Tracciare l'impatto della devianza comunicativa materna sul coinvolgimento congiunto: il ruolo mediatorio dei comportamenti materni di supporto
Titolo originaleTracing the impact of maternal communication deviance on joint engagement: The mediating role of maternal supportive behaviors.
Infant mental health journal · 2027
Sintesi in italiano
Questo studio longitudinale, condotto nel Regno Unito e pubblicato nel 2027, ha esaminato l'influenza della devianza comunicativa materna, valutata durante la gravidanza, sul coinvolgimento congiunto madre-figlio e processi correlati a 30 mesi di età. La devianza comunicativa rappresenta uno stile comunicativo disorganizzato che può interferire con la costruzione condivisa di significato nelle interazioni genitore-figlio ed è considerata un potenziale fattore di rischio per difficoltà di salute mentale future. Hanno partecipato 229 madri prevalentemente di origine britannica (95,9%), con un'età media di 27,6 anni (DS = 6,1). Le osservazioni di interazioni di gioco madre-figlio hanno mostrato che la devianza comunicativa prediceva un minore coinvolgimento congiunto totale, inclusi i sottotipi supportato (β = -0,18, p = 0,007), coordinato (β = -0,17, p = 0,016) e simbolico-infuso (β = -0,24, p < 0,001). Inoltre, era associata a una riduzione significativa dei comportamenti di supporto materni (β = -0,40, p < 0,001) e dell'attenzione diadica (β = -0,31, p < 0,001). Le analisi di mediazione hanno indicato che i comportamenti di supporto materni spiegavano completamente l'associazione tra devianza comunicativa e risultati del coinvolgimento congiunto. Questi risultati sottolineano come la devianza comunicativa possa limitare le opportunità di attenzione condivisa e costruzione di significato del bambino, suggerendo che il miglioramento dei comportamenti materni di supporto potrebbe rappresentare un target per interventi volti a promuovere la salute mentale precoce e la qualità dell'interazione genitore-figlio. Il disegno di studio è longitudinale osservazionale.
Sintesi generata con IA. Verifica metodi, risultati e limiti nella pubblicazione originale.
Abstract originale
Communication Deviance is a disorganized communicative style that can disrupt shared meaning-making in parent-child interactions and is considered a risk factor for later mental health difficulties. This longitudinal study, conducted in the UK, examined whether maternal Communication Deviance assessed during pregnancy predicted joint engagement and related processes. Participants were 229 mothers (95.9% White British; M = 27.6 years, SD = 6.1). Maternal Communication Deviance was assessed antenatally, and mother-child dyads were observed during play at 30-months. Communication Deviance predicted lower total joint engagement, including supported (β = -.18, p = .007), coordinated (β = -.17, p = .016), and symbol-infused joint engagement (β = -.24, p < .001), and was associated with reduced supportive behaviors (β = -.40, p < .001) and dyadic attunement (β = -.31, p < .001). Mediation analyses indicated that supportive behaviors accounted for the associations between Communication Deviance and joint engagement outcomes. These findings suggest that Communication Deviance may constrain children's opportunities for shared attention and meaning making. The study supports developmental models that highlight early communicative processes as foundational to socio-cognitive and relational development and identifies supportive behaviors as a potential target for interventions promoting early mental health and parent-child interaction. KEY FINDINGS: Higher maternal Communication Deviance during pregnancy predicted less Total, Supported, Coordinated, and Symbol-infused joint engagement at 30 months. Reduced maternal supportive behaviors fully explained the link between Communication Deviance and lower joint engagement and dyadic attunement. Maternal Communication Deviance may act as an early environmental risk factor, highlighting the importance of supporting maternal responsiveness to promote healthy development. STATEMENT OF RELEVANCE: This study examines how maternal Communication Deviance measured during pregnancy predicts joint engagement and dyadic attunement at 30 months, processes central to early socio-cognitive and relational development. By identifying maternal supportive behaviors as a key mechanism linking Communication Deviance to interactional quality, the research highlights a modifiable target for early intervention. Findings advance understanding of early environmental risk factors and inform strategies to support parent-child interactions, directly addressing foundational processes relevant to child and early childhood mental health.
Parole chiave
Leggi il testo integrale originale
Testo estratto dalla pubblicazione: sono esclusi immagini, media, supplementi e parti soggette a diritti distinti.
Tracing the impact of maternal communication deviance on joint engagement: The mediating role of maternal supportive behaviors SOUSA et al.
Abstract
Communication Deviance is a disorganized communicative style that can disrupt shared meaning‐making in parent–child interactions and is considered a risk factor for later mental health difficulties. This longitudinal study, conducted in the UK, examined whether maternal Communication Deviance assessed during pregnancy predicted joint engagement and related processes. Participants were 229 mothers (95.9% White British; M = 27.6 years, SD = 6.1). Maternal Communication Deviance was assessed antenatally, and mother–child dyads were observed during play at 30‐months. Communication Deviance predicted lower total joint engagement, including supported (β = –.18, p = .007), coordinated (β = –.17, p = .016), and symbol‐infused joint engagement (β = –.24, p < .001), and was associated with reduced supportive behaviors (β = –.40, p < .001) and dyadic attunement (β = –.31, p < .001). Mediation analyses indicated that supportive behaviors accounted for the associations between Communication Deviance and joint engagement outcomes. These findings suggest that Communication Deviance may constrain children's opportunities for shared attention and meaning making. The study supports developmental models that highlight early communicative processes as foundational to socio‐cognitive and relational development and identifies supportive behaviors as a potential target for interventions promoting early mental health and parent–child interaction.
Key Findings
Higher maternal Communication Deviance during pregnancy predicted less Total, Supported, Coordinated, and Symbol‐infused joint engagement at 30 months.
Reduced maternal supportive behaviors fully explained the link between Communication Deviance and lower joint engagement and dyadic attunement.
Maternal Communication Deviance may act as an early environmental risk factor, highlighting the importance of supporting maternal responsiveness to promote healthy development.
Statement of Relevance
This study examines how maternal Communication Deviance measured during pregnancy predicts joint engagement and dyadic attunement at 30 months, processes central to early socio‐cognitive and relational development. By identifying maternal supportive behaviors as a key mechanism linking Communication Deviance to interactional quality, the research highlights a modifiable target for early intervention. Findings advance understanding of early environmental risk factors and inform strategies to support parent–child interactions, directly addressing foundational processes relevant to child and early childhood mental health.
1COMMUNICATION DEVIANCE AS A RISK FACTOR FOR PSYCHOPATHOLOGY
Communication Deviance refers to a communicative style characterized by vagueness, ambiguity, and disruptions in shared meaning between interlocutors (Singer, 1965a, b; Singer et al., 1978; Tienari & Wahlberg, 2008; Wynne, 1963a; Wynne et al., 1963, 1977b, 1984). It encompasses a range of linguistic and interactional disturbances that can hinder mutual understanding during social exchange (e.g., unclear referents, contradictory statements, or poorly timed contributions).
Communication Deviance has been most extensively studied in the context of schizophrenia‐spectrum disorders, where elevated levels in parents have been associated with increased risk of thought disorder and broader psychopathology in offspring (de Sousa et al., 2014; Roisko et al., 2014). Longitudinal findings indicate that Communication Deviance precedes the onset of such difficulties, supporting its role as an environmental risk factor rather than a response to child characteristics (Goldstein, 1987). However, Communication Deviance is not specific to schizophrenia; similar communicative patterns have been observed in families experiencing other forms of psychopathology, relational distress, and needs, suggesting it is better conceptualized as a dimensional and transdiagnostic construct (Miklowitz & Stackman, 1992).
Evidence from adoption studies further suggests that Communication Deviance interacts with genetic vulnerability, such that its impact on offspring outcomes depends on broader risk context (Wahlberg et al., 1997, 2000). Overall, this body of work supports the view that parental CD constitutes an environmental risk factor in the development of psychopathology. However, the mechanisms through which CD operates in early caregiving contexts remain poorly understood.
2PLAUSIBLE DEVELOPMENTAL MECHANISM
One proposed pathway is through the disruption of early interactional processes in childhood. In particular, Communication Deviance may interfere with joint engagement which refers to episodes of mutual attention, communication, and emotional connection between caregiver and child around a common interest or task (Adamson et al., 2004, 2009, 2012). These interactions are critical for the development of social understanding, emotional regulation, and cognitive skills (Adamson et al., 2019) and the quality of nonverbal and verbal interactions (e.g., dyadic connectedness and fluency) seemed to be important for the child's cognitive and social development (Hirsh‐Pasek et al., 2015).
If caregiver communication is vague, inconsistent, or poorly timed, opportunities for sustained shared attention and meaning making may be reduced. In this way, Communication Deviance may constrain the quality of early interactional experiences that support development. Although this account is theoretically plausible, it has not yet been directly tested.
2.1Joint engagement
Joint engagement refers to sustained, reciprocal interactions between a child and a social partner in which both participants are actively involved in a shared activity and maintain a common focus over time (Adamson, 2018; Adamson et al., 2014; Trevarthen & Aitken, 2001). These engagement states are characterized not only by shared attention, but also by the quality of the interaction, including affective attunement, reciprocity, and the use of gestures, actions, and language. This construct is grounded in sociocultural developmental theory (Vygotsky & Cole, 1978) and early work on infants’ emerging capacity to share attention with caregivers (Scaife & Bruner, 1975).
Within these broader engagement states, joint attention is best conceptualized as a set of social‐communicative skills that enable the coordination of attention between partners, including behaviors such as responding to joint attention, initiating joint attention, gaze following, and pointing (Adamson et al., 2019; Mundy, 2018). Rather than being synonymous with joint engagement, these skills support the initiation, maintenance, and elaboration of joint engagement episodes. Thus, joint attention reflects the child's moment‐to‐moment attentional coordination abilities, whereas joint engagement captures the broader, dynamic context in which these skills are embedded and expressed (Adamson et al., 2019).
Disruptions to both joint attention skills and joint engagement states are well documented in autism and are thought to contribute to broader differences in social communication and cognition (Adamson et al., 2009; Nyström et al., 2019). These early social interactions are supported by developing perceptual‐motor capacities (Butterworth & Jarrett, 1991) and neurocognitive mechanisms that orient the child toward social information (Baron‐Cohen, 1997), but they are fundamentally shaped within relational and affective contexts (Hobson, 2005; Hobson & Hobson, 2011; Racine & Carpendale, 2007).
During the first year of life, infants begin to engage in behaviors that recruit others’ attention, such as gaze following and pointing, with more consistent use emerging toward the end of the first year (Salter & Carpenter, 2025). Across the second year, episodes of joint engagement become increasingly sustained, differentiated, and socially coordinated (Bakeman & Adamson, 1984; Charman et al., 2000; Mundy & Gomes, 1998). These interactions provide a critical context for development, supporting the emergence of social understanding (Mundy, 2018; Tomasello, 2014), language acquisition (Loy et al., 2018; Mundy et al., 2007; Tomasello & Farrar, 1986), cognitive development (Mundy & Newell, 2007; Mundy & Sigman, 2015), and emotion regulation (Morales et al., 2005).
Joint engagement episodes are inherently multi‐modal, integrating gaze, gesture, touch, and language (Schatz et al., 2022; Suarez‐Rivera et al. et al., 2022). Caregivers play a central role in structuring and sustaining these interactions through sensitivity, contingent responding, and scaffolding (Bakeman & Adamson, 1984; Hustedt & Raver, 2002). For example, following the child's focus of attention supports the emergence of coordinated joint engagement, whereas more directive behaviors may characterize supported joint engagement (Mendive et al., 2013). In naturalistic contexts, caregiver involvement, particularly when it is multi‐modal, substantially increases the duration and complexity of play interactions (Schatz et al., 2022). Indeed, caregiver behaviors such as sensitivity, cognitive stimulation, and positive affect are positively associated with both the frequency and duration of joint engagement episodes (Koşkulu et al., 2021), whereas difficulties in affective attunement may reduce the likelihood of initiating shared engagement (Annie Yoon et al., 2014).
Importantly, the quality of parent‐child interaction appears to be a strong predictor of developmental outcomes. For example, dyadic fluency and connectedness during interaction predicts language outcomes beyond the quantity of linguistic input, particularly in socioeconomically disadvantaged contexts (Hirsh‐Pasek et al., 2015). Similarly, early caregiver attunement predicts later joint attention abilities, which in turn mediate associations with executive function development (Brandes‐Aitken et al., 2020). Time spent in joint engagement and caregivers’ responsiveness to the child's attentional focus are also associated with the development of gesture and early language (Carpenter et al., 1998), while infants’ responsiveness to joint attention bids predicts later vocabulary outcomes (Morales et al., 2000).
Across development, joint engagement becomes increasingly symbol‐infused, providing a foundation for language use and meaning‐making (Adamson et al., 2004). By early childhood, these interactions expand to include references to absent objects and events, supporting more complex communicative exchanges about the past and future (Adamson & Bakeman, 2006; Adamson et al., 2014).
3PRESENT STUDY
The study examined whether antenatal maternal Communication Deviance (at 32 weeks’ gestation) predicts reduced Joint Engagement and related interactional processes at 30 months in primiparous mothers. It was hypothesized that higher maternal Communication Deviance would be associated with less time in Joint Engagement states, less Maternal Scaffolding, Symbol Highlighting, Following the Child's focus, less Fluency and Connectedness of communication and Sustainability of Shared topic in the dyad and poorer Expressive Language, and that the relationships between maternal Communication Deviance, Joint Engagement, dyadic attunement variables, and the child's Expressive Language would be mediated by maternal behaviors such as Scaffolding, Symbol Highlighting, and Following the Child's Focus.
4METHOD
4.1Recruitment and sample
The study draws on data from the Wirral Child Health and Development Study (WCHADS; Sharp et al., 2012) which is a longitudinal study that aimed to identify early social, emotional, and biological factors involved in the development of conduct disorder. The study employed a stratified two‐stage cluster design, recruiting a larger sample of primiparous mothers during pregnancy, from which a random subsample was drawn for intensive assessment (n = 316). Participants were English‐speaking, aged ≥18, and sought antenatal care at 12 weeks’ gestation between February 2007 and October 2008 in the Northwest of the UK. A detailed flowchart of sampling, recruitment and attrition rates is available elsewhere (Sharp et al., 2012). Only data from the intensive subsample were used, as this included detailed observational data on mother‐child interactions. Of the 316 participants, 29 declined or were unable to complete the 5‐min speech sample task. Of the 287 who provided a 5‐min speech sample task during pregnancy, 230 attended the 30‐months’ assessment involving mother‐child play observations. Reasons for non‐attendance included withdrawal from the study, illness, and other commitments. Participants were required to be English‐speaking, and the sample was predominantly monolingual English‐speaking; however, formal mother‐tongue language status was not systematically recorded.
5PROCEDURE
5.1weeks speech samples (antenatal)
At 32 weeks’ gestation, mothers were invited to provide speech samples about their unborn child (Leeb et al., 1991). The mother was instructed using the following prompt:
“I would like to hear your thoughts and feelings about your baby at the moment, in your own words without me interrupting. When I ask you to begin, I would like you to speak for 5 minutes, tell me what your impressions have been of your baby whilst you've been pregnant. Have you got any questions before we start?”
All interaction samples were reviewed to ensure they reflected continuous, naturalistic engagement; no minimum word‐count threshold was set a priori for inclusion.
5.2Communication Deviance
Communication Deviance was scored with the Interactional Communication Deviance scoring scheme which was originally developed for family interaction (Velligan, 1985). The scheme focuses on eight types of communication disturbances derived from research on Communication Deviance (Doane & Singer, 1977; Singer, 1965a, b; Wynne, 1963a,b; Wynne et al., 1977, 1963b): (1) abandoned, abruptly ceased, uncorrected remarks, (2) unintelligible remarks, (3) contradictions, denials and retractions, (4) ambiguous referents, (5) extraneous questions and remarks, (6) tangential, inappropriate responses to questions or remarks, (7) odd word usage/odd sentence construction, and (8) reiteration. Code examples are available in a previous publication (de Sousa et al., 2019) and in Table 1.
The scheme demonstrates good reliability and construct validity (Velligan et al., 1990) and has been applied in clinical (Velligan et al., 1995, 1996), high‐risk (Velligan et al., 1988) and natural speech samples (Docherty, 1993), including 5‐min speech samples (Kymalainen & Weisman de Mamani, 2008; Kymalainen et al., 2006).
The Communication Deviance raw score reflects the total frequency of coded Communication Deviance instances across the eight categories. The Communication Deviance ratio was calculated as: (number of Communication Deviance instances ÷ total word count) × 100, providing a rate‐adjusted index of Communication Deviance. The minimum possible score is 0, with no fixed maximum value, as this depends on speech length and frequency of Communication Deviance occurrences (Hirsch & Leff, 1971, 1975).
The speech samples were audio‐recorded and transcribed by the WCHADS team, then coded for Communication Deviance. The period of training involved reading relevant papers in the Communication Deviance field (Singer & Wynne, 1966), the scoring manual (Velligan, 1985), and regular meetings between raters and communication exchanges with other researchers in the field. Raters were blind to any background information about the participants using anonymized transcripts and audio‐recordings. Coders were trained to distinguish dialectal variation from Communication Deviance by focusing on functional communicative clarity rather than conformity to Standard English. Non‐standard grammatical forms or regional linguistic features were not coded as Communication Deviance unless they resulted in ambiguity, disruption of referential coherence, or breakdown in shared meaning within the speech sample. The first author completed all primary coding, and a second independent coder completed reliability coding on approximately 10% of transcripts. The second coder was a local resident of the region, providing contextual familiarity with regional speech patterns. Any discrepancies between coders were resolved through discussion and reference to the coding manual. Some of the categories were very infrequent (e.g., reiteration), but the inter‐rater reliability was good (the intraclass correlations for the items ranged from .77 to .97).
5.330 months play protocol
The mother and child were first presented with three cloth bags numbered 1, 2, and 3, which were left on the floor of the room in order and close to the mother. This structured play procedure is consistent with standardized observational parent‐child play paradigms, such as the “three‐box” task used in the NICHD Early Child Care Study, which provide a standardized context for observing parent‐child interaction across dyads. Although the three‐box task was not developed specifically to assess joint engagement, it similarly offers comparable opportunities for shared attention and coordinated interaction. The mother was then given the following prompt from the research assistant:
“This activity will take about 10 minutes. We would like for your child to spend some time with each of the three toys in the bags. Please have [child] start with the #1 bag and finish with the #3 bag. During this activity, you may play with [child] if you would like. Do you have any questions before we get started?”
Each cloth bag contained a toy (a storybook, a toddler kitchen and a toy house, respectively). At the start, the mother was informed that the sound of a baby crying would be played for 30 seconds during the 10‐min session. This element formed part of the broader WCHADS protocol designed to assess maternal responses to child's distress and was included for all participants under standardized conditions. The present study did not analyze responses to this stimulus. Given its brief duration and uniform administration across participants, this component is unlikely to have systematically influenced the associations examined in the present analyses. After 10 min, the mother was told the task had ended by the researcher calling for “tidy‐up time”. The play was captured by four cameras, providing four angles of the interaction.
5.4Joint engagement
Joint engagement was coded using the Joint Engagement Rating Inventory (JERI; Adamson et al., 2004; 2009; 2012), a 7‐point observational rating system designed to assess the quality and quantity of caregiver–child engagement during interaction (Adamson et al., 2012). The JERI includes multiple dimensions of child engagement, caregiver behavior, child behavior, and dyadic interaction during structured play. A brief description of each selected item is provided in the text, and the full definitions and scoring anchors are presented in Table 2.
The specific JERI items used in this study were selected following detailed discussions with the developer. In addition, the first author spent 1 week in the developer's laboratory receiving training in JERI coding procedures, observing coding practices, and discussing study‐specific adaptations to ensure appropriate application of the measure to the current dataset.
For this study, 10 of the 17 available JERI items were selected based on their relevance to the study hypotheses. Item selection was guided by theoretical relevance to the proposed developmental model linking caregiver communicative behavior, joint engagement, and early socio‐cognitive development, rather than solely on empirical criteria. These included child‐centered engagement states (Total, Supported, Coordinated, and Symbol‐infused Joint Engagement), maternal behaviors (Scaffolding, Symbol Highlighting, and Following the Child's Focus), dyadic interaction measures (Sustainability of Shared Topic and Fluency and Connectedness) and child's Expressive Language. Although caregiver supportive behaviors are theoretically expected to support joint engagement, the JERI distinguishes between caregiver contributions to interaction and dyadic engagement states, permitting examination of their associations while recognizing their interdependence.
Importantly, joint engagement states are defined from the child's perspective, emphasizing that although the parent plays a critical role in shaping the interaction, the child's involvement and orientation toward the parent ultimately determine whether the parent becomes an integrated part of the child's play experience, either explicitly or implicitly.
Engagement states were first identified through systematic observation of the full interaction and then rated on a 7‐point scale that reflects both the proportion of time spent in each state and the qualitative features of engagement within that state, as defined by the JERI. Importantly, ratings capture not only duration but also the quality of engagement, coordination, and responsiveness between caregiver and child, as detailed in Table 2.
Total Joint Engagement captures the overall extent of the child's joint engagement with the caregiver around a shared object, event, or topic across the interaction. It provides a global measure of joint engagement, encompassing periods of both Supported and Coordinated Joint Engagement.
Supported Joint Engagement refers to periods of triadic engagement in which the child and caregiver are jointly engaged with a shared object or topic, with the caregiver providing support within the interaction and influencing the child's experience. This construct does not include situations in which the child and caregiver are merely attending to the same object or activity without the caregiver influencing the child's experience (e.g., parallel play). Ratings also reflect the quality of engagement, with a midpoint score indicating approximately one‐third of the interaction (around 3 min and 20 s) spent in supported joint engagement of moderate quality, or briefer but particularly high‐quality episodes.
Coordinated Joint Engagement describes interaction in which the child and caregiver share focus and the child actively coordinates attention between the object and the caregiver (e.g., through gaze shifts, gestures, or communicative acts). Ratings incorporate both time and quality, with a midpoint score indicating approximately one‐third of the interaction (around 3 min and 20 s) spent in Coordinated Joint Engagement of moderate quality or shorter but especially high‐quality episodes.
Symbolically Infused Joint Engagement captures episodes of supported or coordinated joint engagement that incorporate symbolic communication, such as spoken language, conventional gestures, signs, symbolic gestures (e.g., performing the finger movements for The Incy Wincy Spider while singing), or symbolic pretend play. These episodes extend beyond shared attention by involving symbols that represent objects, actions, or ideas beyond the immediate context. Ratings reflect both the frequency and the richness of symbolic communication demonstrated throughout the interaction.
Scaffolding refers to caregiver behaviors that support and extend the child's participation in interaction, play, or communication by providing appropriately calibrated assistance. This includes adjusting support to the child's level, prompting, modelling, and structuring activities to maintain engagement while encouraging increasing independence and complexity. Ratings reflect the extent and consistency of scaffolding across the interaction.
Symbol Highlighting refers to caregiver behaviors that intentionally draw attention to linguistic or symbolic information, such as emphasizing key words, labelling objects, repeating phrases, or pairing speech with actions to increase salience and meaning. Ratings reflect the frequency and continuity of these strategies.
Following in on the Child's Focus describes the caregiver's tendency to join and build on the child's existing focus of attention rather than redirecting it. This involves observing what the child is engaged with and responding in ways that sustain and extend that focus. Ratings reflect how consistently the caregiver aligns with and maintains the child's attentional focus over time.
Sustainability of Shared Topic refers to the extent to which caregiver and child maintain a consistent focus on the same topic, object, or activity across time. Higher ratings reflect longer and more coherent shared engagement, with a midpoint score indicating that topics are maintained for a moderate duration (several conversational or play turns across the interaction), and higher scores reflecting sustained topics across much or most of the session.
Fluency and Connectedness captures the smoothness, coordination, mutual responsiveness, and balance of the interaction. In the version of the JERI definition used in the present study, higher scores reflect balanced and reciprocal exchanges in which caregiver and child responses are clearly contingent on one another, contributing to a coherent and connected interactional flow. The detailed scoring anchors for this item are provided in Table 2.
Expressive Language refers to the child's demonstrated expressive language within the interaction. Ratings reflect both complexity and frequency of use, ranging from no expressive language to fluent sentence‐level speech. A midpoint score indicates the use of approximately 10 or more intelligible words with limited combinations, while higher scores reflect frequent and fluent use of multi‐word sentences. Full definitions and detailed scoring anchors for all items are provided in Table 2. The mean length of utterance was not calculated within this scoring system.
The first author received formal training in the JERI coding system at the developers’ laboratory, using gold‐standard videos and associated coding materials. Following this training, the first author coded a subgroup of WCHADS videos with guidance and support from staff at the developers’ laboratory. A second coder was subsequently trained by the first author using the same JERI coding materials, and coding guidance provided during the formal training. Although the second coder did not receive direct training from the JERI developers, they had access to the same coding framework used to establish the first author's competency.
To assess inter‐rater reliability, a subset of WCHADS videos (10%) was independently coded by the second coder. Given the ordinal nature of the JERI ratings, weighted Cohen's kappa coefficients with linear weighting were calculated for each domain. Reliability was good to excellent across categories, with weighted kappa values ranging from κw = .709 to .907. In addition, independent coders assigned ratings that were within one point of each other on the 7‐point scale for 91.7%–100% of observations across domains, indicating a high degree of coding agreement.
Coordinated Joint Engagement could not be reliably coded for six participants because portions of the video footage were of insufficient quality or visibility to permit accurate observation of coordinated attention. Other engagement states for these participants were sufficiently observable and were therefore reliably coded.
Given the conceptual and empirical overlap among several caregiver and dyadic interaction variables, composite constructs were specified to capture broader interactional dimensions. In particular, supportive behaviors (Scaffolding, Symbol Highlighting, and Following in on the Child's Focus) and dyadic interaction variables (Sustainability of Shared Topic and Fluency and Connectedness) were expected to reflect higher‐order constructs. These composites were theoretically motivated based on prior literature on caregiver support and interactional synchrony, with Principal Component Analysis used as a data‐reduction technique to derive parsimonious indices and reduce multicollinearity in subsequent analyses. Full details of Principal Component Analysis procedures are provided in the Statistical Analysis section.
6CONFOUNDERS
6.1Index of Multiple Deprivation
Socioeconomic status was determined using the revised Index of Multiple Deprivation (Noble et al., 2004). Postcode areas of England are ranked from the deprived (Index of Multiple Deprivation of 1) to the least deprived (Index of Multiple Deprivation of 32,482) based on seven domains of inequality (e.g., income deprivation, living environment deprivation, crime, etc.). All mothers were scored according to their area postal code.
6.2Reading ability
Maternal reading ability was estimated using the Wechsler Test of Adult Reading (WTAR), a standardized measure of premorbid verbal intelligence. The WTAR requires pronunciation of 50 irregularly spelled words, with testing discontinued after 12 consecutive errors. Each correct response scores one point (maximum = 50), and raw scores are converted to age‐ and education‐adjusted standard scores (Holdnack, 2001). WTAR performance correlates strongly with verbal IQ, verbal comprehension, and full‐scale IQ, supporting the validity as a proxy measure of verbal cognitive ability (Strauss et al., 2006).
7STATISTICAL ANALYSIS
All analyses were conducted in IBM SPSS Statistics (Version 27.0). Descriptive statistics summarized demographic, behavioral, and developmental characteristics, and bivariate correlations assessed associations and multicollinearity. To reduce multicollinearity and derive theoretically meaningful composite constructs, Principal Component Analysis were conducted on sets of highly correlated variables. Maternal supportive behaviors (Scaffolding, Symbol Highlighting, and Following in on Child's Focus) were entered into a Principal Component Analysis to derive a single latent factor representing maternal supportive behaviors. Similarly, Sustainability of Shared Topic and Fluency and Connectedness were entered into a separate Principal Component Analysis to derive a dyadic attunement factor. These composite variables were specified a priori based on theoretical considerations rather than derived purely empirically. Principal Component Analysis was used to confirm the extent to which these variables reflected shared underlying constructs and to provide parsimonious representations for use in mediation models, rather than to identify novel latent structures. In addition, individual variables were retained in initial regression analyses to examine specific associations, ensuring that findings were not solely dependent on composite measures. Linear regressions examined relationships between Communication Deviance and outcomes using unadjusted and adjusted models; adjusted models controlled for maternal age, reading ability, and Index of Multiple Deprivation. Mediation analyses tested indirect effects of Communication Deviance via maternal supportive behaviors. Mediation models were conducted using PROCESS (Hayes, 2013).
8RESULTS
8.1Characteristics of the sample and study variables
The mean age of the mothers was 27.58 years (standard deviation [SD] = 6.09), and they were predominantly White British (95.9%) reflecting the general demographics of the area in the UK (ONS, 2023). The mean reading ability score was 105.79 (SD = 6.53). The mothers were predominantly either married (40.4%) or in cohabitation with a partner (33.4%), with predominantly secondary school qualifications. Regarding the Index of Multiple Deprivation, the mean was 32.07 (SD = 20.1), reflecting high levels of deprivation, consistent with the study's catchment area. Table 3 also shows the means and standard deviations for both the Communication Deviance and the joint engagement states and related variables. Although mean ratings were relatively high across several joint engagement items, the full range of the scale (1–7) was utilized, and standard deviations indicated adequate variability, suggesting no substantial restriction of range or ceiling effects.
8.2Joint engagement
To test whether Communication Deviance was a significant predictor of total time spent on joint engagement, an unadjusted linear regression was performed to test the bivariate relationship. In the unadjusted model, Communication Deviance was a significant predictor Total Joint Engagement, B = −.38, SE = .09, β = −.27, t = −4.18, p < .001. The model accounted for 7.1% of the variance, R2
= .071, F(1228) = 17.44, p < .001. The same model, but adjusted for the mother's age, Index of Multiple Deprivation, and reading ability, was run. The overall model was significant, F(4225) = 5.64, p < .001, explaining 9.1% of the variance (R2
= .091). Communication Deviance continued to significantly predict Total Joint Engagement, B = −.33, SE = .09, β = −.23, t = −3.49, p < .001 but maternal reading ability (B = .10, p = .416), age (B = .19, p = .151), and Index of Multiple Deprivation (B = −.001, p = .718) were non‐significant.
For Supported Joint Engagement, Communication Deviance was again a significant predictor, B = −.32, SE = .1, β = −.21, t = −3.16, p = .002. The unadjusted model accounted for 4.2% of the variance in Supported Joint Engagement, R2 = .042, F(1228) = 10.01, p = .002. The overall adjusted model was again significant, F(4225) = 3.28, p = .012, and explained 5.5% of the observed variance (R2
= .055) with Communication Deviance predicting Supported Joint Engagement, B = −.29, SE = .11, β = −.18, t = −2.71, p = .007. Again, maternal reading ability (B = .02, p = .182) age (B = .01, p = .419), and Index of Multiple Deprivation (B = .003, p = .460) were non‐significant.
The same analysis for Coordinated Joint Engagement was repeated. Communication Deviance was again a significant predictor of this form of engagement, B = −.27, SE = .11, β = −.177, t = −2.69, p < .01. The unadjusted model accounted for 3.1% of the variance in Coordinated Joint Engagement, R2 = .031, F(1222) = 7.21, p = .008. The overall adjusted model was non‐significant, F(4219) = 2.03, p = .091, and explained 3.6% of the observed variance.
Finally, the analysis using Symbol‐infused Joint Engagement as the dependent variable yielded a positive unadjusted model, F(1228) = 19.55, p < .001, and explained 7.9% of the observed variance (R2
= .079) with higher Communication Deviance scores associated with less Symbol‐infused Joint Engagement between the mother and the child, B = −.56, β = −.28, t = −4.42, p < .001. The adjusted model was significant, F(4225) = 6.28, p < .001, explaining 10% of the observed variance (R2 = .10). Communication Deviance remained a significant predictor, B = −.48, SE = .13, β = −.24, t = −3.65, p < .001, and maternal reading ability (B = .03, p = .071), age (B = .01, p = .535), and Index of Multiple Deprivation (B = −.001, p = .886) being non‐significant. Overall, higher maternal Communication Deviance was consistently associated with reduced time in all forms of joint engagement.
8.3Maternal behaviors
For Scaffolding, the unadjusted model accounted for 17% of the variance, R2 = .17, F(1228) = 46.03, p < .001 and Communication Deviance was associated with less Scaffolding, B = −.81, SE = .12, β = −.41, t = −6.78, p < .001. The overall adjusted model was significant, F(4225) = 13.28, p < .001, and explained 19% of the observed variance (R2
= .19) with Communication Deviance predicting reduced Scaffolding, B = −.73, SE = .12, β = −.37, t = −5.94, p < .001 and with the confounders being non‐significant, maternal reading ability (B = .02, p = .316), age (B = .03, p = .125) and Index of Multiple Deprivation (B = −.002, p = .642).
Communication Deviance was also a significant predictor of reduced Symbol Highlighting, B = −.77, SE = .13, β = −.37, t = −6.04, p < .001. The unadjusted model accounted for 14% of the variance, R2 = .14, F(1228) = 36.42, p < .001. The overall adjusted model was also significant, F(4225) = 9.88, p < .001, and explained 15% of the observed variance (R2
= .15) with Communication Deviance again predicting reduced Symbol Highlighting, B = −.71, SE = .13, β = −.34, t = −5.301, p < .001 and with all the confounders being non‐significant, maternal reading ability (B = .02, p = .246), age (B = .01, p = .661), and Index of Multiple Deprivation (B = −.003, p = .609).
Finally, the unadjusted model with Following in on Child's Focus as the dependent variable, yielded significant findings with Communication Deviance being a negative predictor, B = −.76, SE = .12, β = −.398, t = −6.55, p < .001. The model accounted for 16% of the variance, R2 = .16, F(1228) = 42.93, p < .001. The adjusted model was significant, F(4225) = 12.55, p < .001, and explained 18% of the observed variance (R2
= .18) with Communication Deviance predicting reduced Following in on Child's Focus, B = −.69, SE = .12, β = −.36, t = −5.78, p < .001 and with all the confounders being non‐significant, maternal reading ability (B = .01, p = .488), age (B = .03, p = .058) and Index of Multiple Deprivation (B = −.001, p = .804). Therefore, indicating that maternal Communication Deviance strongly predicts reductions in Scaffolding, Symbol Highlighting, and Following in on the Child's Focus. Given the high intercorrelations among these behaviors, subsequent analyses also examined a composite maternal supportive behaviors factor derived via Principal Component Analysis (see Methods).
8.4Sustainability of shared topic and fluency and connectedness
The same analyses were repeated with Sustainability of Shared Topic and Fluency and Connectedness. Regarding Sustainability of Shared Topic, Communication Deviance was a significant predictor of reduced length of Sustainability of Shared Topic in the dyad, B = −.698, SE = .12, β = −.35, t = −5.62, p < .001. The unadjusted model accounted for 12% of the variance, R2 = .12, F(1228) = 31.596, p < .001. The overall adjusted model was significant, F(4225) = 9.46, p < .001, and explained 14% of the observed variance (R2
= .14) with Communication Deviance as a significant predictor, B = −.61, SE = .13, β = −.305, t = −4.75, p < .001 and with all the confounders being non‐significant, maternal reading ability (B = .03, p = .095), age (B = .002, p = .895) and Index of Multiple Deprivation (B = −.005, p = .320).
In relation to Fluency and Connectedness, given that this item was highly correlated with Sustainability of Shared Topic (r = .83), Communication Deviance was unsurprisingly a significant predictor, B = −.71, SE = .13, β = −.33, t = −5.27, p < .001. The unadjusted model accounted for 11% of the variance (R2 = .109), F(1228) = 27.78, p < .001. The overall adjusted model was significant, F(4225) = 7.94, p < .001, and explained 12% of the observed variance (R2
= .12) with Communication Deviance remaining a significant predictor, B = −.63, SE = .14, β = −.295, t = −4.53, p < .001 and with all the confounders being non‐significant, maternal reading ability (B = .02, p = .251), age (B = .02, p = .422), and Index of Multiple Deprivation (B = −.003, p = .663). Given the high correlation between Sustainability of Shared Topic and Fluency and Connectedness (r = .83), subsequent analyses also examined a composite Dyadic Attunement factor derived via Principal Component Analysis (see Statistical Analysis).
8.5Expressive Language
Finally, to examine whether maternal Communication Deviance at 32 weeks’ gestation predicted the child's Expressive Language at 30 months a subsequence analysis was run. The unadjusted model was non‐significant, F(1228) = 1.57, p = .211, and Communication Deviance did not predict Expressive Language in the child, B = −.14, SE = .11, β = −.08, t = −1.25, p = .211. Table 4 shows all linear regressions using Communication Deviance as the predicting variable.
8.6Testing mediation
As specified in the Methods, Principal Component Analysis was used to derive composite variables for maternal supportive behaviors and dyadic attunement. The supportive behaviors variables were strongly intercorrelated (all r > .54), with adequate sampling adequacy (KMO = .698) and a significant Bartlett's test (p < .001). A single factor was extracted, loading most strongly on supportive behaviors, and was used in subsequent mediation analyses. A similar procedure was applied to dyadic attunement variables (r = .83; KMO = .50; Bartlett's p < .001), yielding a single factor.
Separate mediation analyses were conducted using the PROCESS macro in SPSS. Communication Deviance was entered as the predictor, maternal supportive behaviors as the mediator, and joint engagement variables (Total Joint Engagement, Supported Joint Engagement, Coordinated Joint Engagement, and Symbol‐infused Joint Engagement), and the dyadic attunement factor as dependent variables. Maternal age, reading ability, and Index of Multiple Deprivation were included as covariates. Analyses were conducted separately for clearer interpretation as some of the scores on joint engagement inherently trade off against each other. The significance of indirect effects was assessed using 95% bias‐corrected bootstrapped confidence intervals (CI) based on 5000 bootstrap samples. In the analyses, Communication Deviance was again a significant predictor of maternal supportive behaviors factor (B = −.48, SE = .07, 95% CI [−.62, −.34]). The maternal supportive behaviors factor was then tested as a mediator of the relationship between Communication Deviance and the different outcomes. Firstly, the maternal supportive behaviors factor significantly predicted greater Total Joint Engagement (B = .80, SE = .07, 95% CI [.66, .93]). The indirect effect of Communication Deviance on Total Joint Engagement via the supportive behaviors factor was significant (B = −.38, BootSE = .07, 95% CI [−.55, −.26]), with the direct effect being non‐significant (B = .06, SE = .08, 95% CI [−.10, .21]), indicating full mediation; therefore, suggesting that the supportive behaviors factor fully mediates the association between maternal Communication Deviance and Total Joint Engagement. Secondly, the maternal supportive behaviors was a significant predictor of Supported Joint Engagement (B = .74, SE = .08, 95% CI [.58, .91]). The indirect effect of Communication Deviance on Supported Joint Engagement via the supportive behaviors factor was significant (B = −.36, BootSE = .08, 95% CI [−.52, −.23]), with the direct effect being non‐significant (B = .07, SE = .10, 95% CI [−.13, .27]), indicating full mediation; therefore, reductions in supportive behaviors accounted for the association between Communication Deviance and Supported Joint Engagement. Furthermore, the maternal supportive behaviors factor significantly predicted higher Coordinated Joint Engagement (B = .51, SE = .10, 95% CI [.32, .70]). The indirect effect of Communication Deviance on Coordinated Joint Engagement via supportive behaviors was significant (B = −.24, BootSE = .06, 95% CI [−.38, −.14]), with the direct effect being non‐significant (B = −.03, SE = .12, 95% CI [−.26, .20]), indicating full mediation; therefore, suggesting that supportive behaviors may be a mechanism through which Communication Deviance may influence Coordinated Joint Engagement. Additionally, the maternal supportive behaviors factor significantly predicted higher Symbol‐infused Joint Engagement (B = .95, SE = .10, 95% CI [.76, 1.14]). The indirect effect of Communication Deviance on Symbol‐infused Joint Engagement via the supportive behaviors factor was significant (B = −.46, BootSE = .09, 95% CI [−.66, −.31]), with the direct effect being non‐significant (B = −.02, SE = .12, 95% CI [−.26, .22]), indicating full mediation; therefore, suggesting that maternal supportive behaviors mediates the impact of Communication Deviance on Symbol‐infused Joint Engagement. Maternal supportive behaviors significantly predicted the higher dyadic attunement factor too (B = .75, SE = .06, 95% CI [.64, .86]). The indirect effect of Communication Deviance on dyadic attunement via supportive behaviors was significant (B = −.36, BootSE = .06, 95% CI [−.49, −.26]), with the direct effect being non‐significant (B = −.04, SE = .07, 95% CI [−.17, .09]), indicating full mediation. This suggests that maternal supportive behaviors may underpin the association between higher Communication Deviance in the mother and poorer conversational fluency and topic sustainability in the dyad. Lastly, maternal supportive behaviors significantly predicted higher Expressive Language in the child (B = .35, SE = .1, 95% CI [.15, .55]). The indirect effect of Communication Deviance on Expressive Language via supportive behaviors was significant (B = −.17, BootSE = .05, 95% CI [−.28, −.07]), with the direct (B = .09, SE = .12, 95% CI [−.15, .32]), and total effect (B = −.08, SE = .11, 95% CI [−.31, .14]), being both non‐significant indicating indirect‐only mediation. Though Communication Deviance did not directly predict Expressive Language, the indirect effect through supportive behaviors highlights the developmental importance of maternal support. Overall, maternal supportive behaviors was consistently a significant mediator for all forms of joint engagement, dyadic attunement, and Expressive Language. Standardized coefficients for all relationships are summarized in Figure 1.
9DISCUSSION
This study extends research on parental Communication Deviance by providing new empirical evidence regarding its developmental significance. Specifically, our findings contribute to understanding the role of parental Communication Deviance in developmental pathways and its potential implications for child outcomes. It was hypothesized that early caregiver communicative disturbances disrupt joint engagement in infancy, potentially cascading into broader socio‐cognitive difficulties. In a longitudinal cohort, higher maternal Communication Deviance at 32 weeks of pregnancy robustly predicted lower joint engagement at 30 months postpartum across domains (Total, Supported, Coordinated, and Symbol‐infused Joint Engagement states). These associations held after controlling for maternal age, reading ability, and Index of Multiple Deprivation. The findings underscore the enduring influence of maternal communicative style on mother‐child interaction quality. Given that Communication Deviance levels in this cohort were lower than typically reported in high‐risk samples, the restricted range has likely attenuated the effect sizes, so the observed associations may underestimate the strength of the underlying relationships.
9.1Communication Deviance as a predictor of joint engagement
Results suggest that Communication Deviance may compromise the foundational processes of dyadic interaction. Communication Deviance predicted not only less total time spent by the dyad on joint engagement but also reduced Supported and Coordinated Joint Engagement.
The association between higher maternal Communication Deviance and lower Symbol‐infused Joint Engagement suggests that communicative disturbances may reduce opportunities for symbolically enriched shared engagement between caregiver and child. Because Symbol‐infused Joint Engagement is a dyadic construct that incorporates both caregiver and child contributions, this finding may reflect differences in the extent to which symbolic communication is incorporated into shared interaction, rather than deficits in children's symbolic abilities per se. Effects on Symbol‐infused Joint Engagement suggest that higher maternal Communication Deviance may be associated with fewer opportunities for symbolically enriched shared engagement between caregiver and child. Because Symbol‐infused Joint Engagement reflects a dyadic process involving the incorporation of language, symbols, and pretend play into shared interaction, these findings may indicate that communicative disturbances reduce the extent to which symbolic communication becomes integrated into joint engagement episodes. These findings align with theoretical accounts suggesting that shared meaning‐making and symbolic reference rely heavily on the caregiver's capacity to structure and sustain affectively attuned, communicatively rich interactions (Adamson et al., 2004; Hobson, 2005).
9.2Mediating role of maternal supportive behaviors and attunement
Mediation analyses revealed that the effect of Communication Deviance on joint engagement was significantly mediated by maternal supportive behaviors. Communication Deviance was strongly associated with reductions in these behaviors, which in turn predicted less time spent by the dyad on joint engagement and poorer dyadic attunement. These findings are consistent with research finding that maternal supportive behaviors is a key process where the caregiver supports the development of attention, language, and emotion regulation (Brandes‐Aitken et al., 2020; Schatz et al., 2022). The direct effect of maternal Communication Deviance on joint engagement became non‐significant when supportive behaviors was included in the regression model, underscoring supportive behaviors as a possible key behavioral mechanism linking maternal Communication Deviance and early engagement. It is important to note that these constructs are conceptually distinct. Supportive behaviors reflect caregiver‐driven behaviors that structure and support the child's participation in interaction, whereas joint engagement captures the emergent dyadic state of shared attention and reciprocity. The strong association between these constructs is therefore theoretically expected, as supportive behaviors are thought to support and sustain joint engagement episodes rather than constitute them. Nevertheless, because both constructs were coded from the same interaction and derived from the same observational system, the mediation findings should be interpreted cautiously as evidence that supportive behaviors statistically account for the association between Communication Deviance and joint engagement, rather than as definitive evidence of a temporally distinct causal process. Moreover, given the substantial intercorrelations among items, the variance explained by individual models should be interpreted with caution, as these behaviors reflect overlapping aspects of broader interactional constructs.
Moreover, Communication Deviance was a robust predictor of dyadic attunement, operationalized as Sustainability of Shared Topic and Fluency and Connectedness. Findings suggest that maternal Communication Deviance may disrupt communicative clarity and the temporal and emotional synchrony underpinning high‐quality mother–child interactions (Hobson & Hobson, 2011). This synchrony of attunement is thought to form the basis of shared intentionality and reciprocal engagement, central tenets of sociocultural and enactivist models of development (Racine & Carpendale, 2007; Trevarthen & Aitken, 2001).
9.3Communication Deviance as an early environmental risk factor
Findings converge and extend prior research linking maternal Communication Deviance to child psychopathology, highlighting potential developmental pathways through which Communication Deviance may operate during the early vulnerability period. Extending prior longitudinal research linking maternal Communication Deviance to adolescent psychopathology (Tienari et al., 1985; Wahlberg et al., 1997), the present findings show associations with disrupted mother‐child interaction from as early as 30 months postpartum; therefore, supporting that Communication Deviance is not a downstream correlate of offspring vulnerability, or high‐risk status, but may function as an independent environmental risk factor shaping early developmental pathways. Importantly, Communication Deviance did not predict expressive language in the child at 30 months. It is possible that any direct effects of Communication Deviance on language development may emerge later, when more advanced and fluent language abilities are expected. The mean Expressive Language rating in the present sample indicated that most children were producing multiple intelligible words and simple word combinations but had not yet reached the highest levels of fluent sentence production captured by the scale. Furthermore, the JERI Expressive Language item primarily reflects the quantity and complexity of language production rather than the coherence, organization, clarity, or pragmatic appropriateness of language. Given that Communication Deviance is theoretically linked to difficulties in shared meaning and communicative coherence, its influence may be more readily detectable in later emerging aspects of language and communication than in early expressive language production alone.
9.4Implication and future directions
The findings suggest that Communication Deviance in the caregiver may be associated with disruption of early joint engagement and shared symbolic experiences with the child, which are critical for healthy socio‐cognitive development. These results point to potential targets for preventative work with families at risk or parents with diagnoses of severe and enduring mental illness. This is an area of work which is neglected (Davidsen et al., 2015; Harder et al., 2015). Training programs focused on enhancing maternal responsiveness and supportive behaviors during joint engagement may offer a viable pathway for mitigating the developmental impact of Communication Deviance, especially in families with known risk factors. Interventions could prioritize enhancing maternal responsiveness and supportive behaviors during early joint engagement episodes. Practical approaches could include structured coaching on following in on the child's focus, symbol highlighting, and supportive behaviors strategies, particularly for mothers with known psychosocial or mental health risk factors or diagnoses. Future research could use micro‐behavioral and sequential analytic approaches to examine how Communication Deviance influences the moment‐to‐moment dynamics of parent‐child interaction. For example, are interactions less mutually contingent, or do parents respond less contingently to children's communicative bids and symbolic play? In addition, future studies should also include postpartum assessments of Communication Deviance to examine whether communicative style remains stable once the child becomes an active partner, and to clarify the developmental timing of its effects. It would also be interesting in future research to antenatally examine maternal mental health and trauma history as potential covariates and predictors of Communication Deviance, given their likely influence on communicative style and early interactional processes. Another avenue will be to explore the extent to which relational disruptions mediate long‐term outcomes in social cognition, school readiness, and psychopathology. Given that Communication Deviance is not specific to schizophrenia‐spectrum conditions and appears across various mental health risks (Hirsch & Leff, 1971; Miklowitz et al., 1991), these findings support its conceptualization as a transdiagnostic dimensional marker of communicative disturbance. Further investigation into neurobiological correlates and genetic markers (Levy et al., 2010) of Communication Deviance, may clarify how genetic and environmental factors co‐produce developmental risk. Additionally, based on Lavoie and colleagues’ research into social cognition deficits in first‐degree relatives of individuals with schizophrenia (Lavoie et al., 2013, 2014), it may be valuable to examine associations between social cognition, mentalization, and Communication Deviance, and socio‐cognitive development as another potential early environmental risk factor in the emergence of mental illness. Future research should also examine these processes in culturally and linguistically diverse samples to determine whether these associations vary across different caregiving contexts. Finally, future research using micro‐temporal or sequential analytic approaches would be valuable in determining whether supportive behaviors temporally precede and predict transitions into joint engagement states, thereby clarifying the directionality of these effects.
9.5Limitations
Some limitations warrant consideration. Our sample was drawn from a relatively homogeneous sample in a specific geographic area, potentially limiting generalizability. As the sample was almost entirely White British, cultural and linguistic homogeneity may have shaped expression of Communication Deviance, supportive behaviors, and joint engagement, therefore further limiting generalizability to more diverse populations. Moreover, while Communication Deviance was measured prospectively and before the child's birth, all other constructs were assessed concurrently at 30 months, which limits causal inference.
Another limitation relates to the use of Five‐Minute Speech Samples as a methodology to elicit speech samples. Although this has been used previously to assess Communication Deviance, it is possible that the 5‐min sample methodology may underestimate the true variability and contextual expression of Communication Deviance in daily life. It would be interesting in future studies to use more conversational speech samples. Regarding the joint engagement rating approach, although the JERI provides detailed scoring guidance and behavioral examples (Adamson et al., 2020), observational ratings inevitably involve a degree of coder judgement when integrating interactional behaviors across an entire session into global ratings. Consequently, subtle differences in interpretation between coding teams cannot be entirely excluded despite formal training and good inter‐rater reliability.
Additionally, Communication Deviance was assessed antenatally to minimize reverse causality, as maternal communication could be influenced by the child postnatally. Whilst Communication Deviance shows good temporal stability (Nugter et al., 1997; Wahlberg et al., 2001), the expression may shift once the child becomes an active communicative partner. The absence of a post‐partum assessment further limits conclusions about developmental timing and warrants consideration as a limitation. Another limitation is that formal assessment of first‐language status was not available, although Communication Deviance coding focuses on communicative coherence rather than linguistic proficiency.
A consideration when interpreting the JERI findings concerns the coder training procedure. Although the first author received formal training at the JERI developers’ laboratory using gold‐standard videos and subsequently coded the WCHADS material with support from laboratory staff, the second coder was trained by the first author rather than directly by the JERI developers. The second coder nevertheless had access to coding materials used during the first author's training, and inter‐rater reliability was assessed through independent double‐coding. This procedure provides support for the consistency of the coding; however, the shared training pathway means that some influence arising from common training or coder expectations cannot be entirely excluded. This is relevant when interpreting the relatively high JERI scores observed in the present sample.
An additional consideration concerns the relatively high mean ratings observed across several JERI domains, particularly caregiver supportive behavior items and Sustainability of Shared Topic. Although the full range of ratings was used and inter‐rater reliability was good to excellent, the pattern of scores may suggest that some interactional behaviors were coded more liberally than intended in the original JERI framework. This possibility is difficult to evaluate directly because no cross‐site calibration with contemporary JERI coding teams was available. If present, such differences would likely influence the absolute ratings assigned to interactions rather than the pattern of associations observed within the current sample. Accordingly, findings should be interpreted primarily in terms of relative differences between dyads rather than as estimates of absolute levels of engagement or caregiver support.
A further consideration concerns the fact that maternal supportive behaviors and Joint Engagement were derived from observations of the same interaction. Although these constructs are related, they are conceptually distinct within the JERI framework. Supportive behaviors reflect caregiver actions that may facilitate engagement, whereas Joint Engagement is coded from the child's perspective and requires the child to acknowledge the caregiver as a social partner in the shared activity. Thus, caregiver efforts alone are insufficient to determine Joint Engagement ratings. Nevertheless, because both constructs were rated from the same interactional context, some degree of shared observational variance cannot be entirely excluded. The mediation findings should therefore be interpreted as identifying a plausible explanatory pathway rather than providing definitive evidence of a temporally independent causal mechanism.
10CONCLUSION
These findings enrich understanding of how subtle early environmental risk factors may shape socio‐cognitive development in the context of heightened familial vulnerability. Maternal Communication Deviance was directly associated with reduced joint engagement across multiple dimensions and with lower scaffolding, symbol highlighting, and dyadic attunement at 30 months. Mediation analyses further showed that these effects were explained by maternal supportive behaviors, demonstrating the role of these behaviors as a key mechanism linking Communication Deviance to early interactional quality.
Pre‐registration statement
This study was not pre‐registered. The research involved secondary data analysis of observational data collected as part of the Wirral Child Health and Development Study (WCHADS), a longitudinal cohort study. At the time of data collection, pre‐registration was not standard practice for secondary analyses of this kind. Further information about the WCHADS cohort can be found at: https://www.cataloguementalhealth.ac.uk/?content=study&studyid=WCHADS and https://www.liverpool.ac.uk/population‐health/research/groups/first‐steps/
FUNDING STATEMENT
This study was supported by the UK Medical Research Council (Grant G0400577 to Jonathan Hill and Helen Sharp) and the Fundação para a Ciência e a Tecnologia (Grant SFRH/BD/77,379/2011 to Paulo de Sousa). Additional support was provided by Wirral University Teaching Hospital NHS Foundation Trust, Cheshire and Wirral Partnership NHS Foundation Trust, Wirral Community NHS Foundation Trust, the NIHR Biomedical Research Centre for Mental Health at South London and Maudsley NHS Foundation Trust, and King's College London.
CONFLICT OF INTEREST STATEMENT
All authors declare that there are no competing interests.
References
Adamson , L. B. (2018 ). Communication development during infancy . Routledge.
Adamson , L. B. , & Bakeman , R. (2006 ). Development of displaced speech in early mother–child conversations . Child Development , 77 (1 ), 186 –200 .10.1111/j.1467-8624.2006.00864.x 16460533
Adamson , L. B. , Bakeman , R. , & Deckner , D. F. (2004 ). The development of symbol‐infused joint engagement . Child Development , 75 (4 ), 1171 –1187 .10.1111/j.1467-8624.2004.00732.x 15260871
Adamson , L. B. , Bakeman , R. , Deckner , D. F. , & Nelson , P. B. (2012 ). Rating parent–child interactions: Joint engagement, communication dynamics, and shared topics in autism, down syndrome, and typical development . Journal of Autism and Developmental Disorders , 42 , 2622 –2635 .10.1007/s10803-012-1520-1 PMC3445743 22466689
Adamson , L. B. , Bakeman , R. , Deckner , D. F. , & Nelson , P. B. (2014 ). From interactions to conversations: The development of joint engagement during early childhood . Child Development , 85 (3 ), 941 –955 .10.1111/cdev.12189 PMC4019689 24266591
Adamson , L. B. , Bakeman , R. , Deckner , D. F. , & Romski , M. (2009 ). Joint engagement and the emergence of language in children with autism and Down syndrome . Journal of Autism and Developmental Disorders , 39 , 84 –96 .10.1007/s10803-008-0601-7 PMC2640949 18581223
Adamson , L. B. , Bakeman , R. , & Suma , K. (2020 ). The joint engagement rating inventory . Department of Psychology, Georgia State University.
Adamson , L. B. , Bakeman , R. , Suma , K. , & Robins , D. L. (2019 ). An expanded view of joint attention: Skill, engagement, and language in typical development and autism . Child Development , 90 (1 ), e1 –e18 .10.1111/cdev.12973 PMC5891390 28991358
Bakeman , R. , & Adamson , L. B. (1984 ). Coordinating attention to people and objects in mother‐infant and peer‐infant interaction . Child Development , 55 (4 ), 1278 .6488956
Baron‐Cohen , S. (1997 ). Mindblindness: An essay on autism and theory of mind . MIT Press.
Brandes‐Aitken , A. , Braren , S. , Gandhi , J. , Perry , R. E. , Rowe‐Harriott , S. , & Blair , C. (2020 ). Joint attention partially mediates the longitudinal relation between attuned caregiving and executive functions for low‐income children . Developmental Psychology , 56 (10 ), 1829 .10.1037/dev0001089 PMC9275474 32700951
Butterworth , G. , & Jarrett , N. (1991 ). What minds have in common is space: Spatial mechanisms serving joint visual attention in infancy . British Journal of Developmental Psychology , 9 (1 ), 55 –72 .
Carpenter , M. , Nagell , K. , Tomasello , M. , Butterworth , G. , & Moore , C. (1998 ). Social cognition, joint attention, and communicative competence from 9 to 15 months of age . Monographs of the Society for Research in Child Development , 63 (4 ), i.9835078
Charman , T. , Baron‐Cohen , S. , Swettenham , J. , Baird , G. , Cox , A. , & Drew , A. (2000 ). Testing joint attention, imitation, and play as infancy precursors to language and theory of mind . Cognitive Development , 15 (4 ), 481 –498 .
Davidsen , K. A. , Harder , S. , MacBeth , A. , Lundy , J. M. , & Gumley , A. (2015 ). Mother‐infant interaction in schizophrenia: Transmitting risk or resilience? A systematic review of the literature . Social Psychiatry and Psychiatric Epidemiology , 50 (12 ), 1785 –1798 .10.1007/s00127-015-1127-x PMC4654744 26454698
de Sousa , P. , Sellwood , W. , Fien , K. , Sharp , H. , Pickles , A. , Hill , J. , Abbott , K. , Fisher , L. , & Bentall , R. P. (2019 ). Mapping early environment using communication deviance: A longitudinal study of maternal sensitivity toward 6‐month‐old children . Development and Psychopathology , 31 (4 ), 1501 –1511 .10.1017/S0954579418001189 30348239
de Sousa , P. , Varese , F. , Sellwood , W. , & Bentall , R. P. (2014 ). Parental communication and psychosis: A meta‐analysis . Schizophrenia Bulletin , 40 (4 ), 756 –768 .10.1093/schbul/sbt088 PMC4059429 23800431
Doane , J. A. , & Singer , M. (1977 ). Communication deviance scoring manual for use with consensus Rorschach. Unpublished manuscript, University of Rochester .
Docherty , N. M. (1993 ). Communication deviance, attention, and schizotypy in parents of schizophrenic patients . The Journal of Nervous and Mental Disease , 181 (12 ), 750 –756 .10.1097/00005053-199312000-00007 8254327
Goldstein , M. J. (1987 ). The UCLA high‐risk project . Schizophrenia Bulletin , 13 (3 ), 505 –514 .10.1093/schbul/13.3.505 3629204
Harder , S. , Davidsen , K. , MacBeth , A. , Lange , T. , Minnis , H. , Andersen , M. S. , Simonsen , E. , Lundy , J. M. , Nyström‐Hansen , M. , Trier , C. H. , Røhder , K. , & Gumley , A. (2015 ). Wellbeing and resilience: Mechanisms of transmission of health and risk in parents with complex mental health problems and their offspring–The WARM Study . BMC Psychiatry , 15 , 310 .10.1186/s12888-015-0692-6 PMC4674908 26654720
Hayes , A. F. (2013 ). Introduction to mediation, moderation, and conditional process analysis: A regression‐based approach . Guilford Press.
Hirsch , S. R. , & Leff , J. P. (1971 ). Parental abnormalities of verbal communication in the transmission of schizophrenia . Psychological Medicine , 1 (2 ), 118 –127 .10.1017/s0033291700000040 5148763
Hirsch , S. R. , & Leff , J. P. (1975 ). Abnormalities in parents of schizophrenics , (Vol. 22 ). Oxford University Press.
Hirsh‐Pasek , K. , Adamson , L. B. , Bakeman , R. , Owen , M. T. , Golinkoff , R. M. , Pace , A. , Yust , P. K. S. , & Suma , K. (2015 ). The contribution of early communication quality to low‐income children's language success . Psychological Science , 26 (7 ), 1071 –1083 .10.1177/0956797615581493 26048887
Hobson , P. , & Hobson , J. (2011 ). Joint attention or joint engagement? Insights from autism . In A. Seemann (Ed.),
Joint attention: New developments in psychology, philosophy of mind, and social neuroscience
, (pp. 115 –135 ), Boston Review.
Hobson , R. P. (2005 ). What puts the jointness into joint attention .
Joint attention: Communication and Other Minds
, 185 –204 , Clarendon Press/Oxford University Press.
Holdnack , H. A. (2001 ).
Wechsler test of adult reading: WTAR
. In (The Psychological Corporation, Ed.).
Hustedt , J. T. , & Raver , C. C. (2002 ). Scaffolding in low‐income mother‐child dyads: Relations with joint attention and dyadic reciprocity . International Journal of Behavioral Development , 26 (2 ), 113 –119 .
Koşkulu , S. , Küntay , A. C. , & Uzundag , B. A. (2021 ). Maternal behaviors mediate the relationship between socioeconomic status and joint attention . Journal of Applied Developmental Psychology , 75 , 101291.
Kymalainen , J. A. , Weisman , A. G. , Rosales , G. A. , & Armesto , J. C. (2006 ). Ethnicity, expressed emotion, and communication deviance in family members of patients with schizophrenia . The Journal of Nervous and Mental Disease , 194 (6 ), 391 –396 .10.1097/01.nmd.0000221171.42027.5a 16772854
Kymalainen , J. A. , & Weisman de Mamani , A. G. (2008 ). Expressed emotion, communication deviance, and culture in families of patients with schizophrenia: A review of the literature . Cultural Diversity & Ethnic Minority Psychology , 14 (2 ), 85 .10.1037/1099-9809.14.2.85 18426280
Lavoie , M.‐A. , Jackson , P. L. , Godmaire‐Duhaime , F. , Lacroix , J. B. , & Achim , A. M. (2014 ). Performance in multiple domains of social cognition in parents of patients with schizophrenia . Psychiatry Research , 220 (1‐2 ), 118 –124 .10.1016/j.psychres.2014.07.055 25216560
Lavoie , M.‐A. , Plana , I. , Bédard Lacroix , J. , Godmaire‐Duhaime , F. , Jackson , P. L. , & Achim , A. M. (2013 ). Social cognition in first‐degree relatives of people with schizophrenia: A meta‐analysis . Psychiatry Research , 209 (2 ), 129 –135 .10.1016/j.psychres.2012.11.037 23375626
Leeb , B. , Hahlweg , K. , Goldstein , M. J. , Feinstein , E. , Mueller , U. , Dose , M. , & Magana‐Amato , A. (1991 ). Cross‐national reliability, concurrent validity, and stability of a brief method for assessing expressed emotion . Psychiatry Research , 39 (1 ), 25 –31 .10.1016/0165-1781(91)90005-a 1771207
Levy , D. L. , Coleman , M. J. , Sung , H. , Ji , F. , Matthysse , S. , Mendell , N. R. , & Titone , D. (2010 ). The genetic basis of thought disorder and language and communication disturbances in schizophrenia . Journal of Neurolinguistics , 23 (3 ), 176 –192 .10.1016/j.jneuroling.2009.08.003 PMC2821112 20161689
Loy , M. , Masur , E. F. , & Olson , J. (2018 ). Developmental changes in infants’ and mothers’ pathways to achieving joint attention episodes . Infant Behavior and Development , 50 , 264 –273 .10.1016/j.infbeh.2018.02.001 29462746
Mendive , S. , Bornstein , M. H. , & Sebastián , C. (2013 ). The role of maternal attention‐directing strategies in 9‐month‐old infants attaining joint engagement . Infant Behavior and Development , 36 (1 ), 115 –123 .10.1016/j.infbeh.2012.10.002 PMC3618404 23276727
Miklowitz , D. J. , Velligan , D. I. , Goldstein , M. J. , Nuechterlein , K. H. , Gitlin , M. J. , Ranlett , G. , & Doane , J. A. (1991 ). Communication deviance in families of schizophrenic and manic patients . Journal of abnormal psychology , 100 (2 ), 163 –173 .10.1037//0021-843x.100.2.163 2040767
Miklowitz , D. J. , & Stackman , D. (1992 ). Communication deviance in families of schizophrenic and other psychiatric patients: Current state of the construct . Progress in Experimental Personality & Psychopathology Research , 15 , 1 –46 .1343105
Morales , M. , Mundy , P. , Crowson , M. , Neal , A. R. , & Delgado , C. (2005 ). Individual differences in infant attention skills, joint attention, and emotion regulation behaviour . International Journal of Behavioral Development , 29 (3 ), 259 –263 .
Morales , M. , Mundy , P. , Delgado , C. E. F. , Yale , M. , Messinger , D. , Neal , R. , & Schwartz , H. K. (2000 ). Responding to joint attention across the 6‐ through 24‐month age period and early language acquisition . Journal of Applied Developmental Psychology , 21 (3 ), 283 –298 .
Mundy , P. (2018 ). A review of joint attention and social‐cognitive brain systems in typical development and autism spectrum disorder . European Journal of Neuroscience , 47 (6 ), 497 –514 .10.1111/ejn.13720 28922520
Mundy , P. , Block , J. , Delgado , C. , Pomares , Y. , Van Hecke , A. V. , & Parlade , M. V. (2007 ). Individual differences and the development of joint attention in infancy . Child Development , 78 (3 ), 938 –954 .10.1111/j.1467-8624.2007.01042.x PMC2654237 17517014
Mundy , P. , & Gomes , A. (1998 ). Individual differences in joint attention skill development in the second year . Infant Behavior and Development , 21 (3 ), 469 –482 .
Mundy , P. , & Newell , L. (2007 ). Attention, joint attention, and social cognition . Current Directions in Psychological Science , 16 (5 ), 269 –274 .10.1111/j.1467-8721.2007.00518.x PMC2663908 19343102
Mundy , P. , & Sigman , M. (2015 ). Joint attention, social competence, and developmental psychopathology .
Developmental Psychopathology: Volume one: Theory and method
, Vol. 1 293 –332 , John Wiley & Sons.
Noble , M. , Wright , G. , Dibben , C. , Smith , G. A. N. , McLennan , D. , Anttila , C. , Barnes , H. , Mokhtar , C. , Noble , S. , Avenell , D. , Gardner , J. , Covizzi , I. , & Lloyd , M. (2004 ). The English Indices of Deprivation , 2004.
Nugter , M. A. , Dingemans , P. M. A. J. , Linszen , D. H. , Van Der Does , A. J. W. , & Gersons , B. P. R. (1997 ). Parental communication deviance: Its stability and the effect of family treatment in recent‐onset schizophrenia . Acta Psychiatrica Scandinavica , 95 (3 ), 199 –204 .10.1111/j.1600-0447.1997.tb09620.x 9111852
Nyström , P. , Thorup , E. , Bölte , S. , & Falck‐Ytter , T. (2019 ). Joint attention in infancy and the emergence of autism . Biological Psychiatry , 86 (8 ), 631 –638 .10.1016/j.biopsych.2019.05.006 31262432
ONS
. (2023 ).
2021 Census: Aggregate Data
, U. D. Service.
Racine , T. P. , & Carpendale , J. I. (2007 ). The role of shared practice in joint attention . British Journal of Developmental Psychology , 25 (1 ), 3 –25 .
Roisko , R. , Wahlberg , K. E. , Miettunen , J. , & Tienari , P. (2014 ). Association of parental Communication Deviance with offspring's psychiatric and thought disorders. A systematic review and meta‐analysis . European Psychiatry , 29 (1 ), 20 –31 .10.1016/j.eurpsy.2013.05.002 23849397
Salter , G. , & Carpenter , M. (2025 ). The developmental origins of joint attention: Infants' early joint attention bids . Infancy , 30 (2 ), e70012.10.1111/infa.70012 PMC11947298 40140992
Scaife , M. , & Bruner , J. S. (1975 ). The capacity for joint visual attention in the infant . Nature , 253 (5489 ), 265 –266 .10.1038/253265a0 1113842
Schatz , J. L. , Suarez‐Rivera , C. , Kaplan , B. E. , & Tamis‐LeMonda , C. S. (2022 ). Infants’ object interactions are long and complex during everyday joint engagement . Developmental Science , 25 (4 ), e13239.10.1111/desc.13239 PMC10184133 35150058
Sharp , H. , Pickles , A. , Meaney , M. , Marshall , K. , Tibu , F. , & Hill , J. (2012 ).
Frequency of infant stroking reported by mothers moderates the effect of prenatal depression on infant behavioural and physiological outcomes
.10.1371/journal.pone.0045446 PMC3473033 23091594
Singer , M. T. (1965a ). Thought disorder and family relations of schizophrenics: III. Methodology using projective techniques . Archives of General Psychiatry , 12 (2 ), 187 .10.1001/archpsyc.1965.01720320075009 14237629
Singer , M. T. (1965b ). Thought disorder and family relations of schizophrenics: IV. Results and implications . Archives of General Psychiatry , 12 (2 ), 201 .10.1001/archpsyc.1965.01720320089010 14237630
Singer , M. T. , & Wynne , L. C. (1966 ). Principles for scoring communication defects and deviances in parents of schizophrenics: Rorschach and TAT scoring manuals . Psychiatry , 29 (3 ), 260 –288 .5969538
Singer , M. T. , Wynne , L. C. , & Toohey , M. L. (1978 ). Communication disorders and the families of schizophrenics. In L. C. Wynne , R. L. Cromwell , & S. Matthysse (Eds.), The nature of schizophrenia: New approaches to research and treatment , (pp. 499 –511 ). John Wiley & Sons.
Strauss , E. , Sherman , E. M. , & Spreen , O. (2006 ). A compendium of neuropsychological tests: Administration, norms, and commentary . Oxford University Press.
Suarez‐Rivera , C. , Schatz , J. L. , Herzberg , O. , & Tamis‐LeMonda , C. S. (2022 ). Joint engagement in the home environment is frequent, multimodal, timely, and structured . Infancy , 27 (2 ), 232 –254 .10.1111/infa.12446 34990043
Tienari , P. , Sorri , A. , Lahti , I. , Naarala , M. , Wahlberg , K. E. , Rönkkö , T. , Pohjola , J. , & Moring , J. (1985 ). The Finnish adoptive family study of schizophrenia . The Yale Journal of Biology and Medicine , 58 (3 ), 227 –237 .PMC2589875 4049906
Tienari , P. , & Wahlberg , K.‐E. (2008 ). Family environment and psychosis. In C. Morgan , K. McKenzie , & P. Fearon , (Eds.), Society and Psychosis , (pp. 112 –126 , 1 ed.). Cambridge University Press.
Tomasello , M. (2014 ). Joint attention as social cognition , (pp. 103 –130 ). Joint attention. Psychology Press.
Tomasello , M. , & Farrar , M. J. (1986 ). Joint attention and early language . Child Development ,57(6), 1454 –1463 .3802971
Trevarthen , C. , & Aitken , K. J. (2001 ). Infant intersubjectivity: Research, theory, and clinical applications . The Journal of Child Psychology and Psychiatry and Allied Disciplines , 42 (1 ), 3 –48 .11205623
Velligan , D. , Christensen , A. , Goldstein , M. J. , & Margolin , G. (1988 ). Parental communication deviance: Its relationship to parent, child, and family system variables . Psychiatry Research , 26 (3 ), 313 –325 .10.1016/0165-1781(88)90126-6 3222396
Velligan , D. I. (1985 ).
Communication deviance coding manual for coding family confrontations University of California
.
Velligan , D. I. , Funderburg , L. G. , Giesecke , S. L. , & Miller , A. L. (1995 ). Longitudinal analysis of communication deviance in the families of schizophrenic patients . Psychiatry , 58 (1 ), 6 –19 .10.1080/00332747.1995.11024707 7792324
Velligan , D. I. , Goldstein , M. J. , Nuechterlein , K. H. , Miklowitz , D. J. , & Ranlett , G. (1990 ). Can communication deviance be measured in a family problem‐solving interaction?
Family Process , 29 (2 ), 213 –226 .10.1111/j.1545-5300.1990.00213.x 2373216
Velligan , D. I. , Miller , A. L. , Eckert , S. L. , Funderburg , L. G. , True , J. E. , Mahurin , R. K. , Diamond , P. , & Hazelton , B. C. (1996 ). The relationship between parental communication deviance and relapse in schizophrenic patients in the 1‐year period after hospital discharge: A pilot study . The Journal of Nervous and Mental Disease , 184 (8 ), 490 –496 .10.1097/00005053-199608000-00006 8752078
Vygotsky , L. S. , & Cole , M. (1978 ). Mind in society: Development of higher psychological processes . Harvard university press.
Wahlberg , K.‐E. , Wynne , L. C. , Keskitalo , P. , Nieminen , P. , Moring , J. , Läksy , K. , Sorri , A. , Koistinen , P. , Tarvainen , T. , Miettunen , J. , & Tienari , P. (2001 ). Long‐term stability of communication deviance . Journal of Abnormal Psychology , 110 (3 ), 443 –448 .10.1037//0021-843x.110.3.443 11502087
Wahlberg , K. E. , Wynne , L. C. , Oja , H. , Keskitalo , P. , Anais‐Tanner , H. , Koistinen , P. , Tarvainen , T. , Hakko , H. , Lahti , I. , Moring , J. , Naarala , M. , Sorri , A. , & Tienari , P. (2000 ). Thought Disorder Index of Finnish adoptees and communication deviance of their adoptive parents . Psychological Medicine , 30 (1 ), 127 –136 .10.1017/s0033291799001415 10722183
Wahlberg , K. E. , Wynne , L. C. , Oja , H. , Keskitalo , P. , Pykalainen , L. , Lahti , I. , Moring , J. , Naarala , M. , Sorri , A. , Seitamaa , M. , Laksy , K. , Kolassa , J. , & Tienari , P. (1997 ). Gene‐environment interaction in vulnerability to schizophrenia: Findings from the Finnish adoptive family study of schizophrenia . American Journal of Psychiatry , 154 (3 ), 355 –362 .10.1176/ajp.154.3.355 9054783
Wynne , L. , Singer , M. T. , Bartko , J. , & Toohey , M. (1977 ). Schizophrenics and their families: Research on parental communication. In J. M. Tanner (Ed.), Developments in psychiatric research , (pp. 254 –286 ). Hodder & Stoughton.
Wynne , L. C. (1963a ). Thought disorder and family relations of schizophrenics: I. A research strategy . Archives of General Psychiatry , 9 (3 ), 191 .10.1001/archpsyc.1963.01720150001001 14002138
Wynne , L. C. (1963b ). Thought disorder and family relations of schizophrenics: II. A classification of forms of thinking . Archives of General Psychiatry , 9 (3 ), 199 .10.1001/archpsyc.1963.01720150009002 14002139
Wynne , L. C. (1984 ). The epigenesis of relational systems: A model for understanding family development . Family Process , 23 (3 ), 297 –318 .10.1111/j.1545-5300.1984.00297.x 6479297
Yoon A , S. , Kelso A , G. , Lock , A. , & Lyons‐Ruth , K. (2014 ). Mother–infant joint attention and sharing: Relations to disorganized attachment and maternal disrupted communication . The Journal of Genetic Psychology , 175 (6 ), 494 –510 .10.1080/00221325.2014.964170 25412198
Come citare questo studio
Sousa P, Hartley R, Sellwood W, Grace N, Sharp H, Hill J, Bentall RP. (2027). Tracing the impact of maternal communication deviance on joint engagement: The mediating role of maternal supportive behaviors.. Infant mental health journal. DOI: 10.1002/imhj.70126
Consulta la pubblicazione originale ↗