Maternal Depression Affects Infants’ Lexical Processing Abilities in the Second Year of Life

brain sciences Article Maternal Depression Affects Infants’ Lexical Processing Abilities in the Second Year of Life Ruth Brookman 1,2,* , Marina Kalashnikova 1,3,4 , Janet Conti 2, Nan Xu Rattanasone 5, Kerry-Ann Grant 6, Katherine Demuth 5and Denis Burnham 1 1 The MARCS Institute for Brain, Behaviour and Development, Western Sydney University, Locked Bag 1957, Penrith, NSW 2750, Australia; [email protected] (M.K.); [email protected] (D.B.) 2School of Psychology, Western Sydney University, Locked Bag 1957, Penrith, NSW 2750, Australia; [email protected] 3Basque Centre for Cognition, Brain and Language, Paseo Mikeletegi 69, 2º, 20009 Donostia-San Sebastián, Spain 4IKERBASQUE, Basque Foundation for Science, Plaza Euskadi 5, 48009 Bilbao, Spain 5Department of Linguistics, Macquarie University, Balaclava Road, North Ryde, NSW 2109, Australia; [email protected] (N.X.R.); [email protected] (K.D.) 6Health Education and Training Institute, Locked Bag 7118, Parramatta BC, NSW 2124, Australia; Kerry[email protected] *Correspondence: [email protected] Received: 8 November 2020; Accepted: 9 December 2020; Published: 12 December 2020   Abstract: Maternal depression and anxiety have been proposed to increase the risk of adverse outcomes of language development in the early years of life. This study investigated the effects of maternal depression and anxiety on language development using two approaches: (i) a categorical approach that compared lexical abilities in two groups of children, a risk group (mothers with clinical-level symptomatology) and a control non-risk group, and (ii) a continuous approach that assessed the relation between individual mothers’ clinical and subclinical symptomatology and their infants’ lexical abilities. Infants’ lexical abilities were assessed at 18 months of age using an objective lexical processing measure and a parental report of expressive vocabulary. Infants in the risk group exhibited lower lexical processing abilities compared to controls, and maternal depression scores were negatively correlated to infants’ lexical processing and vocabulary measures. Furthermore, maternal depression (not anxiety) explained the variance in infants’ individual lexical processing performance above the variance explained by their individual expressive vocabulary size. These results suggest that significant differences are emerging in 18-month-old infants’ lexical processing abilities, and this appears to be related, in part, to their mothers’ depression and anxiety symptomatology during the postnatal period. Keywords: language development; lexical processing; vocabulary; postnatal; maternal depression; maternal anxiety; infancy 1. Introduction An infant’s developmental trajectory is shaped by both environmental and genetic factors [ 1 ]. Communication with a primary caregiver, usually the mother, is an infant’s first social and linguistic experience and exerts a strong influence on the infant’s language development [ 2 , 3 ]. However, infancy is also associated with an increased risk of maternal depression and anxiety, the two most common emotional health conditions experienced by mothers after giving birth, e.g., [ 4 , 5 ]. Postnatal depression and anxiety can have long-lasting effects on children’s socio-emotional development, e.g., [ 6 , 7 ]. Recent Brain Sci. 2020,10, 977; doi:10.3390/brainsci10120977 www.mdpi.com/journal/brainsci Brain Sci. 2020,10, 977 2 of 18 research suggests that children’s cognitive development can be impacted in the first years of life. Specifically, in the domain of language acquisition, recent studies have shown negative effects of maternal depression and anxiety on infants’ expressive vocabulary skills at 12 and 18 months of age [ 8 , 9 ]. Importantly, findings suggest that effects may be dependent on the severity of maternal depression and anxiety symptoms and are not restricted to infants of mothers with clinical levels of these disorders [ 10 ]. This study investigates the relation between maternal depression and anxiety at both clinical and subclinical levels and infants’ lexical development by including an assessment of infants’ expressive vocabulary size as well as an objective measure of receptive vocabulary and efficiency of lexical processing, which have not been investigated in this population. 1.1. Maternal Depression and Anxiety and Links with Language Development One-quarter to one-third of women in Western societies experience one major depressive episode in their lifetime [ 11 ]. In clinical settings, health professionals frequently use the term postnatal depression (PND) to refer to depression symptoms experienced in the postnatal period that are similar to those of major depressive disorder (MDD). According to the Diagnostic and Statistical Manual of Mental Disorders, Fifth edition (DSM-5: [ 12 ]), PND is referred to as a specifier of MDD, with symptoms including a loss of appetite, sleep disturbance, weight changes, loss of interest in pleasurable activities, agitation, concentration difficulties, feelings of guilt and worthlessness, and suicidal thoughts. In addition to depression, there is a high incidence of anxiety disorders in this population, with prevalence estimates ranging from 3% to 43% in Western societies [ 13 ]. The DSM-5 notes that anxiety disorders can include, for example, generalized anxiety, specific phobias, and panic disorders [ 12 ]. In a similar manner to postnatal depression, postnatal anxiety symptoms can lead to a significant impairment in daily functioning [ 14 ]. Importantly, these two conditions often co-occur in the postnatal period [ 15 ], and it is not always possible to isolate their independent effects on maternal mental health and possible subsequent effects on infants’ cognitive development. For this reason, this study considers the two conditions together. There is growing evidence that maternal depression can yield early and long-lasting effects on children’s linguistic development. The National Institute of Child Health and Human Development (NICHD) Early Child Care Research Network conducted a longitudinal study that followed mother–infant dyads from six to 15, 24, and 36 months, and assessed children’s language and general cognitive abilities at 36 months [ 16 ]. After controlling for demographic risk factors, children of mothers with chronically or occasionally elevated depression scores performed less optimally compared to children of non-depressed mothers on measures of verbal comprehension and expressive language in addition to an assessment of school readiness. More recently, Kaplan et al. [ 10 ] found a significant negative correlation between self-reported maternal depression scores and 12-month-old infants’ percentile scores on the Bayley Expressive Communication subscale but not the general cognitive and receptive communication subscales of the Bayley-III [ 17 ]. A similar study focusing on maternal anxiety examined the emerging language abilities of 12-month-old infants with mothers who met the criteria for an anxiety disorder in the postnatal period [ 9 ]. This study showed that infants scored significantly lower than controls in the language subtests of the Bayley-III [ 17 ]. Taken together, the above studies suggest that the onset of language delay in the children of mothers affected by depression or anxiety may commence quite early during infancy and be reflected in infants’ early pre-lexical and lexical skills. Overall, compared to maternal depression, there is a scarcity of research examining the link between maternal anxiety and adverse developmental findings [ 13 ]. Inconsistencies in findings from previous studies examining maternal anxiety have been attributed to the lack of homogeneity in defining and measuring anxiety and its high co-morbidity with depression [ 13 ]. Furthermore, postnatal depression symptoms may mask anxiety symptoms, leading to a risk of postnatal anxiety disorders remaining undetected and untreated [ 18 , 19 ]. In this study, we assess both maternal depression and anxiety. Our analyses consider these conditions together for the purpose of classifying mothers into Brain Sci. 2020,10, 977 3 of 18 groups of those who are and are not affected by a depression and/or anxiety diagnosis, and also separately, to measure the relation of each condition to infants’ early language development. 1.2. Language Learning and Social–Environmental Influences The observed delays in language outcomes have been attributed to the effects of maternal mental health concerns on the quality of infants’ early linguistic environment [ 8 ]. This view emphasizes the important role that infants’ early environments and socio-communicative interactions have in supporting their language development. It also posits that language learning is socially mediated by caregiver interactions with their children [ 20 ]. Accordingly, it is said that mothers support their infants’ communication development through the quality and quantity of their speech input [ 21 ], both of which can contribute positively to a rich and stimulating home language environment [ 22 ]. This supporting role, however, may be challenged by the presence of maternal mental health concerns affecting the quality and quantity of day-to-day interactions with the infant and leading to long-lasting effects in the infant’s development. A recent study by Brookman et al [ 8 ] provides direct evidence for the proposal that adverse outcomes in infants’ language development are due to the disruptions to the quantity of the mother–infant interaction related to maternal depression [ 23 ]. This longitudinal study examined the home language environments of a group of infants at risk for language delays by virtue of having mothers affected by emotional health concerns, compared with a group of not-at-risk controls. Analyses of day-long audio recordings conducted when the infants were six and 12 months old revealed that at-risk infants were exposed to significantly fewer conversational turns and vocalized significantly less than controls. Importantly, over and above maternal depression and anxiety measures, individual variations in infants’ early home language environment (conversational turns and vocalization counts) predicted infants’ expressive vocabulary size at 18 months. From these studies, it can be concluded that mothers who speak more to their infants tend to have children who develop more advanced language abilities, e.g., [ 24 , 25 ], and that this is the case for mothers who do and do not exhibit clinical levels of mental health concerns. Moreover, research suggests that the effects of frequency of mother–infant interactions can be observed not only on children’s developing vocabulary size but also on lexical processing efficiency—infants’ ability to access the meaning of familiar words stored in the lexicon. Weisleder and Fernald [ 26 ] assessed the quantity of early language input and lexical processing efficiency in children from low and high socio-economic status (SES) backgrounds. Differences emerged at 18 months between SES groups in both vocabulary development and language processing efficiency. In addition, the effect of the amount of speech input on expressive vocabulary was mediated by language processing efficiency, suggesting that exposure to richer or higher-quality speech input supports infants’ processing abilities and, therefore, their vocabulary growth. Infants’ lexical processing efficiency abilities undergo a significant change in their second year of life, whereby infants become increasingly fast, more accurate, and more efficient at recognizing familiar words in continuous speech and in different contexts [ 27 – 29 ]. Individual differences observed in this process are developmentally meaningful as they predict individual vocabulary growth and the development of more advanced linguistic abilities in early childhood and even school age [ 29 ]. Word recognition skills demonstrated by toddlers at 25 months of age have been found to predict later language skills at eight years of age [ 30 ]. In addition, individual differences in processing speed in pre-term infants tested at 18 months of age were shown to predict receptive vocabulary at 3 years [ 31 ] and global language and cognitive abilities at 4.5 years [ 32 ]. Infants’ lexical processing abilities have been found to have high predictive validity for later literacy skills, school readiness, and academic performance in primary school, e.g., [33] Lexical processing efficiency is typically assessed using Looking-While-Listening (LWL) paradigms and is an important research concept when exploring individual differences in infants’ language development [ 34 ]. Infants’ gaze patterns are recorded while infants hear familiar words and look at Brain Sci. 2020,10, 977 4 of 18 referents on a screen that do or do not correspond to these words (e.g., hearing the phrase “look at the apple” while seeing a picture of an apple and a shoe on a screen). This procedure uses real-time eye-tracking measures of children’s gaze patterns in response to speech and imposes low cognitive demands, which makes it well suited for infants in their second year of life, in particular here, for objectively measuring young infants’ lexical competence. Previous research involving infants of mothers affected by depression has primarily included language measures based on standardized IQ batteries and parental reports [ 8 , 10 ], both of which can be problematic when used with populations at risk for developmental language delays. For example, while parental reports are a well-established method for obtaining valuable information about infants’ early language development [ 35 ], they can be vulnerable to parents’ underor over-estimations of their child’s abilities [ 36 ]. The LWL procedure provides a valuable objective assessment and, thus, complements parental reports of vocabulary size [ 26 ]. The LWL procedure can provide a more detailed picture of young infants’ ability to comprehend or find meaning in spoken language (see [37]). 1.3. The Present Study This study investigated the speech processing efficiency and vocabulary size of infants whose mothers have depression and anxiety symptoms. The study included a subset of mother–infant dyads from the longitudinal cohort from Brookman et al [ 8 ]. Infants’ vocabulary size and lexical processing abilities were assessed when they were 18 months old, a time that is associated with the “vocabulary spurt”, e.g., [ 24 ]. Infants’ vocabulary size was assessed using a parental report and lexical processing was assessed using the LWL procedure. Maternal depression and anxiety were measured at multiple time points in the postnatal period. This enabled the employment of two complementary analytical approaches. First, we employed a categorical approach by allocating mother–infant dyads to a risk or no-risk control group. The aim of this was to assess the differences in early language abilities (lexical processing and vocabulary size) between infants whose mothers were affected by clinical levels of symptomatology of either depression, anxiety, or both versus a no-risk control group. Mothers in the risk group had a current diagnosis of depression and anxiety and/or elevated symptoms (reaching a clinical threshold) reported in the postnatal period. Second, we employed a continuous approach, in which the aim was to assess the relationship between individual differences in the severity of maternal depression and anxiety symptoms on the one hand and infants’ language outcomes on the other. In this case, data from the entire sample were collapsed, placing mothers on a continuum of depression and anxiety measures and assessing the predictive validity of levels of emotional health concerns on infants’ language abilities. Two predictions were tested. First, it was predicted that the categorical approach comparisons would reveal deficits in lexical processing abilities and lower expressive vocabulary sizes in infants in the risk group compared with infants in the control group, e.g., [ 9 ]. Second, it was predicted that the correlational and regression analyses of the continuous approach would demonstrate that infants’ speed and accuracy of lexical processing and their expressive vocabulary size would be correlated with maternal depression and anxiety scores. That is, the effects of maternal depression and anxiety symptoms were predicted to be incremental with subclinical expression and not only present in families where the mother displays clinical levels of depression or anxiety symptomatology, e.g., [8,10]. 2. Materials and Methods 2.1. Participants Forty-six mother–infant dyads (24 female and 22 male infants) participated. Lexical processing data for 10 infants were excluded from the final analysis due to equipment failure (n=2), an acquired hearing loss (n=1), failure to contribute sufficient gaze data for analyses (see Data processing in Section 2.3.1 for details, n=2), and failure to complete the task (n=5). The remaining 36 infants (17 female and 19 male) were in the risk group (n=17) or the control group (n=19) based on the Brain Sci. 2020,10, 977 5 of 18 mothers’ current diagnosis or elevated symptoms of depression and anxiety (see Section 2.2 Maternal Measures below). All infants were born into households with two heterosexual parents, full term (37–42 weeks), with normal birth weight, and no history of birth or postnatal complications. The infants were acquiring English in a monolingual context and were not exposed to a second language, and they had no reported hearing difficulties, neurological conditions, or health problems. At the time of recruitment, mothers’ ages ranged from 26 to 41 years (Mean [M]=33, Standard Deviation [SD]=4.05). Maternal education levels for all mothers ranged from high school to postgraduate degree. A comparison between groups indicated that mothers from the risk group had significantly higher education levels than controls; Mann–Whitney U=103, p=0.040. This difference was due to more at-risk mothers having postgraduate degrees (Risk =7>Control =5) while more control mothers had undergraduate degrees (Control =14 >Risk =12). Nevertheless, the median education level of both groups was an undergraduate university degree. Household income was also assessed by using the estimated average weekly household income levels based on participating families’ place of residence (Australian Bureau of Statistics). There was no significant group difference in estimated household income levels; Mann–Whitney U=106, p=0.051. See Table 1for descriptive and inferential statistics of participants’ demographic information. Table 1. Descriptive and inferential statistics for participants’ demographic information. Mother and Infant Characteristics Total Sample (n=36) Control Group (n=19) Risk Group (n=17) Infant Gender: n (%) Male Female 19 (53) 17 (47) 12 (63) 7 (37) 7 (41) 10 (59) Birth weight (kg) Range M(SD) 2.38–4.89 3.49 (0.52) 2.38–4.74 3.38 (0.53) 3.01–4.89 3.61 (0.51) Birth order: n(%) First-born 26 (72) 13 (68) 13 (77) Maternal Age Range M(SD) 26–41 32.69 (4.05) 26–40 32.74 (3.57) 26–41 32.65 (4.64) Maternal education Mean rank 15.42 21.94 Household income Mean rank 15.58 21.76 M: Mean, SD: Standard Deviation. Initial recruitment for this study was conducted by contacting a community sample of mothers who were previously enrolled in a large-scale longitudinal project examining the effects of prenatal anxiety on infants’ development and cortisol levels at 12 weeks postpartum, and whose infants were within the correct age range for this study. This sample included mothers who did and did not have elevated depression and anxiety scores. Fourteen mother–infant dyads were recruited using this method. The remaining mother–infant dyads were recruited from an infant laboratory database and via the distribution of study flyers through community notice boards, libraries, playgroups, community agencies, and social media. This study was approved by the Western Sydney University Human Research Ethics Committee (approval number: H11703). All mothers provided written informed consent prior to participating in the study. Brain Sci. 2020,10, 977 6 of 18 2.2. Maternal Measures Given the high comorbidity of depression and anxiety, all mothers were administered measures of both conditions. The Center for Epidemiologic Studies Depression Scale—Revised (CESD-R) [ 38 ] was used to measure depression symptoms. The CESD-R is a 20-item self-reported scale of depressive symptoms and has been used widely with maternal populations in the perinatal period [ 39 ]. It has excellent psychometric properties, including strong factor loadings, high internal consistency, and theoretically consistent convergent and divergent validity [ 40 ]. The State Scale of the State-Trait Anxiety Inventory (STAI) [ 41 ] is a reliable and valid measure of current self-reported anxiety that is one of the most commonly used measures to capture clinical and subclinical anxiety levels [ 41 , 42 ]. The STAI has demonstrated good internal consistency in prior Australian studies with child-bearing women, e.g., [18]. Mothers commenced the study when their infants were six months old. To account for the potential variation in the severity and persistence of symptoms of maternal emotional health concerns during the postnatal period, which is known to influence child development outcomes, mean postnatal depression scores and mean postnatal anxiety scores (six, nine, 12, and 18 months) were calculated for each participant by averaging scores obtained at each data collection point (see Supplementary Materials Table S1, for details). Mothers were allocated to the risk group if they had (i) a self-reported current diagnosis of depression or anxiety; or (ii) depression and/or anxiety scores of which either or both exceeded the clinical threshold (i.e., CESD-R ≥ 16; STAI ≥ 40) at any of the four testing time points. The remaining mothers were allocated to the control group. Mother–infant dyads remained in the risk group for the duration of the study despite potential changes to their emotional health scores, due to evidence that impaired mother–infant interactions styles can persist beyond the remission of depression with ongoing risks to infant developmental outcomes, e.g., [43]. 2.3. Infant Measures 2.3.1. Lexical Processing Stimuli. Six target words (ball, book, car, cup, hat, and shoe) were selected on the basis of previous research and familiarity to 18-month-old infants [ 29 , 44 ]. An Australian-English-speaking female was recorded producing the target words embedded in the carrier phrases “Where is the [target]?” and “Look at the [target]” in a lively, child-directed manner. In order to verify that the stimuli were comparable to other studies, an acoustic analysis of the phrases was conducted using Praat, version 6.0.40 [ 45 ] (see Supplementary Materials Table S2, for details). Results indicated that pitch and word duration were in the usual range for infant-directed speech (IDS) and consistent with stimuli used in previous research [46–48]. Visual stimuli for test trials consisted of 6 colorful pictures of familiar target objects that were presented in pairs placed side-by-side (target and distracter). The pictures were presented on a white background and selected to be comparable in size and visual salience. The images appeared in blocked pairs: ball–cup, hat–book, and shoe–car. The audio and visual stimuli were combined into video clips containing 24 test trials using iMovie software version 10.0.9 (Apple Inc.). Each clip’s duration was 6 s, and it was sub-divided into two 3-s phases: pre-naming and post-naming (Figure 1). Brain Sci. 2020,10, 977 7 of 18 Brain Sci. 2020, 10, x FOR PEER REVIEW 7 of 18 Figure 1. Structure of a sample trial from the lexical processing looking-while-listening (LWL) paradigm. Procedure. Infants’ lexical processing skills were assessed using the LWL procedure (see [37]). Infants sat on their mother’s lap approximately 60 cm away from a 19-inch square LG monitor and listened to auditory stimuli presented through an Edirol MA-15D speaker (Roland Corporation) placed directly below the monitor. Mothers listened to mixed auditory sounds via headphones to prevent them from interfering with the infants’ performance. In addition, mothers were instructed to divert their eyes away from the computer monitor to prevent their own gaze from interfering with the eye-tracker’s recording, and they were asked to avoid pointing to the screen or talking to their infant during the task. Infants’ gaze duration and direction were recorded using a Tobii X120 eye tracker(Tobii Tachnology AB). A researcher observed the infant from an adjoining control room via a live feed using a webcam located on the top of the computer monitor and directed towards the infant’s face. Prior to beginning the task, infants completed a five-point infant calibration routine. Next, infants proceeded to the test, which consisted of 24 test trials and 4 filler trials [49]. The 24 test trials were divided into 4 blocks. Each block consisted of six trials, one for each target word. Infants were, therefore, exposed to each target word 4 times. The order of trials within each block was randomized. In two blocks, the target words were presented using the carrier phrase “Where is the (target)?”, and in two blocks, they were presented using the phrase “Look at the (target)!”. In each block, the target picture was presented an equal number of times on the left and right side. Before the commencement of each trial, infants were presented with an attention-getter stimulus. The presentation of each trial was controlled by the researcher upon the infant fixating their gaze on the attention-getter stimulus. Between each block, the infants were presented with a filler trial aimed to maintain their attention to the screen, which displayed the images of four familiar images (that were not part of the test trials) accompanied by the exclamations ‘wow’ and ‘look’ recorded by the same speaker. The total duration of the task was approximately five minutes. Figure 1. Structure of a sample trial from the lexical processing looking-while-listening (LWL) paradigm. Procedure. Infants’ lexical processing skills were assessed using the LWL procedure (see [ 37 ]). Infants sat on their mother’s lap approximately 60 cm away from a 19-inch square LG monitor and listened to auditory stimuli presented through an Edirol MA-15D speaker (Roland Corporation) placed directly below the monitor. Mothers listened to mixed auditory sounds via headphones to prevent them from interfering with the infants’ performance. In addition, mothers were instructed to divert their eyes away from the computer monitor to prevent their own gaze from interfering with the eye-tracker’s recording, and they were asked to avoid pointing to the screen or talking to their infant during the task. Infants’ gaze duration and direction were recorded using a Tobii X120 eye tracker(Tobii Tachnology AB). A researcher observed the infant from an adjoining control room via a live feed using a webcam located on the top of the computer monitor and directed towards the infant’s face. Prior to beginning the task, infants completed a five-point infant calibration routine. Next, infants proceeded to the test, which consisted of 24 test trials and 4 filler trials [ 49 ]. The 24 test trials were divided into 4 blocks. Each block consisted of six trials, one for each target word. Infants were, therefore, exposed to each target word 4 times. The order of trials within each block was randomized. In two blocks, the target words were presented using the carrier phrase “Where is the (target)?”, and in two blocks, they were presented using the phrase “Look at the (target)!”. In each block, the target picture was presented an equal number of times on the left and right side. Before the commencement of each trial, infants were presented with an attention-getter stimulus. The presentation of each trial was controlled by the researcher upon the infant fixating their gaze on the attention-getter stimulus. Between each block, the infants were presented with a filler trial aimed to maintain their attention to the screen, which displayed the images of four familiar images (that were not part of the test trials) accompanied by the exclamations ‘wow’ and ‘look’ recorded by the same speaker. The total duration of the task was approximately five minutes. Data processing. Two rectangular areas of interest (AoIs) encompassing each object were defined for each test trial. The AoIs were located on the right and left sides of the screen encompassing the visual referent. The size and position of the AoIs were identical across all trials. Raw gaze data were extracted from Tobii Studio software and were processed using Eye Tracking RinR[ 50 ]. First, segments with data loss were identified, and trials with less than 25% detected gaze were excluded from all subsequent analyses. On average, infants contributed trials with 72% of Brain Sci. 2020,10, 977 8 of 18 detected gaze per trial, and this did not differ between the risk and the control groups, t(29) = − 0.321, p=0.750. Next, two response windows were defined: the pre-naming window and the post-naming window. The pre-naming window was from 0 to 3300 ms which included looks to the target and the distracter prior to the presentation of the target label. The post-naming window is the critical response window for analysis of the accuracy and latency of infants’ looking behaviors. This window was from 3300 to 4800 ms, which included looks to the target and the distracter in response to the target label. A 1500-ms response window was used for analyses as looking behaviors later in the trial are unlikely to represent responses to the target label [ 29 ]. Proportion of time looking (PTL) to the target out of the total looking time to the target (T) and the distracter (D) was calculated [PTL =T/(T +D)] for each time window and was used as the dependent variable in all analyses. Finally, trials were categorized based on whether the infant had fixated the target object (target-initial trials) or distracter object (distracter-initial trials) at the onset of the target word [ 37 ]. This enabled an onset contingent analysis to assess the latency (response time) required to switch gaze from distracter to target object after hearing the target name. 2.3.2. Expressive Vocabulary Checklist At 18 months of age of the children, their mothers completed the Australian English (OZI) [ 51 ] adaptation of the MacArthur–Bates Communicative Inventory [ 52 ]. Mothers were asked to identify words on the checklist that their child could produce. OZI data were unavailable for two infants due to the mother’s failure to complete the OZI. 3. Results First, our categorical approach employed independent t-tests to compare groups on accuracy and latency scores to test our first prediction that infants’ lexical processing abilities and expressive vocabulary sizes would be lower in the risk group compared with the control group. Further, one-sample t-tests were conducted to compare group accuracy scores against chance levels. Second, our continuous approach employed a correlational analysis to examine the extent to which mean depression and anxiety scores were correlated with infant accuracy and latency scores. Next, hierarchical regression analyses were used to test our second prediction that infants’ speed and accuracy of lexical processing, and their expressive vocabulary size, would be negatively correlated with maternal depression and anxiety scores. 3.1. Lexical Processing The time course of looking behaviors by the control and risk infants during the experimental trials is presented in Figure 2. First, infants’ PTL to the target during the pre-naming window was compared between the control and risk groups to ensure that infants in the two groups did not differ in their baseline gaze patterns to the two objects before the onset of the target label. An independent samples t-test showed no significant group differences; t(30) = − 0.306, p=0.762, d= − 0.112. Furthermore, one-sample t-tests confirmed that infants in the control, t(17) =−0.872, p=0.395, d=−0.209, and risk groups, t(13) = − 0.272, p=0.790, d= − 0.074, did not look at the targets significantly above chance levels (chance =0.5) prior to hearing its label. These analyses demonstrate that infants in the two groups did not exhibit any visual preferences to the referents used in the task that may have impacted their post-naming looking behaviors. Brain Sci. 2020,10, 977 9 of 18 Brain Sci. 2020, 10, x FOR PEER REVIEW 9 of 18 Figure 2. Time course plot of the proportion of looking time to the target across lexical processing trials (samples taken every 100 ms; shading around the line represents the 95% confidence intervals). Shaded rectangle area represents the post-naming analysis window. Next, post-naming performance was compared between the two groups on measures of accuracy and latency upon hearing the target label. 3.1.1. Accuracy The term accuracy refers to infants’ proportions of correct fixation (i.e., the proportion of looking to the target versus distracter following hearing the target label). An independent samples t-test showed that infants in the control group were significantly more likely to fixate the target than infants in the risk group after hearing its label; t(30) = 2.085, p = 0.046, d = 0.761. One-sample t-tests were conducted to determine whether infants were fixating the target above chance levels (0.5) (a Bonferroni correction was used to adjust the p-value to 0.025 to account for multiple comparisons). Infants in the control group fixated the target significantly above chance levels, t(17) = 5.082, p < 0.001, d = 1.181, but this was not the case for the risk group, t(13) = 2.265, p = 0.041, d = 0.602 (Figure 3). Figure 2. Time course plot of the proportion of looking time to the target across lexical processing trials (samples taken every 100 ms; shading around the line represents the 95% confidence intervals). Shaded rectangle area represents the post-naming analysis window. Next, post-naming performance was compared between the two groups on measures of accuracy and latency upon hearing the target label. 3.1.1. Accuracy The term accuracy refers to infants’ proportions of correct fixation (i.e., the proportion of looking to the target versus distracter following hearing the target label). An independent samples t-test showed that infants in the control group were significantly more likely to fixate the target than infants in the risk group after hearing its label; t(30) =2.085, p=0.046, d=0.761. One-sample t-tests were conducted to determine whether infants were fixating the target above chance levels (0.5) (a Bonferroni correction was used to adjust the p-value to 0.025 to account for multiple comparisons). Infants in the control group fixated the target significantly above chance levels, t(17) =5.082, p<0.001, d=1.181, but this was not the case for the risk group, t(13) =2.265, p=0.041, d=0.602 (Figure 3). Brain Sci. 2020,10, 977 16 of 18 16. NICHD Early Child Care Research Network. Chronicity of maternal depressive symptoms, maternal sensitivity, and child functioning at 36 months. Dev. Psychol. 1999,35, 1297–1310. [CrossRef] 17. Bayley, N. Bayley Scales of Infant and Toddler Development; PsychCorp, Pearson: San Antonio, TX, USA, 2006. 18. Grant, K.-A.; McMahon, C.; Austin, M.-P. Maternal anxiety during the transition to parenthood: A prospective study. J. Affect. Disord. 2008,108, 101–111. [CrossRef] [PubMed] 19. Kaitz, M.; Maytal, H.R.; Devor, N.; Bergman, L.; Mankuta, D. Maternal anxiety, mother–infant interactions, and infants’ response to challenge. Infant Behav. Dev. 2010,33, 136–148. [CrossRef] 20. Kuhl, P.K. Is speech learning “gated” by the social brain? Dev. Sci. 2007,10, 110–120. [CrossRef] 21. Rowe, M.L.; Snow, C.E. Analyzing input quality along three dimensions: Interactive, linguistic, and conceptual. J. Child Lang. 2019,1, 17. [CrossRef] 22. Kuhl, P.K. Early language acquisition: Cracking the speech code. Nat. Rev. Neurosci. 2004 ,5, 831–843. [CrossRef] 23. Huang, Z.J.; Lewin, A.; Mitchell, S.J.; Zhang, J. Variations in the relationship between maternal depression, maternal sensitivity, and child attachment by race/ethnicity and nativity: Findings from a nationally representative cohort study. Matern. Child Health J. 2012,16, 40–50. [CrossRef] 24. Gilkerson, J.; Richards, J.A.; Warren, S.F.; Oller, D.K.; Russo, R.; Vohr, B. Language Experience in the Second Year of Life and Language Outcomes in Late Childhood. Pediatrics 2018,142, e20174276. [CrossRef] 25. Romeo, R.R.; Leonard, J.A.; Robinson, S.T.; West, M.R.; Mackey, A.P.; Rowe, M.L.; Gabrieli, J.D.E. Beyond the 30-Million-Word Gap: Children’s Conversational Exposure Is Associated With Language-Related Brain Function. Psychol. Sci. 2018,29, 700–710. [CrossRef] [PubMed] 26. Weisleder, A.; Fernald, A. Talking to children matters: Early language experience strengthens processing and builds vocabulary. Psychol. Sci. 2013,24, 2143–2152. [CrossRef] [PubMed] 27. Fernald, A.; Pinto, J.P.; Swingley, D.; Weinberg, A.; McRoberts, G.W. Rapid Gains in Speed of Verbal Processing by infants in the 2nd year. Psychol. Sci. 1998,9, 228–231. [CrossRef] 28. Fernald, A. Speech to infants as hyperspeech: Knowledge-driven processes in early word recognition. Phonetica 2000,57, 242–254. [CrossRef] 29. Fernald, A.; Perfors, A.; Marchman, V.A. Picking up speed in understanding: Speech processing efficiency and vocabulary growth across the 2nd year. Dev. Psychol. 2006,42, 98. [CrossRef] 30. Marchman, V.A.; Fernald, A. Speed of word recognition and vocabulary knowledge in infancy predict cognitive and language outcomes in later childhood. Dev. Sci. 2008,11, F9–F16. [CrossRef] 31. Marchman, V.A.; Adams, K.A.; Loi, E.C.; Fernald, A.; Feldman, H.M. Early language processing efficiency predicts later receptive vocabulary outcomes in children born preterm. Child Neuropsychol. 2016 ,22, 649–665. [CrossRef] 32. Marchman, V.A.; Loi, E.C.; Adams, K.A.; Ashland, M.; Fernald, A.; Feldman, H.M. Speed of language comprehension at 18 months old predicts school-relevant outcomes at 54 months old in children born Preterm. J. Dev. Behav. Pediatr. 2018,39, 246–253. [CrossRef] 33. Morgan, P.L.; Farkas, G.; Hillemeier, M.M.; Hammer, C.S.; Maczuga, S. 24-Month-Old Children With Larger Oral Vocabularies Display Greater Academic and Behavioral Functioning at Kindergarten Entry. Child Dev. 2016,86, 1351–1370. [CrossRef] 34. Donnelly, S.; Kidd, E. Individual differences in lexical processing efficiency and vocabulary in toddlers: A longitudinal investigation. J. Exp. Child Psychol. 2020,192, 104781. [CrossRef] 35. Fenson, L.; Marchman, V.A.; Thal, D.J.; Dale, P.S.; Reznick, J.S. MacArthur-Bates Communicative Development Inventories; Paul H. Brookes Publishing Co.: Baltimore, MD, USA, 2007. 36. Feldman, H.M.; Dollaghan, C.A.; Campbell, T.F.; Kurs-Lasky, M.; Janosky, J.E.; Paradise, J.L. Measurement properties of the MacArthur Communicative Development Inventories at ages one and two years. Child Dev. 2000,71, 310–322. [CrossRef] [PubMed] 37. Fernald, A.; Zangl, R.; Portillo, A.L.; Marchman, V.A. Looking while listening: Using eye movements to monitor spoken language. In Developmental Psycholinguistics: On-Line Methods in Children’s Language Processing; John Benjamins Publishing: Amsterdam, The Netherlands, 2008; Volume 44, p. 97. 38. Eaton, W.W.; Smith, C.; Ybarra, M.; Muntaner, C.; Tien, A. Center for Epidemiologic Studies Depression Scale: Review and revision (CESD and CESD-R). In The Use of Psychological Testing for Treatment Planning and Outcomes Assessment: Instruments for Adults; Lawrence Erlbaum Associates Publishers: Mahwah, NJ, USA, 2004. Brain Sci. 2020,10, 977 17 of 18 39. Engle, P.L. Maternal mental health: Program and policy implications. Am. J. Clin. Nutr. 2009 ,89, 963S–966S. [CrossRef] [PubMed] 40. Van Dam, N.T.; Earleywine, M. Validation of the Center for Epidemiologic Studies Depression Scale—Revised (CESD-R): Pragmatic depression assessment in the general population. Psychiatry Res. 2011 ,186, 128–132. [CrossRef] [PubMed] 41. Spielberger, C.D.; Gorsuch, R.L.; Lushene, R.; Vagg, P.R.; Jacobs, G.A. Manual for the State-Trait Anxiety Inventory; Consulting Psychologists Press: Palo Alto, CA, USA, 1983. 42. Glasheen, C.; Richardson, G.A.; Fabio, A. A systematic review of the effects of postnatal maternal anxiety on children. Arch. Women’s Ment. Health 2010,13, 61–74. [CrossRef] 43. Murray, L.; Halligan, S.; Cooper, P. Effects of postnatal depression on mother–infant interactions and child development. In Handbook of Infant Development, 2nd ed.; Bremner, G., Wachs, T., Eds.; Wiley-Blackwell: Chichester, UK, 2010; Volume 2, pp. 192–220. 44. Song, J.Y.; Demuth, K.; Morgan, J. Effects of the acoustic properties of infant-directed speech on infant word recognition. J. Acoust. Soc. Am. 2010,128, 389–400. [CrossRef] [PubMed] 45. Boersma, P.; Weenink, D. Praat, a system for doing phonetics by computer, version 3.4. Inst. Phon. Sci. Univ. Amst. Rep. 1996,132, 182. 46. Fernald, A.; Taeschner, T.; Dunn, J.; Papousek, M.; de Boysson-Bardies, B.; Fukui, I. A cross-language study of prosodic modifications in mothers’ and fathers’ speech to preverbal infants. J. Child Lang. 1989 ,16, 477–501. [CrossRef] 47. Ma, W.; Golinkoff, R.M.; Houston, D.M.; Hirsh-Pasek, K. Word Learning in Infantand Adult-Directed Speech. Lang. Learn. Dev. 2011,7, 185–201. [CrossRef] 48. Thiessen, E.D.; Hill, E.A.; Saffran, J.R. Infant-directed speech facilitates word segmentation. Infancy 2005,7, 53–71. [CrossRef] 49. Swingley, D.; Aslin, R.N. Spoken word recognition and lexical representation in very young children. Cognition 2000,76, 147–166. [CrossRef] 50. Dink, J.W.; Ferguson, B. eyetrackingR: An R Library for Eye-Tracking Data Analysis. 2015. Available online: www.eyetracking-r.com (accessed on 1 April 2019). 51. Kalashnikova, M.; Schwarz, I.C.; Burnham, D. OZI: Australian English Communicative Development Inventory. First Lang. 2016,36, 407–427. [CrossRef] 52. Fenson Larry Dale, P.S.; Reznick, J.S.; Bates, E.; Thal, D.J.; Pethick, S.J.; Stiles, J. Variability in early communicative development. Monogr. Soc. Res. Child Dev. 1994. [CrossRef] 53. Hurtado, N.; Marchman, V.A.; Fernald, A. Does input influence uptake? Links between maternal talk, processing speed and vocabulary size in Spanish-learning children. Dev. Sci. 2008. [CrossRef] [PubMed] 54. Egger, J.; Rowland, C.F.; Bergmann, C. Improving the robustness of infant lexical processing speed measures. Behav. Res. Methods 2020,52, 2188–2201. [CrossRef] 55. Milgrom, J.; Westley, D.T.; Gemmill, A.W. The mediating role of maternal responsiveness in some longer term effects of postnatal depression on infant development. Infant Behav. Dev. 2004 ,27, 443–454. [CrossRef] 56. Masi, G.; Millepiedi, S.; Mucci, M.; Poli, P.; Bertini, N.; Milantoni, L. Generalized anxiety disorder in referred children and adolescents. J. Am. Acad. Child Adolesc. Psychiatry 2004,43, 752–760. [CrossRef] 57. Golinkoff, R.M.; Can, D.D.; Soderstrom, M.; Hirsh-Pasek, K. (Baby) Talk to Me: The Social Context of Infant-Directed Speech and Its Effects on Early Language Acquisition. Curr. Dir. Psychol. Sci. 2015 ,24, 339–344. [CrossRef] 58. Bettes, B.A. Maternal Depression and Motherese: Temporal and Intonational Features. Child Dev. 1988 ,59, 1089–1096. [CrossRef] 59. Lam-Cassettari, C.; Kohlhoff, J. Effect of maternal depression on infant-directed speech to prelinguistic infants: Implications for language development. PLoS ONE 2020,15, e0236787. [CrossRef] 60. Porritt, L.; Zinser, M.; Bachorowski, J.; Kaplan, P. Depression diagnoses and fundamental frequency-based acoustic cues in maternal infant-directed speech. Lang. Learn. Dev. 2015,10, 51–67. [CrossRef] 61. Kaplan, P.; Jung, P.; Ryther, J.; Zarlengo-Strouse, P.; Zar, P.; We, F.; Scheuneman, D. Infant-Directed Versus Adult-Directed Speech as Signals for Faces. Dev. Psychol. 1996,32, 880–891. [CrossRef] 62. Kaplan, P.S.; Bachorowski, J.; Zarlengo-Strouse, P. Child-directed speech produced by mothers with symptoms of depression fails to promote associative learning in 4-month-old infants. Child Dev. 1999 ,70, 560–570. [CrossRef] [PubMed] Brain Sci. 2020,10, 977 18 of 18 63. Kaplan, P.S.; Dungan, J.K.; Zinser, M.C. Infants of chronically depressed mothers learn in response to male, but not female, infant-directed speech. Dev. Psychol. 2004,40, 140–148. [CrossRef] [PubMed] 64. Austin, M.P.; Priest, S.R. Clinical issues in perinatal mental health: New developments in the detection and treatment of perinatal mood and anxiety disorders. Acta Psychiatr. Scand. 2005,112, 97–104. [CrossRef] 65. Dennis, C.L.; Stewart, D.E. Treatment of postpartum depression, part 1: A critical review of biological interventions. J. Clin. Psychiatry 2004,65, 1242–1251. [CrossRef] 66. Field, T. Maternal Depression Effects on Infants and Early Interventions 1. Prev. Med. 1998 ,203, 200–203. [CrossRef] 67. Murray, L.; Cooper, P. Intergenerational transmission of affective and cognitive processes associated with depression: Infancy and the pre-school years. In Unipolar Depression: A Lifespan Perspective; Oxford University Press: Oxford, UK, 2003; pp. 17–46. 68. Nielsen, D.; Videbech, P.; Hedegaard, M.; Dalby, J.; Secher, N.J. Postpartum depression: Identification of women at risk. BJOG Int. J. Obstet. Gynaecol. 2000,107, 1210–1217. [CrossRef] 69. Roberts, M.Y.; Curtis, P.R.; Sone, B.J.; Hampton, L.H. Association of parent training with child language development: A systematic review and meta-analysis. JAMA Pediatr. 2019,173, 671–680. [CrossRef] Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).