Sustained Attention in Infancy: A Foundation for the Development of Multiple Aspects of Self-Regulation for Children in Poverty

Abstract

There are many avenues by which early life poverty relates to the development of school readiness. Few studies, however, have examined the extent to which sustained attention, a central component of self-regulation in infancy, mediates relations between poverty-related risk and cognitive and emotional self-regulation at school entry. To investigate longitudinal relations among poverty-related risk, sustained attention in infancy, and self-regulation prior to school entry, we analyzed data from The Family Life Project, a large prospective longitudinal sample (N=1,292) of children and their primary caregivers in predominantly low-income and nonurban communities. We used structural equation modeling to assess the extent to which a latent variable of infant sustained attention, measured in a naturalistic setting, mediated the associations between cumulative poverty-related risk and three domains of self-regulation. We constructed a latent variable of infant sustained attention composed of a measure of global sustained attention and a task-based sustained attention measure at ages 7 and 15 months. Results indicated that infant sustained attention was negatively associated with poverty-related risk and positively associated with a direct assessment of executive function abilities and teacher reported effortful control and emotion regulation in pre-kindergarten. Mediation analysis indicated that the association between poverty-related risk and each self-regulation outcome was partially mediated by infant attention. These results provide support for a developmental model of self-regulation whereby attentional abilities in infancy act as a mechanism linking the effects of early-life socioeconomic adversity with multiple aspects of self-regulation in early childhood.

Increasingly, research demonstrates that self-regulation is a critical component of healthy child development (Blair & Raver, 2016; Ursache, Blair, & Raver, 2012). Self-regulation describes the growing ability of the child to adapt behavior to the context in which the behavior is occurring—to be studious and attentive in the classroom during a lesson in mathematics, but to then run and play on the playground during recess. More specifically, self-regulation is an integrative construct that describes the multiple ways in which social, emotional, behavioral, cognitive, and physiological aspects of the person are organized and influence a response to a given stimulus (Posner & Rothbart, 2007; Rothbart, 2004). While research on self-regulation has increased since the turn of the 21^st century, more studies are needed on the early antecedents of self-regulation that contribute to variation in its emotional and cognitive aspects. Work in this area is particularly warranted to understand why children living in poverty demonstrate disparities in early school readiness. Research and theory suggest that lower-order cognitive abilities, such as attention, act as early precursors for more advanced, higher-order self-regulation processes (Posner & Rothbart, 2000; Ursache, Blair, Stifter, & Voegtline, 2013). Therefore, the goal of the current work was to investigate whether attention processes emerging in early infancy may set the foundation for the development of multiple aspects of self-regulation, particularly for children who live in socioeconomic risk.

Theoretical Models of Self-Regulation

Self-regulation is comprised of cognitive, behavioral, and emotional components and best characterized by the by the reciprocal interaction of conscious, effortful, and reflective aspects of the person with non-conscious, automatic, and reactive aspects of emotional and physiological responses to stimulation (Blair & Ursache, 2014; Blair & Raver, 2016). The origin of this model of self-regulation is found in the theoretical model of temperament proposed by Rothbart, Posner, and collaborators (Derryberry & Rothbart, 1997; Posner & Rothbart, 2000; Rothbart & Ahadi, 2001). Broadly, this model conceptualizes self-regulation as the balance and interplay between bottom-up reactivity and top-down regulation, which are predominantly enduring, biologically-based behaviors present at birth. Research supports a neurobiological model of this relation suggesting that self-regulation relies on a bidirectional relationship between the limbic system and the prefrontal cortex (PFC) in which the top-down activity of the PFC regulates the bottom-up reactivity in the limbic system (Arnsten, 2009). Notably, both the PFC and limbic system are susceptible to environmental influence, especially early in development (Hensch, 2005; Mackey, Raizada, & Bunge, 2014). Variability within the child’s environmental context can bias the PFC-limbic network to be more or less reactive or reflective, thereby supporting or undermining self-regulation capacities (Blair, 2010). In this regard, the PFC is of particular importance given its ability to down-regulate reactivity within the limbic system and its prolonged susceptibility to environmental influences throughout childhood. Moreover, the protracted nature of PFC development highlights the importance of early life experience in promoting the development of adaptive top-down control over bottom-up reactivity. Therefore, while temperament models typically highlight reactivity and regulation to be trait-like and relatively fixed, more recent developmental and neuroscience research suggests that processes related to self-regulation and reactivity are sensitive to environmental influences (Diamond, 2009). Evidence for the plasticity of self-regulation indicates that its development, especially early in life, is influenced by the context of the home and family environment, and the quality of early parenting (Blair, 2010; Raver, 2004; Zeytinoglu, Calkins, Swingler, & Leerkes, 2017). Although the relations between environmental experience and self-regulation have been extensively studied, the developmental origins of these associations remain largely unknown. Given the malleability of the PFC in early life, self-regulation disparities may emerge around the same time that some of the earliest PFC-dependent cognitive control processes, such as attention, begin to develop in infancy.

The Development and Measurement of Infant Sustained Attention

Attention, like self-regulation, is a multidimensional construct. Sustained attention, in particular, emerges within the first year of life and continues to develop across childhood (Amso & Scerif, 2015; Reynolds & Romano, 2016; Rose, Feldman, & Jankowski, 2001; Ruff & Lawson, 1990). An infant’s ability to sustain attention is a core component of self-regulation and thus, is critical to development (Casey & Richards, 1988; Ruff, 1986; Swingler, Perry, & Calkins, 2015). For instance, infants focus and sustain their attention to stimuli in their environment to support their volitional control of behavior (Mary K Rothbart & Rueda, 2005; H. A. Ruff & Capozzoli, 2003). Specifically, by practicing sustained attention, infants’ are better able to resolve internal or external conflicts and guide adaptive responses and decisions (Rothbart, Sheese, Rueda, & Posner, 2011), thus setting the stage for self-regulation.

Measurement of sustained attention in infancy is difficult, with most studies using indirect methods of assessment to elicit and infer attentional processes. For instance, sustained attention in infancy has been indexed behaviorally during looking paradigms in the laboratory. Research in this domain has found that sustained attention manifests behaviorally in the form of prolonged gaze, decreased distractibility, and object manipulation (Ruff & Capozzoli, 2003; Ruff and Rothbart, 1996). Ruff (1986) has postulated that the amount of time an infant spends attending to an object promotes stimuli processing and object learning. Further, research has demonstrated that infants visually focused to objects for sustained periods were less prone to distractions (Oakes & Tellinghuisen, 1994).

Consequently, sustained attention has often been indexed by look duration and frequency during structured computer-based tasks or semi-naturalistic interactions in the laboratory (B. Casey & Richards, 1988; Pérez-Edgar et al., 2010). These lab-based tasks frequently rely on attention that is intentionally elicited. However, sustained attention has also been measured as it occurs spontaneously during naturalistic paradigms in the home by observing global patterns of sustained attention during unstructured tasks, such as free play or parent-child interactions (Johansson, Marciszko, Brocki, & Bohlin, 2016; Rothbart, Sheese, Rueda, & Posner, 2011; Towe-Goodman, Stifter, Coccia, Cox, & Family Life Project Key Investigators, 2011).

The methodological distinction between lab-based elicited attention and home-based spontaneous attention is important because, as it functions in the real world, infant attention is largely guided by social contexts and social interactions especially in the home (e.g., Colombo & Salley, 2015; Kopp, 1982; Miller, Ables, King, & West, 2009; Parrinello & Ruff, 1988; Sigel, 2002; Spruijt, Dekker, Ziermans, & Swaab, 2018; Vygotsky, 1978). However, relatively little research has studied sustained attention as it naturally occurs, outside of the laboratory, in an infant’s day-to-day environment. During everyday interactions in the home, caregivers offer affordances and set constraints to guide children’s sustained attention and avoid distractions (Suarez-Rivera, Smith, & Yu, 2019; S. V Wass et al., 2018; Yu & Smith, 2016). As such, in order to better understand the process and development of sustained attention, it is important to assess sustained attention in contexts in which it naturally occurs and develops.

Attention in Infancy and Self-Regulation at School Entry

In the current analysis we focus on three manifestations of self-regulation: executive function, emotion regulation, and effortful control, and their relation to infant sustained attention. Executive functions refers to cognitive abilities associated with inhibitory control, working memory, and attention shifting, which are recruited to execute problem solving and goal-directed planning (Miyake et al., 2000). Emotion regulation is an interrelated, yet distinct component of self-regulation, which involves the behavioral and cognitive modulation of affective experiences and expressions (Calkins & Hill, 2007). At the intersection of emotion regulation and executive functions lies effortful control, which can be conceptualized as the ability to inhibit a dominant or impulsive response, especially in emotionally valenced contexts (Blair & Ursache, 2011). Although these three domains of self-regulation are interrelated, effortful control is generally considered to be more behavioral (Rothbart & Rueda, 2005), while executive function is more cognitive (Blair, 2016; Posner & Rothbart, 2000) (Posner & Rothbart, 2000), and emotion regulation is primarily affective (Derryberry & Rothbart, 1997; Ursache, Blair, Stifter, & Voegtline, 2013). Further, previous studies have demonstrated only moderate correlations among these constructs, suggesting they constitute distinct, but interdependent components of self-regulation (Blair, Ursache, Greenberg, & Vernon-Feagans, 2015; Liew, 2012; Zhou, Chen, & Main, 2012).

The developmental link between early attention and self-regulation requires further investigation. However, theory and research suggest that the association between attention in infancy and self-regulation in childhood may be supported by a common mechanism underlying both processes, such as the ability to inhibit behavior, ignore distractions, and focus on relevant stimuli in the environment (Colombo & Cheatham, 2006; Fox & Calkins, 2003; Posner, 2012). Multiple studies have found associations with increased visual processing efficiency (a component of infant attention) and domains of self-regulation (Cuevas & Bell, 2014; Rose, Feldman, & Jankowski, 2012; Sigman, Cohen, & Beckwith, 1997). However, researchers have indicated the need to differentiate speed of information processing from the more complex aspects of focused or sustained attention and their relation to cognitive and behavioral outcomes (Courage, Reynolds, & Richards, 2006). To this end, some research suggests that sustained attention observed during free play in infancy is positively related working memory performance (Johansson et al., 2016) and global executive function performance in toddlerhood (Frick et al., 2018; Johansson, Marciszko, Gredebäck, Nyström, & Bohlin, 2015). This longitudinal work offers preliminary support for infant sustained attention as an important component of executive function development (Posner & Rothbart, 2000; Rothbart et al., 1994).

Further, by measuring look duration during a semi-structured play task, research found infant sustained attention to predict parent reported effortful control in toddlerhood (Kochanska, Murray, and Harlan 2000). Additional evidence from eye-tracking research corroborated these findings by demonstrating a positive association between duration of visual fixation to a computerized multimodal object in infancy and parent reported effortful control at age 4 (Papageorgiou et al., 2014). Finally, researchers measuring global patterns of sustained attention during a free play task found a positive relation with effortful control in toddlerhood (Johansson et al., 2015). These findings collectively support the theory that infant sustained attention, indexed through both task-specific paradigms and global observations are longitudinally associated with effortful control.

Finally, attentional control is closely related to children’s ability to regulate their emotions. For instance, previous work has demonstrated that infants with lower levels of sustained attention were more likely to demonstrate social anxiety during a structured social interaction scenario in adolescence (Pérez-Edgar et al., 2010). Additional support from neurobiological studies shows that higher frontal electroencephalography (EEG) power (a neural marker of top-down attention control) during an attention task at 10 months of age negatively predicted frustration response to an emotional challenge at 3 years of age. Importantly, this association between EEG power and frustration was mediated by observed infant sustained attention at 10 months (Perry, Swingler, Calkins, & Bell, 2016). These findings demonstrate additional biologically based evidence for the role of infant attention in modulating development of emotion regulation. Together, these findings suggest that attention serves as a gatekeeper for regulating processes underlying emotion regulation.

Attention & Self-Regulation in the Context of Poverty

Children living in poverty are more likely to live in conditions that are less supportive of self-regulation development (Blair & Raver, 2015; Blair & Razza, 2007; Ursache et al., 2012). Given that self-regulation is fundamental to children’s school readiness (McClelland et al., 2007), it has been proposed as a mechanism explaining gaps in school readiness associated with poverty (Blair & Raver, 2015). Indeed, substantial evidence has found that by the time children enter the school system, they show considerable disparities in the cognitive, behavioral, and emotional domains of self-regulation (Blair, Raver, Granger, Mills-Koonce, & Hibel, 2011; Blair, Granger, et al., 2011; Raver, Blair, & Willoughby, 2013). Critically, the biological underpinnings of early self-regulation components, such as attention, develop rapidly in the first year of life and as such are especially susceptible to environments of early life adversity and stress (Cerqueira, Mailliet, Almeida, Jay, & Sousa, 2007; Grossmann, 2013; Hodel, 2018). As a result, poverty-related risk factors may compromise the development of self-regulation and a potential mechanism through which these effects may occur is early attention. Only recently have the effects of psychosocial adversity on attention in infancy been demonstrated, however. Clearfield & Jedd (2013) found that across the first year of life, infants from lower socioeconomic status (SES) backgrounds spent less time demonstrating focused attention towards people and toys during a free play task. Moreover, Lipina, Martelli, Vuelta & Colombo (2005) demonstrated similar disparities in early life executive function skills, measured with the A-not-B task, based on income background. These behavioral findings are substantiated by neuroimaging research during infancy and toddlerhood suggesting that socioeconomic factors are associated with reduced gray matter growth in frontal and parietal lobes (Hanson et al. 2013) and decreased EEG activity in frequency bands associated with sustained attention (Tomalski, 2013) and executive attention (Conejero, 2016). Collectively, the existing literature provides preliminary support for the susceptibility of frontal cortical-dependent processes to early psychosocial adversity. However, to date, prior research has not explicitly examined sustained attention in infancy as a mediating pathway linking early-life poverty-related adversity and self-regulation in early childhood.

The Present Study

Most prior research has measured attention using a single attention variable derived from standard laboratory tasks. However, the measurement of attention in infancy is subject to error and the generalizability from the lab to the home environment is uncertain (Hughes, 2011; Schmuckler, 2001). Here we aim to extend current knowledge of attention in infancy by examining how a multimethod, multi-time point measure of naturalistic sustained attention in infancy is associated with a cumulative measure of poverty-related risk, and whether sustained attention mediates the influence of poverty-related risk on multiple aspects of self-regulation at school entry. In doing so, we examine sustained attention as an early precursor of self-regulation abilities that may be malleable and therefore serve as a focus for efforts to promote healthy child development in the infant and toddler periods. Therefore, the purpose of the present work was to longitudinally examine how early poverty-related adversity related to infant sustained attention with potential implications for the development of self-regulation in early childhood. We broadly conceptualize infant sustained attention in the current analysis by using more global measurements as a means to improve external validity and expand on previous laboratory-based attention research. Specifically, to investigate naturalistic infant sustained attention, we used two ecologically valid, home-based measures of sustained attention. First, we assessed task-specific sustained attention during a caregiver-infant book reading task in their home (see examples of similar task-specific methods, Frick et al., 2018; Marcovitch, Clearfield, Swingler, Calkins, & Bell, 2016; Suarez-Rivera et al., 2019; S. V Wass et al., 2018; Yu, Suanda, & Smith, 2019). Second, we assessed global infant sustained attention behavior to gauge how the infant typically practices attention behaviors in their natural environment towards objects, activities, and people throughout the home visit (see examples of similar global observation methods, Johansson et al., 2016; Towe-Goodman et al., 2011).

To investigate our main questions of interest, we evaluated associations between early-life poverty-related risk and three separate measures of self-regulation in pre-kindergarten (preK), executive function, effortful control, and emotion regulation. Consistent with the prior literature and the theoretical model of self-regulation described above, we hypothesized that: (1) poverty-related risk would be negatively associated with sustained attention in infancy; (2) sustained attention in infancy would be positively associated with all three measures of self-regulation; and (3) sustained attention in infancy would mediate the relationship between poverty-related risk and measures of self-regulation at school entry.

Method

Participants

The Family Life Project (FLP) is a prospective longitudinal study of families residing in six low-wealth counties in Eastern North Carolina and Central Pennsylvania (three counties per state) that were selected to be indicative of the Black South and Appalachia, respectively. The FLP adopted a developmental epidemiological design whereby complex sampling procedures were used to recruit a representative sample of 1,292 children whose families resided in one of the six counties at the time of the child’s birth. Detailed descriptions of the participating families and communities are available in Vernon-Feagans, Cox, and the FLP Investigators (2013).

Procedures

As part of the larger protocol, home visits were conducted when children were 7, 15, and 60-months-old; at 60 months, children were also seen in preK. During each home visit, primary caregivers provided information on demographics and numerous aspects of family life and relationships. In addition, at 7 and 15 months, children and their primary caregiver participated in a book reading task from which observational indicators of infant attention were derived. Immediately following the home visits, research assistants (RAs) completed ratings of the child’s attention during the 2–3 hours of the data collection period. At 60 months, children were administered a battery of executive function tasks in the home and preK teachers completed ratings of children’s effortful control and emotion regulation.

Measures

Cumulative Risk.

Cumulative poverty-related risk was measured by creating an aggregate variable composed of multiple measures of poverty-related risk to encompass the many contributing environmental factors that are associated with living in poverty. Given the high likelihood of co-occurrence between risk factors associated with poverty and the difficulty in parsing them from each other, a cumulative risk model can better encompass the multidimensional nature of poverty-related risk. Thus, as with prior FLP data (see Vernon-Feagans et al., 2013), we created a cumulative risk index of seven measures collected at 7 months, including family income-to-needs ratio, maternal education, consistent partner, hours of employment, occupational prestige, household density, and neighborhood noise and safety (See Table 1 for descriptive statistics). These variables were chosen as indicators of social and economic resources that previous research has demonstrated are significantly related to the context of poverty, especially in rural communities (Burchinal, Roberts, Hooper, & Zeisel 2000; Evans, 2004; Dill, 1999; Vernon-Feagans et al., 2013). Principal components analysis confirmed that these seven indicators each loaded significantly onto a single factor (Vernon-Feagans et al., 2013). The cumulative risk index was calculated by z-scoring each variable, reverse-scoring positively framed variables, and averaging the seven factors. For the present analysis, we used the cumulative risk scores to obtain a variable of early-life exposure to poverty-related cumulative risk.

Table 1.

Descriptive Statistics for Cumulative Risk Variables

	N	Mean/%	St. Dev.	Min	Max
Income-to-Needs Ratio	1,102	1.92	1.7	0	16.49
Maternal Education	1,204	14.44	2.82	6	22
Consistent Partner	1,292	57.1%
Employment Hours	1,204	31.64	21.78	0	104
Occupational Prestige	1,096	39.94	12.04	16.78	86.05
Household Density	1,092	0.88	0.36	1	4
Neighborhood Safety	1,195	2.99	0.58	0.36	3.33

Infant Sustained Attention.

Task-specific infant sustained attention was assessed at 7 and 15 months using a wordless picture-book task: the Early Attention to Reading Situations (EARS; Feagans, Kipp, & Blood, 1994). At each time point caregivers were given a wordless picture book to review before the session began. The caregiver was asked to go through the book and talk to the child about the book as she might normally do. Caregivers and their children shared the books Baby Faces (2002) at 6 months and No David! (Shannon, 1998) or David Gets in Trouble (Mayer, 1973) at 15 months. After approximately 10 min, the RA asked the mother to stop if the picture book task had not ended. The RA live-coded for sustained attention using Observer Software (Noldus, Inc.) on a laptop computer that was programmed to receive observational ratings every 5 seconds. The RA rated the child’s focus of attention at 5 second interval using one of four mutually exclusive categories: (1) a look at the book when the infant’s eyes are focused on the cover or pages of the book; (2) a look at the caregiver when the infant’s eyes are focused on the caregiver’s face who is going through the book; (3) a gaze aversion when the child’s eyes are distracted from the task; (4) a look at the coder when the infant is looking at the coder or camera; (5) a leave if the infant leaves the task momentarily or drops the book. RAs practiced the rating system with pilot children until acceptable reliability was reached with the master rater (Cohen’s Kappas of at least .70). Five selected video recordings of the book sharing activity were then rated by the master rater every 6 months, to confirm and maintain a Cohen’s Kappa of at least .70 for each rater. Because the length of the book sharing activity varied between participants, raw frequency scores were converted to proportion scores to be used in analyses. We created a mean proportion score at each time point that represented the proportion of 5 second intervals in which a look at book was observed.

Global sustained attention was assessed at 7 and 15 months of age with the Infant Behavior Record (IBR; Bayley, 1969) as adapted for use by Stifter and Corey (2001) and completed independently by both RAs. The IBR was applied to infant behavior observed globally across the entire (2–3 hr) home visit. The IBR consists of 11 items rated on a 9-point scale, with higher scores indicating greater sustained attention skills. As was conducted in prior analyses with these data (Towe-Goodman, Stifter, Coccia, Cox, & Family Life Project Key Investigators, 2011), three items were used in the current analysis to index child’s global sustained attention: attention to objects, which assessed the degree to which the child demonstrated sustained interest in toys, test materials, or other objects (a score of 1 indicated the child did not look at or in any way indicate interest in objects, whereas a score of 9 indicated sustained interest in objects, to the point at which they were reluctantly relinquished); attention to activities, which assessed the child’s persistence in attending to 13 activities with toys, objects or persons (a score of 1 indicated the child showed a fleeting attention span, whereas a score of 9 indicated long-continued absorption); and overall attention, which assessed the child’s attention across the demands of the home visit (a score of 1 indicated the child tired easily and quickly regresses to lower levels of functioning, whereas a score of 9 indicated that the child continued to respond well and with interest, even during prolonged tasks at difficult levels). These items were chosen as the most relevant to probing behaviors related to sustained attention, in contrast to other IBR items that pertain more to behaviors related to arousal, mood, and reactivity. The mean of both home visitors’ ratings were used for each item; intraclass correlations ranged from 0.66 to 0.80. The final global attention score was calculated by summing the means across RAs for the three attention-related items.

Executive Functions.

At 60 months of age children were administered an executive function battery consisting of two working memory tasks, three inhibitory control tasks, and one attention shifting task. Preceding the test trials in each task, RAs administered training trials and children completed up to three practice trials. RAs discontinued the task for those children who did not demonstrate an understanding of the task. Each task was presented by a RA in an open spiral-bound flipbook with pages that measured 8 in. × 14 in. Details on the tasks and administration procedures as well as psychometric characteristics are available in Willoughby et al. (2011) and Willoughby et al. (2010). We provide a brief description of the tasks here. Working Memory Span (working memory). In the span task children are shown the outline of a house with an animal and a colored dot inside it, and are prompted to name the animal and the color of the dot. Then they are shown a blank house and asked to either report the animal or the color they had seen in the previous house. In order to perform correctly, children must hold two pieces of information in mind (i.e., the animal and the color), but only recall the prompted feature (e.g., animal). Pick the Picture Game (working memory). In this self-ordered pointing task, children are presented with sets of items. For each set, the same items appear on two sequential pages in a different arrangement. On the first page children are asked to “pick one” item. On each subsequent page, children are instructed to pick an item that was not previously picked so that each picture “gets a turn.” Difficulty increases as more items, up to six, are added to sets. Silly Sounds Stroop (inhibitory control). Children are presented with pictures of cats and dogs and asked to make the sound opposite of that which is typically associated with that animal (e.g., when showed a dog, a correct response would be to make a cat sound). Spatial Conflict Arrows (inhibitory control). In this Simon-like task children are presented response cards with two black circles (“buttons”) on either side of the page and an arrow on either the left or right side of the page. Children are instructed to touch the button corresponding to the side to which the arrow is pointing. The task proceeds in difficulty from displaying left-pointing arrows on the left and right-pointing arrows on the right side of the page (congruent trials), to most arrows pointing to the side opposite from which they are positioned (incongruent trials). Animal Go/No-Go (inhibitory control). In this standard go/no-go task children are instructed to click a button (which made an audible sound) every time they see an animal (go trials) unless the animal is a pig (no-go trials). Varying numbers of go trials appear prior to each no-go trial, including, in standard order, 1-go, 3-go, 3-go, 5-go, 1-go, 1-go, and 3-go trials. Something’s the Same Game (attention shifting). In this task, children are presented with a pair of pictures for matching on a single dimension (e.g., the same color). Subsequently, a third picture was presented, and children were asked to identify which of the first two pictures was similar to the new picture. This task requires the child to shift attention from the initial dimension to a new dimension of similarity (e.g., from color to size). Executive Function Task Scoring and Composite Formation. Children needed to complete at least 75% of trials for each task in order for their performance to be analyzed. Tasks were scored using item response theory as this is a more precise way to estimate children’s executive function abilities than percent correct scores. Expected a posteriori (EAP) scores were derived for each task and averaged to obtain a composite score (Willoughby, Wirth, & Blair, 2011). Z-scores were calculated to reflect accuracy on each of the six executive function assessments. The total score reflected the mean of all completed z-scored individual scores. We used a formative composite, as it has been found to more appropriately represent the overarching construct of executive function than a latent factor, which is limited to measurement of the shared variance between tasks that are only weakly to moderately correlated (Willoughby et al., 2017). Prior studies using this battery with the same population have demonstrated acceptable psychometric properties with the composite executive function score (Willoughby, Blair, Wirth, & Greenberg, 2012). As is typical of executive function measures (Willoughby, Holochwost, Blanton, & Blair, 2014) the reliability coefficient for the composite was relatively low, α = 0.50.

Effortful Control.

The child’s preK teacher reported on child effortful control using the Children’s Behavior Questionnaire (CBQ) at 60 months of age (Rothbart, Ahadi, Hershey, & Fisher, 2001). The two CBQ dimensions that most directly relate to effortful control were assessed: Attentional Focusing and Inhibitory Control. Attentional Focusing consisted of seven items, for example, “When picking up toys or other jobs, usually keeps at the task until it’s done.” Inhibitory Control was comprised of five items, for example, “Can lower his or her voice when asked to do so.” Teachers rated each item, using a 7-point Likert scale ranging from extremely untrue of the child to extremely true of the child. Reliability estimates in this sample for Attentional Focusing were α = .77 and α = .78 for Inhibitory Control. These are very similar to those reported by Rothbart, Ahadi, Hershey, & Fisher (2001). The scales were highly correlated within reporter, r = .77, p < .001 for teachers and were averaged into a single indicator of effortful control.

Emotion Regulation.

Children’s ability to regulate emotions was indexed through teacher reports of emotion regulation skills on a subscale of the Social Competence Scale (SCS; Conduct Problems Prevention Research Group, 1995) when the child was in PreK. The SCS is a 14-item measure comprised of three subscales: emotion regulation skills, prosocial skills, and aggressive-oppositional behaviors. Each item was scored on a scale of 1 (almost never) to 6 (almost always). Only the emotion regulation subscale was used in the present analysis. Items representing the emotional regulation subscale (5 items; e.g., “copes well with disappointment or frustration”) were taken from the SCS developed for the Fast Track Project (CPPRG, 1995) and had an internal reliability estimate of α = 0.84.

Demographic covariates.

State of residence (PA = 0; NC = 1), sex (0 = Male; 1 = Female) and race (0 = White; 1 = African American) of the child were included as covariates to control for site and demographic differences in study variables. Covariates were included in all subsequent models.

Data Analysis

The total sample size recruited at study entry was 1,292 with 1,204 children seen at age 7 months, 1,169 at 15 months, and 1,099 at 60 months. To avoid bias in estimates associated with missing data, we used full information maximum likelihood (FIML) for all analyses. We were specifically interested in estimating: (1) the direct paths from cumulative risk to infant sustained attention, executive functions, effortful control and emotion regulation; (2) the direct paths from infant attention to executive functions, effortful control and emotion regulation; and (3) the indirect paths from cumulative risk to executive functions, effortful control, and emotion regulation via infant sustained attention. To address our primary research question, we used structural equation modeling. The construct of infant sustained attention was assessed using confirmatory factor analysis. We first assessed model fit for the infant attention measurement model. We then defined our hypothesized structural model and used bootstrapping with 5000 samples to generate bias-corrected confidence intervals for indirect effects (Shrout & Bolger, 2002). The predictor variable was cumulative risk and the outcome variables were the EAP executive functions factor score, CBQ mean effortful control score, and the SCS emotion regulation score. We controlled for the effects of sex, state, and race on all paths in our model. All analyses were conducted in the R environment (R Core Team, 2013). We tested overall model fit using root mean square error of approximation (RMSEA), standardized root mean square residual (SRMR), and comparative fit index (CFI) (Hu & Bentler, 1999). All parameter estimates are standardized estimates and thus, indicate how much the dependent variables would be expected to change for a single standard deviation change in the predictor variable.

Results

Preliminary Analyses

Tables 2 and 3 present descriptive statistics and correlations for our analysis variables, respectively. Table 3 indicates a significant positive correlation between each attention variable, suggesting reasonable reliability within each measure over time and validity between measures. Moreover, task-specific sustained attention at 15 months was positively correlated with effortful control in preK. Global sustained attention at 7 months was positively correlated with executive function in preK. Global sustained attention at 15 months had a positive correlation with executive function, effortful control, and emotion regulation in preK. Further, cumulative risk was negatively correlated with global sustained attention at 7 and 15 months, and moderately negatively correlated with executive functions, effortful control, and emotion regulation in preK. Notably, there was a significant positive correlation between each outcome measure of self-regulation in preK. These preliminary findings provided support for our hypothesized structural model.

Table 2.

Descriptive Statistics for Variables in the Analysis

	N	Mean/%	St. Dev.	Min	Median	Max
Task-Based Sustained Attention 7mos (EARS)	1,172	0.77	0.19	0.00	0.82	1.00
Task-Based Sustained Attention 15mos (EARS)	1,134	0.70	0.22	0.00	0.75	1.00
Global Sustained Attention 7mos (IBR)	1,196	17.68	2.47	4.50	18.00	24.00
Global Sustained Attention 15mos (IBR)	1,155	17.71	2.70	4.00	18.00	23.50
Executive Functions preK	1,038	0.29	0.48	−1.98	0.35	1.40
Effortful Control Mean preK (CBQ)	793	4.92	1.17	1.00	5.08	7.00
Emotion Regulation preK (SCS)	816	3.91	1.03	1.20	4.00	6.00
Cumulative Risk 7mo	1204	0.01	0.66	−2.93	0.01	1.98
Child’s Race (% African American)	1292	42%
Child’s Sex (% Female)	1292	51%
State of Residence (% NC)	1292	60%

PreK, Pre-Kindergarten; mo, months; IBR, Infant Behavior Record; EARS, Early Attention to Reading Situations; CBQ, Children’s Behavior Questionnaire; SCS, Social Competence Scale.

Table 3.

Correlations among Variables

	1.	2.	3.	4.	5.	6.	7.	8.
1. Task-Based Sustained Attention 7mos (EARS)	1
2. Task-Based Sustained Attention 15mos (EARS)	0.10**	1
3. Global Sustained Attention 7mos (IBR)	0.11**	0.08*	1
4. Global Sustained Attention 15mos (IBR)	0.04*	0.22**	0.21**	1
5. Executive Function preK	0.01	0.03	0.09*	0.10*	1
6. Effortful Control Mean preK (CBQ)	0.01	0.07*	0.01	0.12*	0.45**	1
7. Emotion Regulation preK (SCS)	0.06	0.04	−0.01	0.06**	0.25**	0.59**	1
8. Cumulative Risk	−0.03	0.03	−0.11**	−0.07*	−0.32**	−0.28**	−0.20**	1

^**p≤.01,

^*p≤.05;

PreK, Pre-Kindergarten; mos, months; IBR, Infant Behavior Record; EARS, Early Attention to Reading Situations; CBQ, Children’s Behavior Questionnaire; SCS, Social Competence Scale.

Cumulative Risk and Self-Regulation in PreK

To assess the direct effects of poverty-related risk on the three domains of self-regulation (without infant sustained attention in the model), we constructed a regression of executive functions, effortful control, and emotion regulation at preK on cumulative risk at 7 months, controlling for all demographic covariates. Results from our model indicate that poverty-related risk was significantly negatively associated with executive functions β=−.27, p<.001, effortful control β=−.23, p<.001, and emotion regulation β=−.15, p<.001.

Measurement Model of Attention in Infancy

We evaluated a latent variable of infant sustained attention at 7 and 15 months using confirmatory factors analysis. The model fit was adequate, χ²(2)=16.24, p=.008, CFI=.91, RMSEA=.076, SRMR=.028. Inspection of parameter estimates indicated that all of the factor loadings were statistically significant and in the expected direction (task-specific sustained attention at 7 mos: β=.16, p=.001; task-specific sustained attention at 15 mos: β=.35, p<.001; global sustained attention at 7 mos: β=.34, p<.001; global sustained attention behavior at 15 mos: β=.68, p<.001).

Structural Mediation Model

In the hypothesized structural model, we examined the direct effects of cumulative risk on each self-regulation outcome as well as the indirect effects of risk on executive functions, effortful control and emotion regulation through infant sustained attention while controlling for demographic covariates. The estimated structural model is shown in Figure 1. This model fit the data adequately, χ²(25)=101.84, p<.0001, CFI=.92, RMSEA=.049, SRMR=.034. First, significant negative direct effects were observed for cumulative risk on executive functions, β=−.23, p<.001, effortful control, β=−.10, p<.001, and emotion regulation, β=−.11, p<.001. Compared to the preliminary model without infant sustained attention in the model, the three coefficients for self-regulation were somewhat reduced after infant sustained attention was included in the model, suggesting potential partial mediation.

Figure 1.

The observed mediation model relating cumulative risk to attention and self-regulation outcomes. All coefficients are standardized (β). **p<.01, *p<.05; PreK, Pre-Kindergarten; mo, months. Covariates are not shown.

Second, the model demonstrated a direct effect of cumulative risk on the infant sustained attention latent variable such that cumulative risk was negatively related to infant sustained attention, β=−.22, p<.001. Third, there was a direct effect between infant sustained attention and self-regulation skills such that sustained attention was positively associated with preK executive functions, β=.16, p=.001, effortful control, β=.22, p<.001, and emotion regulation, β=.13, p=.009. Results from tests of indirect effects of cumulative risk on self-regulation through infant sustained attention (in Table 4) indicated that sustained attention partially mediated the relationship between cumulative risk and executive functions, β=−.033, p=.019, 95% CI [−.049,−.008] effortful control, β=−.047, p=.010, 95% CI [−.156,−.031] and emotion regulation β=−.028, p=.029, 95% CI [−.090,−.012].

Table 4.

Standardized Indirect Effects

	β	S.E.	95% Confidence Interval
Cumulative Risk → Infant Attention → Executive function	−0.033	0.011	(−0.049, −0.008)
Cumulative Risk → Infant Attention → Effortful Control	−0.047	0.033	(−0.156, −0.031)
Cumulative Risk → Infant Attention → Emotion Regulation	−0.028	0.020	(−0.090,−0.012)

Discussion

In the current study, we assessed relations between early life poverty-related risk, sustained attention in infancy, and multiple domains of self-regulation in preK. Specifically, we investigated the extent to which infant sustained attention mediated the association between cumulative risk and executive functions, effortful control, and emotion regulation. Based on previous theoretical and empirical evidence, we hypothesized that early poverty-related risk would be negatively associated with reduced sustained attention, which would in turn positively predict higher-order self-regulation difficulties in prekindergarten. This research was motivated by an understanding of infant attention as a foundational aspect of self-regulation that is affected by the context in which the child is developing. While previous research has reported poverty to be negatively associated with early attention (Clearfield & Jedd, 2006; Conejero, Guerra, Abundis-Gutiérrez, & Rueda, 2016; Lipina, Martelli, Vuelta, & Colombo, 2005), and early attention to be positively associated with childhood cognitive (Cuevas & Bell, 2014; Johansson et al., 2015; Rose, Feldman, & Jankowski, 2012) and emotion regulation development (Johansson et al., 2015; Pérez-Edgar et al., 2010; Perry et al., 2016), this is the first study, to the best of our knowledge that has tested a model of infant sustained attention as a mediator of the effects of socioeconomic adversity on multiple domains of self-regulation at school entry. Importantly, one of the strengths of our analysis is that we used two observational measures of attention that were collected during naturalistic, semi-structured paradigms in the home, over a period of 2–3 hours. This method of measurement assesses more global sustained attention behavior, which is in contrast to other relevant studies that measured specific attention behaviors during unfamiliar, structured tasks in the laboratory. Nonetheless, our results are in line with other longitudinal research done in more controlled settings (Cuevas & Bell, 2014; Papageorgiou et al., 2014; Pérez-Edgar et al., 2010; Rose, Feldman, & Jankowski, 2012; Sigman, Cohen, & Beckwith, 1997).

Poverty and Attention in Infancy

Our findings contribute to the extant literature in several ways. First, we found that poverty-related adversity is negatively associated with sustained attention processes in the first year-and-a-half of life. This finding adds to a growing literature demonstrating that the effects of adversity on attention are evident in infancy (Clearfield & Jedd, 2006; Conejero, Guerra, Abundis-Gutiérrez, & Rueda, 2018; Hanson et al., 2013; Lipina et al., 2005; Tomalski et al., 2013). Here, we extend previous lab-based studies by measuring infant sustained attention in the home environment and by including both a task-specific measure in addition to a global measure to create an ecologically valid, inclusive construct of infant sustained attention. Furthermore, much of the prior literature takes a comparative and cross-sectional approach, testing distinct groups of children from low- versus middle-income backgrounds at single time points. In contrast, we analyzed a continuous measure of early adversity and demonstrated an association between variation in adversity and variation in sustained attention over the child’s first year-and–a-half of life.

More broadly, our results are also consistent with a growing body of evidence indicating that early poverty-related risk affects the developing brain in areas known to be associated with top-down emotion regulation and cognitive control. (Hair, Hanson, Wolfe, & Pollak, 2015; Luby, Barch, Whalen, Tillman, & Belden, 2017; Noble et al., 2015). In particular, our findings are in line with structural imaging work demonstrating the effect of poverty on brain development from a very early age in life (Hanson et al., 2013). Specifically, Hanson et al. (2013) found large SES-related reductions in grey matter in the frontal and parietal regions associated with cognitive control abilities (Hanson et al., 2013). Similarly, a study using EEG with 6–9-month-old infants found reduced high frequency (gamma) oscillations in the PFC among children in poverty relative to children from higher income homes (Tomalski et al., 2013). Corroborating neural evidence in a second study indicates that toddlers in low-SES homes demonstrated decreased theta power (Conejero et al., 2018). Importantly, both frontal gamma and theta power are spectral frequencies considered to support attentional control processes (Engel, Fries, & Singer, 2001; Tsujimoto, Shimazu, & Isomura, 2006).

These neuroimaging findings also highlight the protracted development and malleability of PFC in infancy and childhood (Casey, Giedd, & Thomas, 2000; Holmes & Wellman, 2009; Sheridan, Sarsour, Jutte, D ‘esposito, & Boyce, 2012). The first years of brain development are characterized by rapid growth and increased plasticity in the PFC (Grossmann, 2013; Hodel, 2018), which are characterized by heighted susceptibility to environmental influences (Hensch, 2005). It follows that PFC-dependent processes like attention may be differentially affected by environments of adversity within the first years of life. While studies explicitly assessing the importance of developmental timing are lacking, our findings indicate that behavioral differences associated with poverty are observable within the first year-and-a-half of life. These findings have implications for the need for continued efforts to identify windows of plasticity and malleability around which to gauge the relative impacts of efforts to foster the wellbeing of children and families in poverty.

Attention in Infancy and Self-Regulation at School Entry

Our findings also contribute to the growing literature demonstrating that attention in infancy is associated with multiple aspects of self-regulation in early childhood. Specifically, similar to other longitudinal studies, we found a positive relation between infant sustained attention and executive function at school entry (Cuevas & Bell, 2014; Frick et al., 2018; Johansson et al., 2016, 2015; Rose, Feldman, & Jankowski, 2012). These findings are consistent with a theoretical model in which early attention skills are essential in supporting executive functions (Bell & Deater-Deckard, 2007; Colombo & Cheatham, 2006; Rueda, Posner, & Rothbart, 2005). Moreover, our finding that sustained attention in infancy was positively associated with effortful control in preK is largely consistent with prior theoretical and empirical work using Rothbart and Posner’s conceptualization of effortful control (Derryberry & Reed, 1994; Posner & Rothbart, 2000; Rothbart & Rueda, 2005). In particular, attention is necessary for children’s volitional control over prepotent behaviors, such as inhibiting an inappropriate response, particularly in school settings. For instance, when responding to a question posed by the teacher, a child recruits effortful, attentional control to raise her hand instead of shouting out an answer. Such an interpretation aligns with prior studies demonstrating associations between observational ratings of attention in infancy and effortful control at two years of age (Johansson et al., 2015; Kochanska et al., 2000; Papageorgiou et al., 2014). The reported associations between infant sustained attention and executive function and effortful control suggest that basic cognitive abilities developed in infancy precede the emergence of more advanced cognitive and behavioral abilities.

Our finding that infant sustained attention predicts emotion regulation ability is also consistent with the theoretical and biological models of attention and self-regulation as well as with prior longitudinal research demonstrating relations between attention in infancy and emotion regulation in childhood (Pérez-Edgar et al., 2010; Posner & Rothbart, 2000). In a prior study, Pérez-Edgar and colleagues (2010) theorized that sustained attention in the first year of life forms the core of a regulation mechanism to temper emotional responses to environmental stimuli. This theory is supported by research suggesting that frontal cortical activation during an attention task in infancy is indirectly associated with emotion regulation at 3 years of age (Perry et al., 2016). Alternatively, control of attention may allow children to learn how to selectively attend to salient aspects of their environment, enhancing one’s ability to discriminate between threatening and nonthreatening environmental stimuli (Sroufe, 1996). Thus, we theorize that the development of attention control in infancy is likely contributing to the development of emotion regulation.

Attention Mediates Effects of Early Risk on Self-Regulation

Our mediational findings support our central hypothesis that the ability to sustain attention in infancy is a foundational early mechanism linking early life experience with later self-regulation. Although small in size, the indirect effects demonstrate that infant attention is one pathway through which environments of risk exert effects on the development of self-regulation skills in early childhood. This finding is well situated within a rich body of literature linking early life adversity with disparities in the development of brain and behavior as well as research connecting early cognitive control with later self-regulation outcomes (Cuevas & Bell, 2014; Papageorgiou et al., 2014; Raver et al., 2013). We postulate that because infant PFC neurodevelopment is especially susceptible to early life experiences, exposure to poverty-related risk in infancy disrupts attention from an early age. These early alterations in attention may have enduring effects on the development of self-regulation abilities when considering the empirical research supporting a cascade model of infant PFC development (Amso & Scerif, 2015; Colombo & Cheatham, 2006; Susan A. Rose et al., 2012). Broadly, we speculate that infant sustained attention is an important antecedent to more complex self-regulation processes, which may be impacted by early life adversity through attention.

Although the reported pathway is correlational, our model is longitudinal, thus establishing temporal precedence that supports a potentially causal relation. Indeed, evidence from intervention efforts suggests a causal role for infant sustained attention in later self-regulation. For instance, recent training efforts have found that relative to an active control group, infants in an attention training protocol demonstrated enhanced development of distal cognitive control processes in toddlerhood (Wass, Porayska-Pomsta, & Johnson, 2011; Wass, Scerif, & Johnson, 2012). While this finding is limited to toddlerhood and basic cognitive control abilities, these preliminary studies offer promising evidence for infant attention as a viable target mechanism for promoting higher-order cognitive abilities, particularly when infants are in high-risk environments. Collectively, this research has implications for prevention research aimed at cultivating a supportive home environment to promote the development of early cognitive control.

Limitations and Future Directions

There are several limitations that preclude our ability to come to strong inferential conclusions based on our results. First, there are caveats associated with the use of observational ratings to measure infant sustained attention. Observer ratings of sustained attention are subjective and could reflect aspects of child behavior that are associated with attentiveness, such as interest, motivation, sociability, or reactivity. Similarly, aspects of the home environment, such as number of distractors, could be confounding our association between cumulative risk and sustained attention. This potential problem is minimized somewhat by the fact that we combined the ratings of two highly trained RAs that independently rated infant behavior over a 2–3 hour home visit across two child ages. This consideration is an important one, however, given that our infant sustained attention latent variable was primarily driven by global sustained attention behavior at 15 months. Sustained attention at 15 months becomes more sophisticated and intentional as the executive attention network functionally matures around this time (Colombo & Cheatham, 2006; Rueda, Posner, & Rothbart, 2005). As such, attention manifests more prominently and is likely easier to observe as a naturally occurring behavior. While this paper focuses on the cognitive antecedents of self-regulation, future analyses would benefit from considering the role of related behaviors such as reactivity or motivation during infancy as well. Similarly, controlling for chaos in the home and parent behaviors during the task would strengthen future studies aimed at assessing infant sustained attention in naturalistic environments. Further, measurement of sustained attention during the book reading task would be improved by the inclusion of a looking duration code (in addition to frequency) and a physiological measure of heart rate to validate infant’s attention state. In addition, measures of effortful control and emotion regulation relied solely on teacher report, which may limit the generalizability of these findings beyond classroom contexts. The analysis would benefit from the inclusion of direct assessments to measure these constructs. Importantly, the longitudinal model presented here is correlational. While the consideration and inclusion of covariates in our model improves our ability to draw inferences, causal conclusions about the relations among variables are not possible. Therefore, future intervention and cross-species research is needed to develop causal conclusions about relations between attention, context and the development of higher-order cognitive and emotion abilities.

The current study highlights infant sustained attention as a foundation for the development of more complex self-regulatory processes in contexts of poverty. This work provides a mechanistic model by which early environments of cumulative risk may be negatively affecting the development of basic attention processes in infancy, initiating later disparities in higher-order self-regulation abilities. Given the plastic nature of attention networks early in life, amelioration efforts could productively focus on cultivating supportive and nurturing home environments for parents and their infants to support the development of foundational cognitive processes.

Highlights.

• Early poverty-related risk is negatively associated with sustained attention in infancy

• Infant sustained attention is positively related to multiple domains of self-regulation in pre-kindergarten

• Poverty-related risk during infancy predicted lower self-regulation, partially, through infant sustained attention