. Each participant was told that they were free to move around the room and use any of the toys or props as needed. However, participants were not provided a caregiver microchip while completing the simulated infant task. Without the microchip, the crying could only be stopped manually using computer software located in an adjacent room. Participants were then left alone with the infant simulator, and a research assistant in the adjacent room programed the infant simulator to cry continuously for 30 min.

Given the potentially stressful nature of this procedure, participants were allowed to stop the task at any time. After 30 min (or upon requesting to terminate the simulator task), a research assistant in the control room turned off the infant simulator. Next, participants were asked to rate their affective state and completed several additional measures. Upon completion of all measures, participants were debriefed and paid $30 for their participation.

Measures/Instruments

Infant simulator

The RealCare Baby-3 is an infant simulator (weight = 7 lbs., height = 22 inch) that is remotely (i.e., wirelessly) controlled through computer software. Two infant simulators were available for use in the present study: one was a Caucasian model and the other was an African American model. The cry sounds emitted by the infant simulator were recordings of actual infant cries that peaked at 80–85 dB. The RealCare Baby-3 infant simulators could be programmed to be responsive (when the caregiver is wearing the caregiver microchip) to several caregiving behaviors such as feeding, rocking, burping, and changing. In this study, the infant simulators were used to portray an inconsolable infant (i.e., programmed to cry continuously for 30 min and be unresponsive to soothing attempts). During the 30-min simulated infant task, the simulator was programmed to cycle through six cry segments that lasted approximately 5 min each. The simulated infant was dressed in gender neutral clothing (yellow pajamas) and was referred to as “Chris.” Convergent validity for the simulated infant task is supported by research showing that caregiving quality displayed during a simulated infant task is correlated with caregiving quality observed during mothers’ interactions with their own children (Bakermans-Kranenburg et al., 2015; Hechler et al., 2019). Additionally, soothing techniques used by mothers while attempting to soothe the simulated crying infant predicted aspects of their own infants’ temperament four and eight months after completing the simulated infant task (McKay et al., 2019).

Child abuse potential inventory

The CAP Inventory is a self-report screening tool for assessing child physical abuse risk (Milner, 1986). Respondents are asked to indicate whether they agree or disagree with 160 statements, of which 77 items comprise the Abuse scale. The Abuse scale has six subscales: distress (36 items), rigidity (14 items), unhappiness (11 items), problems with child and self (6 items), problems with family (4 items), and problems from others (6 items). Scores on the Abuse scale range from 0 to 486, with higher scores indicating greater potential for child physical abuse. The CAP Inventory also contains three validity indices that indicate whether respondents are randomly responding, faking good or faking bad. In the present study, participants were classified as low CPA risk if their Abuse scores were below the signal detection theory cut score of 166 (Milner, 1986) and no validity indices (i.e., random responding, faking good, or faking bad) were elevated. Respondents were classified as high CPA risk if their Abuse scores were at or above 166 and the CAP Inventory validity indices indicated that they were not randomly responding or faking bad.

The CAP Abuse scale has excellent internal consistency, ranging from 0.92 to 0.95 (for general population and maltreating parents; α = 0.93 in the current sample). Adequate test-retest reliabilities in general population samples have been reported (0.91 for 1-day, 0.90 for 1-week, 0.83 for 1-month, and 0.75 for 3-month intervals; Milner, 1986). Numerous studies report construct validity data for the CAP Abuse scale (for reviews see Milner, 1986, 1993, Milner & Crouch, 2017). For example, CAP Abuse scores are positively correlated with measures of aggression in parents (Crouch et al., 2012; Rodriguez, 2010) and are predictive of subsequent child physical abuse (Chaffin & Valle, 2003; Milner et al., 1984). Based on discriminant analyses, correct classification rates range from 80% to 90% for confirmed child physical abusers and comparison groups (Milner, 1986). Studies examining the CAP Abuse scale’s specificity indicate 100% correct classification of nurturing foster parents, low-risk mothers, and nurturing mothers.

Parent-child early relational assessment

The PCERA is a 28-item coding scheme used to rate the quality of caregivers’ behaviors, verbalizations, and emotions during parent-child interactions (PCERA; Clark, 1985, 1999). The PCERA total score is comprised of six subscales: tone of voice (e.g., warm, hostile), parental affect (e.g., positive affect, negative affect), parental mood (e.g., cheerful, depressed), expressed attitudes (e.g., enjoyment, displeasure), affective and behavioral involvement (e.g., quality of verbalizations, amount of physical contact), and parenting style (e.g., flexibility, intrusiveness). Each PCERA item was coded on a 5-point scale, ranging from 1 (less positive/more negative affect or behavior) to 5 (more positive/less negative affect or behavior). Thus, higher scores represented higher quality caregiving. PCERA items were coded based on the caregiving behaviors displayed during each 5-min segment of the simulated infant task. Specifically, each 5-min segment was viewed simultaneously by two coders, who separately rated each PCERA item and then resolved any discrepancies to finalize the PCERA ratings for each 5-min video segment. If differences could not be resolved, a senior investigator viewed the segment in question with the coders, and together, reached consensus on how the item would be coded. This coding procedure ensured that interrater reliability was 100% for all 28 items.

Each coder completed 40-hours of training on the PCERA coding task and also practiced coding videos until an interrater reliability of at least .80 was obtained. All coders were blind to participant CPA risk status. Item 19 (i.e., contingent responsivity to child’s positive and age-appropriate behavior) could not be rated due to the fact that the simulated infant was programmed to cry continuously (i.e., it did not display positive behaviors).

For the purpose of the present study, the mean rating across the 28 PCERA items was used to calculate the PCERA total and subscale scores. The PCERA total scores estimated the overall caregiving quality for each 5-min segment of the infant simulator session. In the present study, the PCERA total scores had excellent internal consistency across each of the six 5-min caregiving segments, with a mean Cronbach’s alpha across the six caregiving segments of 0.95. PCERA ratings have been shown to be sensitive to individual differences in functioning (e.g., bi-polar disorder status) among mothers of 3-month-old infants (Anke et al., 2019). Specifically, bi-polar mothers, compared to mothers with no mental disorders, were rated (using the PCERA) as displaying lower quality caregiving during a 5-min free play interaction.

Negative affect ratings

State negative affect was assessed using three items (i.e., distress, upset, and irritable) from the Positive and Negative Affect Rating Scale (PANAS; Watson et al., 1988). Participants were asked to think about how they were feeling at this moment and to respond to each item using a 5-point scale ranging from 1 = very slightly or not at all to 5 = extremely. Negative affect ratings were obtained at the beginning of the data collection session (baseline); after completing several performance-based tasks but before initiating the simulated infant task (pre-simulated infant task); and after completing the infant simulator task (post-simulated infant task). Composite negative affect scores were created by averaging the affect ratings at each time point. Correlations among the three ratings at each time point ranged from 0.27 to 0.50 (mean r = 0.36; α′s = 0.57 at Time 1, 0.63 at Time 2, 0.67 at Time 3), suggesting that the three items were assessing related but distinct aspects of negative affect. As expected, the three composite state negative affect scores were each modestly correlated with a measure of trait negative affect (i.e., the Distress subscale of the CAP Inventory, r’s ranged from 0.26 to .37. p’s < .01) and were unrelated to a measure of cognitive rigidity (i.e., the Rigidity subscale of the CAP Inventory, r’s ranged from 0.07 to .11, p’s > .05).

Analytic Strategy

Descriptive statistics were generated to examine demographic (age, race, gender) and psychosocial characteristics (marital status, educational attainment, annual household income) of the sample. Chi-square analysis was used to determine whether rates of early termination of the simulated infant task (i.e., electing to end the task before 30 min) differed for low CPA risk caregivers and high CPA risk caregivers. To test our a priori hypotheses, a 2 (CPA risk: low, high) × 2 (Gender: female, male) x 6 (Time: Blocks 1 through 6) mixed factorial ANCOVA (with repeated measures on the last factor) was conducted for the PCERA total scores. Participant age and race (coded Caucasian vs. African American/other) were entered as covariates in the aforementioned ANCOVA analyses. Partial eta squared (η²_p) and Cramer’s V (V) were used to estimate effect sizes for ANOVA and chi-square analyses, respectively. Supplemental analyses examined the quality of caregiving exhibited during the first 10 min of the task by high-risk participants who terminated the simulated infant session early.

Results

Demographic Characteristics

Table 1 presents the demographic and psychosocial functioning characteristics of the final sample. Of the 184 caregivers included in the final sample, 94 were classified as low CPA risk and 90 were classified as high CPA risk. On average, high-risk caregivers were younger (M = 32.8 years, SD = 10.9) than low-risk caregivers (M = 40.2 years, SD = 11.8), F (1, 180) = 19.23, p < .001, η²_p = 0.097. With respect to race (coded Caucasian vs. African American/other), high-risk caregivers, compared to low-risk caregivers, were more often African American/other, X²(N = 183, df = 1) = 15.60, p < .001, V = 0.292. In terms of psychosocial functioning, high-risk caregivers, compared to low-risk caregivers, were more often single/separated/divorced, X²(N = 184, df = 1) = 13.45, p < .001, V = 0.270, reported lower levels of educational attainment, X²(N = 181, df = 2) = 20.56, p < .001, V = 0.337, and had lower annual household income, X²(N = 183, df = 2) = 30.83, p < .001, V = 0.410. Mean CAP Inventory abuse scores for the low-risk caregivers and the high-risk caregivers were 67.6 (SD = 40.1) and 255.0 (SD = 64.4), respectively, F (1, 182) = 566.46, p < .001, η²_p = 0.757.

Table 1.

Demographic Characteristics by Participant Gender and CPA Risk Status (N = 184).

Characteristics	Gender		CPA Risk
Characteristics	Women (n = 107)	Men (n = 77)	Low risk (n = 94)	High risk (n = 90)
Race (%)
Caucasian	51.9	37.7	60.2	31.1
African American	43.4	51.9	34.4	60.0
Multi-race	3.8	6.5	2.2	7.8
Other	0.9	3.9	3.2	1.1
Ethnicity (%)
Hispanic	10.5	8.9	10.3	9.4
Marital status (%)
Single/separated/divorced	57.0	63.6	46.8	73.3
Married/Cohabitating	43.0	36.4	53.2	26.7
Education (%)
12 or fewer years	27.9	46.8	23.7	48.9
Associates degree or some college	42.3	35.0	38.7	39.8
Bachelor’s degree or higher	29.8	18.2	37.6	11.4
Annual household income (%)
Less than $10,000	34.0	35.0	17.0	52.8
$10,000-$29,999	26.4	37.7	33.0	29.2
$30,000 or higher	39.6	27.3	50.0	18.0

Note. CPA = child physical abuse.

Within the final sample (N = 184) there were 77 (41.8%) men and 107 (58.2%) women. Men and women in this sample did not differ in age, F (1, 180) = 1.46, p = 0.227, η²_p = .008, men: M = 37.8 years, SD = 12.6; women: M = 35.7 years, SD = 11.4. Men tended to more often be African American/other compared to women, X²(N = 183, df = 1) = 3.63, p = .057, V = 0.141. With regard to psychosocial functioning, men and women in this sample did not differ in terms of marital status, X²(N = 184, df = 1) = 0.82, p = 0.366, V = 0.067, or annual household income, X²(N = 183, df = 2) = 3.80, p = 0.149, V = 0.144. Gender differences in educational attainment were significant, X²(N = 181, df = 2) = 7.38, p = .025, V = 0.202, such that men were more likely to report 12 or fewer years of education, whereas women were more likely to report higher levels of education (e.g., BA or higher). Men and women did not differ significantly in terms of CPA risk status, X²(N = 184, df = 1) = 0.04, p = .843, V = 0.015; men: 51.9% low risk, 48.1% high risk; women: 50.5% low risk, 49.5% high risk.

Simulated Infant Task Completion Rates

The mean number of minutes spent participating in the simulated infant task was 27.8 min (SD = 5.8). Overall, 83.2% of participants completed the entire 30-min simulated infant task, whereas 16.8% of participants chose to end the task early. Completion rates did not differ by participant gender, X²(N = 184, df = 1) = 1.41, p = .235, V = 0.088. However, high risk parents were more likely than low-risk parents to discontinue the simulated infant session early (see Figure 1). Among low-risk parents, only 7 (7.4%) parents elected to end the simulated infant session early, whereas 24 (26.7%) high-risk parents elected to end the simulated infant session early, X²(N = 184, df = 1) = 12.12, p < .001, V = 0.257.

Figure 1.

Cumulative percentage of low-risk caregivers and high-risk caregivers who discontinued the 30-min simulated infant task over time.

In an effort to understand the characteristics of parents who did not complete the simulated infant session, we compared the CAP Inventory abuse scores for completers and noncompleters among low-risk and high-risk parents. For low-risk parents, CAP Inventory abuse scores did not differ significantly between completers (M = 66.8, SD = 39.1) and noncompleters (M = 71.3, SD = 48.1), F (1, 92) = 0.08, p = .775, η²_p = 0.001. For high-risk parents, CAP Inventory abuse scores tended to be higher among high-risk noncompleters (M = 272.5, SD = 67.3) compared to high-risk completers (M = 245.6, SD = 61.8), F (1, 88) = 3.17, p = .078, η²_p = 0.035. Given that the majority (77.4%) of noncompleters were high-risk caregivers, and the high-risk noncompleters differed systematically from high-risk completers (i.e., tended to have higher abuse scores), we elected to treat the high-risk noncompleters as a distinct group in supplemental analyses (as described below) rather than attempting to estimate their missing behavioral data.

Quality of Caregiving over Time

Quality of caregiving over time was analyzed using a 2 (CPA risk: low, high) × 2 (Gender: female, male) × 6 (Time: Blocks 1 through 6) mixed factorial ANCOVA (with repeated measures on the last factor) using the PCERA total score as an estimate of caregiving quality during each 5-min segment. Age and race (coded Caucasian vs. African American/other) were entered as covariates. These planned analyses were conducted using the subset of participants who completed all six segments of the simulated infant session.

For overall quality of caregiving behavior, none of the three-way (CPA risk × Gender × Time) or two-way (CPA risk x Time, Time x Gender) interactions were significant, nor were the covariates (age and race), or the main effect of Gender (all p’s > .05 and η²_p’s < 0.01). As expected, significant main effects for CPA risk and Time were observed.

Compared to low-risk caregivers, high-risk caregivers exhibited lower quality caregiving throughout the simulated infant task, F (1, 147) = 6.12, p = .015, η²_p = 0.040, low-risk: M_adj = 3.96, SE = 0.051; high-risk: M_adj = 3.76, SE = 0.059. Follow-up tests were conducted to examine CPA risk group differences for each of the subscales that comprise the PCERA caregiving quality total score (see Table 2). High-risk, compared to low-risk, caregivers obtained significantly lower subscale scores for mood, affective and behavioral involvement, and parenting style. For the remaining PCERA subscales (tone of voice, affect, and attitudes), the direction of differences was as expected; however, these differences were small in magnitude and not statistically significant.

Table 2.

Descriptive and Inferential Statistics for CPA Risk Effects for PCERA Subscales.

PCERA subscale^a	Low risk M (SE)	High risk M (SE)	F-value (1, 144)	η ² _p
Tone of voice (α = .69)	4.0 (.06)	3.8 (.07)	2.40	.017
Affect (α = .72)	3.7 (.07)	3.5 (.08)	2.46	.017
Mood (α = .51)	4.2 (.04)	4.0 (.05)	5.26*	.035
Attitudes (α = .72)	3.5 (.07)	3.3 (.08)	2.68	.018
Affective and behavioral involvement (α = .94)	3.8 (.06)	3.5 (.07)	6.10*	.040
Parenting style (α = .60)	4.4 (.04)	4.2 (.05)	10.76**	.070

Note. CPA = child physical abuse. PCERA = parent child early relational assessment. * p < .05; ** p < .01.

^aHigher PCERA scores reflect higher quality caregiving.

Overall, the quality of caregiving behavior decreased significantly across time, F (5, 735) = 13.68, p < .001, η²_p = 0.085 (see Figure 2). Follow-up tests were conducted to examine pairwise comparisons of caregiving quality across time blocks. Results revealed the following pattern across the six 5-min blocks (blocks with the same superscript did not differ significantly with respect to caregiving quality): Block 1 (M_adj = 4.28; SE = 0.038) > Block 2 (M_adj = 4.04; SE = 0.041) > Block 3^a (M_adj = 3.87; SE = 0.040), Block 4^ab (M_adj = 3.75; SE = 0.045), Block 5^bc (M_adj = 3.65; SE = 0.043), Block 6^c (M_adj = 3.57; SE = 0.045).

Figure 2.

Mean caregiving quality scores for low-risk caregivers, high-risk completers, and high-risk noncompleters across 5-min time blocks during the simulated infant task.

Supplemental Analyses. As noted above, approximately one quarter (26.7%) of the high-risk caregivers elected to discontinue the simulated infant session early, limiting the amount of behavioral data available for analyses involving this subgroup. Nonetheless, inspection of the behavioral data for the high-risk noncompleters at the outset of the session (e.g., first 10 min) provides insights into how this subgroup of caregivers reacted to the simulated infant task. Toward this end, we conducted a supplemental ANOVA designed to compare quality of caregiving exhibited by high-risk completers and high-risk noncompleters during the first 10-min of the infant simulator session. Specifically, PCERA total scores for 66 high-risk completers and 22 high-risk noncompleters were compared using 2 (Completion status: completers, noncompleters) × 2 (Time: Blocks 1 and 2) mixed factorial ANCOVAs (with repeated measures on the last factor). Parent gender was not included in this analysis due to small cell sizes. Results revealed significant main effects for Time and CPA risk for PCERA total scores. Overall, the quality of caregiving behavior decreased from Block 1 to Block 2, F (1, 84) = 8.50, p = .005, η²_p = 0.092 (see Figure 2). Of particular relevance in the present study, high-risk noncompleters exhibited lower quality caregiving than high-risk completers during the first 10-min of the simulated infant session, F (1, 84) = 4.70, p = .033, η²_p = 0.053 (see Figure 2).

Negative Affect over Time

As described above, negative affect ratings were reported at the beginning of the data collection session (i.e., baseline), immediately before the simulated infant task began, and immediately after the simulated infant task ended. Mean negative affect scores were analyzed using a 2 (CPA risk: low, high) × 2 (Gender: female, male) × 3 (Time: baseline, pre-Sim task, post-Sim task) mixed factorial ANOVA (with repeated measures on the last factor). To explore the negative affect ratings of noncompleters, supplemental analyses were conducted in which we compared the negative affect ratings of high-risk completers and high-risk noncompleters.

The planned analyses examining mean negative affect ratings revealed that none of the three-way (CPA risk x Gender x Time) or two-way (CPA risk x Time, Gender x Time, CPA risk x Gender) interactions were significant (all p’s > .05 and η²_p’s < 0.01). The main effect of Gender was not significant, F (1, 180) = 1.57, p = .212, η²_p = 0.009. As expected, significant main effects for Time and CPA risk were observed. Mean negative affect scores increased significantly across time, F (2, 360) = 14.50, p < .001, η²_p = 0.075. Pairwise comparisons revealed the following pattern for negative affect scores over time: Time 1 (M_adj = 1.29; SE = 0.036) < Time 2 (M _adj = 1.41; SE = 0.037) < Time 3 (M _adj = 1.52; SE = 0.045). Compared to low-risk caregivers, high-risk caregivers reported higher levels of negative affect, F (1, 180) = 20.66, p < .001, η²_p = 0.103 (see Figure 3). Supplemental analyses revealed that the negative affect scores of high-risk completers and high-risk noncompleters did not differ significantly at baseline, F (1, 88) = 0.87, p = .354, η²_p = 0.010; before the simulated infant task, F (1, 88) = 0.99, p = .321, η²_p = 0.011; or after the simulated infant task: F (1, 88) = 0.09, p = .768, η²_p = 0.001.

Figure 3.

Mean negative affect scores for low-risk caregivers and high-risk caregivers as reported at baseline, before the simulated infant task, and after the simulated infant task.

Discussion

The present study examined how caregivers with varying levels of CPA risk responded to prolonged, inconsolable crying, as modeled by an infant simulator. On average, caregiving quality observed during the simulated infant task was high; however, the quality of caregiving diminished over the course of the 30-min simulated infant task, with scores declining toward the midpoint of the rating scale over time. The present findings help illuminate the impact of prolonged infant crying on caregiver behavior over the span of 30 min in a controlled laboratory setting. In real life, the challenges of dealing with infant crying may be even greater given that infant crying bouts may last as long as 2 hours at the peak of the crying curve and occur under less controlled conditions as caregivers struggle to deal with competing demands (e.g., caring for other children) and additional stressors (e.g., resource strain, relationship difficulties).

Findings from this study also revealed that high-risk caregivers, compared to low-risk caregivers, exhibited lower quality caregiving as they tried to soothe the inconsolable, simulated crying infant. Inspection of the subscale scores for the PCERA revealed that, compared to low-risk caregivers, high-risk caregivers displayed more negative mood (i.e., appeared more depressed, withdrawn, and less enthusiastic) during the simulated infant task. High-risk caregivers also displayed less positive affective and behavioral involvement, as evidenced by fewer instances of positive physical contact, lower quality verbalizations, and fewer attempts to structure the environment as they tried to soothe the simulated infant. In addition, the parenting style of high-risk caregivers was more intrusive, less consistent, and less predictable.

The lower caregiving quality exhibited by high-risk caregivers, compared to low-risk caregivers, is noteworthy in light of the fact that quality of parental responses to infant distress is theorized to significantly influence children’s subsequent behavioral adjustment, affect regulation, and social competence (Leerkes et al., 2009). Thus, the lower quality caregiving exhibited by high-risk caregivers during the simulated infant task may help explain results from prospective research indicating that higher CPA risk in new and expectant mothers is associated with higher levels of neonatal morbidity and subsequent developmental problems (Dukewich et al., 1999; Zelenko et al., 2001). For example, Zelenko et al. (2001) found that prenatal levels of maternal CPA risk predicted neonatal morbidity and this association remained significant even after obstetric risk factors were statistically controlled. Moreover, higher CPA risk among mothers of young children (i.e., 1 and 3 years of age) prospectively predicted lower levels of adaptive behavior and intelligence in children 2 years later (Dukewich et al., 1999).

When interpreting the CPA risk group differences observed in the present study it is important to note that differential completion rates occurred, such that high-risk caregivers, compared to low-risk caregivers, more often elected to stop the simulated infant task before the conclusion of the 30-min session. Indeed, roughly one quarter (26.7%) of high-risk caregivers chose to end the simulated infant task early, whereas only 7.4% of low-risk caregivers asked to stop before completing the task. Compared to high-risk caregivers who completed the 30-min simulated infant task, high-risk noncompleters tended to have higher CAP abuse scores and they exhibited lower quality caregiving during the initial 10 min of the simulated infant task (see Figure 2). Thus, behavioral data were incomplete for some of the highest risk caregivers who on average exhibited the poorest quality caregiving at the beginning of the task. Collectively, these findings suggest that the CPA risk group differences observed in the present study may under-represent differences in caregiving quality between high-risk caregivers and low-risk caregivers.

The lack of persistence observed among high-risk parents during the simulated infant task is similar to findings reported by Rodriguez et al. (2015), who found that CPA risk scores were inversely associated with time spent attempting to solve an unsolvable maze while listening to recordings of a crying baby or an upset toddler. Although there are a variety of potential explanations for lack of persistence during laboratory tasks (e.g., disinterest, low motivation, depression), Rodriguez and colleagues suggested that high-risk parents withdraw from challenging tasks because they have difficulty tolerating frustration. Although we did not assess frustration levels during the simulated infant task, caregivers were asked to rate how much negative affect they felt at baseline, immediately before, and immediately after the simulated infant task. As expected, negative affect increased over time and was higher among high-risk, as compared to low-risk, caregivers (see Figure 3). However, negative affect ratings did not differ for high-risk completers compared to high-risk non-completers, suggesting that factors other than negative affective states impacted persistence among high-risk parents during the simulated infant task.

Nonetheless, it is noteworthy that high-risk, compared to low-risk, caregivers reported higher levels of negative affect throughout the data collection session, with the highest level of negative affect reported by high-risk caregivers after the simulated infant task. As has been theorized by others, negative affective states may, in turn, trigger negative attributions, evaluations, and expectations, which may intensify negative internal states and potentially trigger an aggressive response (Berkowitz, 1990; Milner, 1993, 2000). In keeping with this theorizing, Fairbrother et al. (2014) reported evidence that, in addition to eliciting negative affective states (e.g., distress, irritation, frustration), exposure to infant crying may trigger thoughts of infant-directed harm and urges to withdraw (escape) from the crying infant. Although no overtly aggressive behaviors were observed in the present study, approximately one quarter of the high-risk caregivers opted to withdraw from the simulated crying infant prior to completing the task. Regarding the absence of aggressive behaviors, it seems reasonable to assume that participants were aware that they were being observed/recorded throughout the simulated infant session. As a result, participants may have intentionally limited expression of less desirable thoughts, feelings and behaviors in the laboratory context.

Despite the aforementioned concerns, the present study represents a methodological advance over prior research examining how high-risk caregivers respond to infant crying. Specifically, the present study provided an opportunity for caregivers to actively attempt to soothe a simulated inconsolable infant over an extended time period (i.e., 30 min). Although the differential attrition observed in the present study was, in and of itself, an interesting finding, the loss of roughly one quarter of the high-risk caregivers prior to completion of the simulated infant task limited our ability to observe the behavior of this important subgroup. Additional research clarifying the factors that lead some high-risk caregivers to withdraw prematurely from challenging tasks could have important implications for both clinical practice and future research.

The present study also was limited with respect to the methods used to assess negative affect. Although self-reported negative affect ratings were obtained at three time points (baseline, pre-simulated infant task, and post-simulated infant task), it is possible that these ratings do not accurately capture negative affect experienced while the simulated infant was crying. For example, the pre-simulated infant task ratings could reflect some degree of anticipatory anxiety, while the post-simulated infant task ratings may reflect some degree of relief given that the infant simulator task concluded before participants made their final affect ratings. Although a continuous rating of negative affect during the simulated infant task was desirable (e.g., see Barr et al., 2014, for an example), asking participants to stop and rate their feelings would have interfered with the primary task of attempting to soothe the crying simulated infant. At a minimum, future research would benefit from including items assessing additional negative affective states (e.g., frustration, anger) in an effort to extend the findings of this study.

Although the sample recruited in the present study was diverse with respect to its demographic characteristics, the overall sample size limited our ability to conduct subgroup analyses. Analyses exploring gender differences revealed little evidence of differences between men and women in their behavioral and affective responses to the simulated crying infant, suggesting that the impact of prolonged infant crying on men and women may be more similar than different.

Summary

Prolonged infant crying is a stressor encountered by caregivers around the world, and the quality of the caregiving response (ranging from aggression to sensitive soothing) can have an enduring impact on an infant’s subsequent development. The simulated infant paradigm used in the present study provided an opportunity to observe caregivers as they attempted to soothe an inconsolable crying infant. Generally, caregiving quality diminished over the course of the 30 min task, which likely reflects the challenging nature of the task. The present findings reveal that high-risk caregivers were more likely to withdraw early from the caregiving task and provided lower quality care when they endeavored to soothe the crying simulated infant. In addition, high-risk, compared to low-risk, caregivers reported higher levels of negative affect throughout the data collection session, with the highest level of negative affect reported by high-risk caregivers after completing the simulated infant task. Additional research is needed to examine the extent to which other aspects of functioning (e.g., cognitions, executive functioning) influence how high-risk caregivers respond to prolonged infant crying. Collectively, the present findings highlight the importance of early intervention efforts (e.g., Shah et al., 2016) designed to support caregivers’ abilities to respond effectively to prolonged infant crying.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Institute of Child Health and Human Development (grant number 1R15HD080041).

ORCID iD

Julie L. Crouch

Author Biographies

Julie L. Crouch is the Director of the Center for the Study of Family Violence and Sexual Assault at NIU.

David Bridgett is a Professor in the Department of Psychology and a research associate at the Center for the Study of Family Violence and Sexual Assault at NIU.

Joel S. Milner is a Distinguished Research Professor at the Center for the Study of Family Violence and Sexual Assault at NIU.

Kreila Cote was a research assistant at the Center for the Study of Family Violence and Sexual Assault at NIU.

Gabriela Lelakowska was a research assistant at the Center for the Study of Family Violence and Sexual Assault at NIU.

America L. Davila was a graduate student in the Department of Psychology and research assistant at the Center for the Study of Family Violence and Sexual Assault at NIU.

Erin McKay is a graduate student in the Department of Psychology and research assistant at the Center for the Study of Family Violence and Sexual Assault at NIU.

Shelby Savoree is a research assistant at the Center for the Study of Family Violence and Sexual Assault at NIU.

References

Adamsbaum

Grabar

Mejean

Rey-Salmon

(2010). Abusive head trauma: Judicial admissions highlight violent and repetitive shaking. Pediatrics, 126(3), 546–555. https://doi.org/10.1542/peds.209-3647

Anke

T. M. S.

Slinning

Moe

Brunborg

Siqveland

T. S.

Skjelstad

D. V.

(2019). Mothers with and without bipolar disorder and their infants: Group differences in mother-infant interaction patterns at three months postpartum. BMC Psychiatry, 19(1), 292. https://doi.org/10.1186/s12888-019-2275-4

Bakermans-Kranenburg

M. J.

Alink

L. R. A.

Biro

Voorthuis

van Ijzendoorn

M. H.

(2015). The Leiden infant simulator sensitivity assessment (LISSA): Parenting an infant simulator as your own baby. Infant and Child Development, 24(3), 220–227. https://doi.org/10.1002/icd.1905

Barnett

J. E.

Hurst

C. S.

(2003). Abstinence education for rural youth: An evaluation of the life’s walk program. Journal of School Health, 73(7), 264–268. https://doi.org/10.1111/j.1746-1561.2003.tb06576.x

Barr

R. G.

(1990). The normal crying curve: What do we really know? Developmental Medicine and Child Neurology, 32(4), 356–362. https://doi.org/10.1111/j.1469-8749.1990.tb16949.x

Barr

R. G.

(2012). Preventing abusive head trauma resulting from a failure of normal interaction between infants and their caregivers. Proceedings of the National Academy of Sciences, 109(Suppl 2), 17294–17301. https://doi.org/10.1073/pnas.1121267109

Barr

R. G.

Fairbrother

Pauwels

Green

Chen

Brant

(2014). Maternal frustration, emotional and behavioral responses to prolonged infant crying. Infant Behavior and Development, 37(4), 652–664. https://doi.org/10.1016/j.infbeh.2014.08.012

Berkowitz

(1990). On the formation and regulation of anger and aggression: A cognitive-neoassociationistic analysis. American Psychologist, 45(4), 494–503. https://doi.org/10.1037/0003-066X.45.4.494

Cerezo

M. A.

Pons-Salvador

Trenado

R. M.

(2008). Mother–infant interaction and children’s socio-emotional development with high- and low-risk mothers. Infant Behavior and Development, 31(4), 578–589. https://doi.org/10.1016/j.infbeh.2008.07.010

10.

Chaffin

Valle

L. A.

(2003). Dynamic prediction characteristics of the child abuse potential inventory. Child Abuse & Neglect, 27(5), 463–481. https://doi.org/10.1016/S0145-2134(03)00036-X

11.

Clark

(1985). The parent-child early relational assessment: Instrument and manual. University of Wisconsin Medical School, Department of Psychiatry.

12.

Clark

(1999). The parent-child early relational assessment: A factorial validity study. Educational and Psychological Measurement, 59(5), 821–846. https://doi.org/10.1177/00131649921970161

13.

Compier-de Block

L. H. C. G.

Alink

L. R. A.

Reijman

Werner

C. D.

Maras

Rijnberk

van Ijzendoorn

M. H.

Bakermans-Kranenburg

M. J.

(2015). Handgrip force of maltreating mothers in reaction to infant signals. Child Abuse & Neglect, 40, 124–131. https://doi.org/10.1016/j.chiabu.2014.03.006

14.

Crouch

J. L.

Irwin

L. M.

Wells

B. M.

Shelton

C. R.

Skowronski

J. J.

Milner

J. S.

(2012). The Word Game: An innovative strategy for assessing implicit processes in parents at risk for child physical abuse. Child Abuse & Neglect, 36(6), 498–509. https://doi.org/10.1016/j.chiabu.2012.04.004

15.

Crouch

J. L.

Skowronski

J. J.

Milner

J. S.

Harris

(2008). Parental responses to infant crying: The influence of child physical abuse risk and hostile priming. Child Abuse & Neglect, 32(7), 702–710. https://doi.org/10.1016/j.chiabu.2007.11.002

16.

Crowe

H. P.

Zeskind

P. S.

(1992). Psychophysiological and perceptual responses to infant cries varying in pitch: Comparison of adults with low and high scores on the Child Abuse Potential Inventory. Child Abuse & Neglect, 16(1), 19–29. https://doi.org/10.1016/0145-2134(92)90005-C

17.

Dukewich

T. L.

Borkowski

J. G.

Whitman

T. L.

(1999). A longitudinal analysis of maternal abuse potential and developmental delays in children of adolescent mothers. Child Abuse & Neglect, 23(5), 405–420. https://doi.org/10.1016/S0145-2134(99)00019-8

18.

Eismann

E. A.

Theuerling

Cassedy

Curry

P. A.

Colliers

Makoroff

K. L.

(2020). Early developmental, behavioral, and quality of life outcomes following abusive head trauma in infants. Child Abuse & Neglect, 108, 104643. https://doi.org/10.1016/j.chiabu.2020.104643

19.

Fairbrother

Barr

R. G.

Pauwels

Brant

Green

(2014). Maternal thoughts of harm in response to infant crying: An experimental analysis. Archives of Women's Mental Health, 18(3), 447–455. https://doi.org/10.1007/s00737-014-0471-2

20.

Frodi

A. M.

Lamb

M. E.

(1980). Child abusers' responses to infant smiles and cries. Child Development, 51(1), 238–241. https://doi.org/10.2307/1129612

21.

Fujiwara

Barr

R. G.

Brant

Barr

(2011). Infant distress at five weeks of age and caregiver frustration. Journal of Pediatrics, 159(3), 425–430. https://doi.org/10.1016/j.jpeds.2011.02.010

22.

Hechler

Beijers

Riksen-Walraven

de Weerth

(2019). Prenatal predictors of postnatal quality of caregiving in mothers and fathers. Parenting: Science and Practice, 19(1–2), 101–119. https://doi.org/10.1080/15295192.2019.1556010

23.

Leerkes

E. M.

Blankson

A. N.

O’Brien

(2009). Differential effects of maternal sensitivity to infant distress and nondistress on socio-emotional functioning. Child Development, 80(3), 762–775. https://doi.org/10.1111/j.1467-8624.2009.01296.x

24.

Leventhal

J. M.

Martin

K. D.

Asnes

A. G.

(2010). Fractures and traumatic brain injuries: Abuse versus accidents in a US database of hospitalized children. Pediatrics, 126(1), Article e104–e115. https://doi.org/10.1542/peds.2009-1076

25.

Lind

Toure

Brugel

Meyer

Laurent-Vannier

Chevignard

(2016). Extended follow-up of neurological, cognitive, behavioral and academic outcomes after severe abusive head trauma. Child Abuse & Neglect, 51, 358–367. https://doi.org/10.1016/j.chiabu.2015.08.001

26.

Manfield

Oakley

Macey

J. A.

Waugh

M. C.

(2021). Understanding the five-year outcomes of abusive head trauma in children: A retrospective cohort study. Developmental Neurorehabilitation, 24(6), 361–367. https://doi.org/10.1080/17518423.2020.1869340

27.

McKay

E. R.

Rosinski

Mayes

L. C.

Rutherford

H. J. V.

Bridgett

D. J.

(2019). Maternal positive responses to a distressed infant simulator predict subsequent negative affect in infants. Infant Behavior and Development, 56, Article e101299. https://doi.org/10.1016/j.infbeh.2018.12.004

28.

Milner

J. S.

(1986). The child abuse potential inventory: Manual. Psytec.

29.

Milner

J. S.

(1993). Social information processing and physical child abuse. Clinical Psychology Review, 13(3), 275–294. https://doi.org/10.1016/0272-7358(93)90024-G

30.

Milner

J. S.

(2000). Social information processing and child physical abuse: Theory and research. In Hansen

D. J.

(Ed.), Nebraska Symposium on Motivation (Vol. 46, pp. 39–84). University of Nebraska Press.

31.

Milner

J. S.

Crouch

J. L.

(2017). Child physical abuse risk assessment: Parent and family evaluations. In Campbell

J. C.

Messing

J. T.

(Eds.), Assessing dangerousness: Domestic violence offenders and child abusers (3rd ed.), pp. 55–87). Springer. https://doi.org/10.1891/9780826133274.0003

32.

Milner

J. S.

Gold

R. G.

Ayoub

C. A.

Jacewitz

M. M.

(1984). Predictive validity of the child abuse potential inventory. Journal of Consulting and Clinical Psychology, 52(5), 879–884. https://doi.org/10.1037/0022-066X.52.5.879

33.

Narang

Clarke

(2014). Abusive head trauma: Past, present, and future. Journal of Child Neurology, 29(12), 1747–1756. https://doi:10.1177/0883073814549995

34.

Palusci

V. J.

Covington

T. M.

(2014). Child maltreatment deaths in the U.S. National child death review case reporting system. Child Abuse & Neglect, 38(1), 25–36. https://doi.org/10.1016/j.chiabu.2013.08.014

35.

Reijneveld

S. A.

van der Wal

M. F.

Brugman

Sing

R. A. H.

Verloove-Vanhorick

S. P.

(2004). Infant crying and abuse. Lancet, 364(9442), 1340–1342. https://doi.org/10.1016/S0140-6736(04)17191-2

36.

Roberts

S. W.

McCowan

R. J.

(2004). The effectiveness of infant simulators. Adolescence, 39(155), 475–487.

37.

Rodriguez

C. M.

(2010). Parent-child aggression: Association with child abuse potential and parenting styles. Violence and Victims, 25(6), 728–741. https://doi.org/10.1891/0886-6708.25.6.728

38.

Rodriguez

C. M.

Russa

M. B.

Kircher

J. C.

(2015). Analog assessment of frustration tolerance: Association with self-reported child abuse risk and physiological reactivity. Child Abuse & Neglect, 46, 121–131. https://doi.org/10.1016/j.chiabu.2015.02.017

39.

Rutherford

H. J. V.

(2019). The infant simulator: A novel approach for the measurement of parenting. Infant Behavior and Development, 56, 101348. https://doi.org/10-1016/j.infbeh.2019.101348

40.

Rutherford

H. J. V.

Goldberg

Luyten

Bridgett

D. J.

Mayes

L. C.

(2013). Parental reflective functioning is associated with tolerance of infant distress but not general distress: Evidence for a specific relationship using a simulated baby paradigm. Infant Behavior and Development, 36(4), 534–547. https://doi.org/10.1016/j.infbeh.2013.06.008

41.

Scribano

P. V.

Makoroff

K. L.

Feldman

K. W.

Berger

R. P.

(2013). Association of perpetrator relationship to abusive head trauma clinical outcomes. Child Abuse & Neglect, 37(10), 771–777. https://doi.org/10.1016/j.chiabu.2013.04.011

42.

Shah

Kennedy

Clark

M. D.

Bauer

S. C.

Schwartz

(2016). Primary care-based interventions to promote positive parenting behaviors: A meta-analysis. Pediatrics, 13(5), Article e20153393. https://doi.org/10.1542/peds.2015-3393

43.

Voorthuis

Out

van der Veen

Bhandari

van IJzendoorn

M. H.

Bakermans-Kranenburg

M. J.

(2013). One doll fits all: Validation of the Leiden infant simulator sensitivity assessment (LISSA). Attachment & Human Development, 15(5-6), 603–617. https://doi.org/10.1080/14616734.2013.820897

44.

Watson

Clark

L. A.

Tellegen

(1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54(6), 1063–1070. https://doi.org/10.1037/0022-3514.54.6.1063

45.

Zelenko

M. A.

Huffman

L. C.

Brown

B. W.

Jr. Daniels

Lock

Kennedy

Steiner

(2001). The Child Abuse Potential Inventory and pregnancy outcome in expectant adolescent mothers. Child Abuse & Neglect, 25(11), 1481–1495. https://doi.org/10.1016/S0145-2134(01)00284-8