The transition to the third year of life represents a pivotal turning point for language development. Children's vocabulary tends to experience a significant increase at the end of the second year of life and many children start to combine different words (Rantalainen et al., 2021). This phase is often accompanied by a substantial increase in predicates production. Nonetheless, this developmental milestone contributes to individual differences that are only partly explained. In the present study, we explored how early forms of attention at 12 months (i.e., exogenous attention allocation and joint attention initiation) might modulate individual differences in 24 months language development in typically developing children.

Children's vocabulary shows a dramatic increase in the second year of life, as infants go from learning their first words around their first birthday to an average of 300 words by their second birthday, with significant individual variability (Frank et al., 2017). Moreover, a common pattern can be found in the qualitative composition of this first vocabulary. Namely, in the very first phases of language development a larger proportion of words learned are related to first social interaction and everyday life routines whereas as the vocabulary grows a larger proportion of words learned are object names and subsequently, as combinatory abilities emerge, action (such as verbs and adjectives) and function words tend to augment. This pattern has been reported in different languages (Caselli et al., 1999; Choi and Gopnik, 1995; Maital et al., 2000; Tardif et al., 1999). A restricted percentage of predicates has also been proposed as an early marker of language delay in Italian speaking children (Camaioni and Longobardi, 1995; D'Odorico et al., 2001) and the number of verbs that children produce at age two has been shown to be a better predictor of later grammatical skills than the number of nouns (Hadley et al., 2016). Nevertheless, limited research as focused on predictors of vocabulary composition or on potential mechanisms of predicates learning (Horvath et al., 2021; Horvath and Arunachalam, 2019).

Children learn language in environments that are both rich with information and yet ambiguous, where every label can have multiple possible referents (Raz et al., 2019). We know that the parents' ability to provide labels contingently with the child's attentional focus to a specific object can support learning (Suanda et al., 2016; Yurovsky et al., 2013) and early interventions focused on parent-infant interactions capitalize on this core assumption (Buschmann et al., 2015; Fong et al., 2012; Roberts and Kaiser, 2015). Consistent with expectations, numerous studies have documented links between children's attentional abilities and their language development. This association appears to hold true across various attentional skills (Salley et al., 2013).

Attention allocation is guided by endogenous mechanisms entailing the ability to actively maintaining focus on a stimulus (sustained attention) and exogenous mechanisms leading to shifts to environmental stimuli or distractors (distractibility) (Colombo, 2001; Colombo and Cheatham, 2006). Endogenous attention allocation develops early in the first year of life but continues to improve in the second half of the first year, enhancing visual engagement, and attention allocation control (Lansink et al., 2000; Reynolds and Romano, 2016).

Higher levels of sustained attention in 9-month-olds have been linked with better language outcomes 6 months later (Yu et al., 2019). Consistently, 12-month-old infants showing frequent exogenous attention allocation to external stimuli have been described as achieving less optimal language development 6 months later (Testa et al., 2023). Notably, prior investigations in this domain have been confined to controlled laboratory environments. Consequently, the influence of early exogenous attention allocation within naturalistic, interactive social contexts involving infants and their caregivers remains largely unexplored. To bridge this knowledge gap, future research should elucidate how this facet of attention relates to subsequent language development when assessed in ecologically valid settings.

Furthermore, between 9 and 12 months, infants show significant advancements in social attention development, including joint attention engagement i.e., the ability to share their attentional focus on external stimuli with the caregivers (Bradley, 2023). In this developmental window infants are sensitive to other's gaze direction and show gradual increases in gaze- and point-following, responding to joint attention cues (Tang et al., 2024) and further expanding their ability to initiate joint attention, using their own gaze and other social cues to intentionally share attention with others (Stephenson et al., 2021). Links between joint attention and language development have been described, too (Tomasello and Todd, 1983) and social-pragmatic cues remain important sources of information about word meanings even for adults. For instance, infants learn to use the adult's gaze to acquire new words using this cue to develop word-to-world mappings (Hollich et al., 2000). Coherently, significant associations have been reported between language development and child ability to both respond to joint-attention (Delgado et al., 2002; Morales et al., 2000) and initiate joint attention by sending specific bids in the shape of gaze triangulation and early communicative behaviors (e.g., pointing; Bavin et al., 2008; Brooks and Meltzoff, 2008; McGillion et al., 2017). As previously highlighted, evidence of a link between early life exogenous attention allocation and later cognitive and language development is present in the literature (Colombo and Mitchell, 2009; Hendry et al., 2019; Wass, 2015). Although multiple studies have shown that children are active agents in their environment and their abilities and motivations greatly impact learning mechanisms (Thelen, 1995), previous research has only partly focused on the role played by children's initiation of joint attention in influencing later language outcomes.

Underlying causes of individual variability in language development are still partly unknown, as many studies on language development predictors only explain a small portion of the difference in vocabulary size (Reilly et al., 2018). Moreover, studies often investigate the main effect of potential predictors and don't account for interaction effects between predictors; conversely recent studies have highlighted the presence of complex and cumulative effects of risk factors in later development (Eadie et al., 2022). For instance, the presence of multiple risk factors has been shown to increase the probability of language delays in the general population (Hayiou-Thomas et al., 2021). However, limited research has investigated potential interactive effects between attention related measures and later language development. This dearth of evidence is regrettable considering that language development (as development in general) is shaped by diverse, interconnected, interdependent mechanisms, and its nature should prompt us to embrace complexity (D'souza et al., 2017). These dynamic interconnected effects are even more relevant in the case of attentional risk and protective factors as attentional mechanisms can both support one another (e.g., Fisher, 2019) or act in competition (e.g., Lee and Schumacher, 2024). For instance, limited studies have investigated the potential moderating effect of joint attention on the link between early individual cognitive differences and later language development (e.g., Canfield and Saudino, 2016; Salley et al., 2013) with contrasting results. For instance, recent studies have shown that infant sustained attention in the context of joint attention, but not joint attention itself, seem to be a stronger predictor of later vocabulary size, suggesting that joint attention may not just co-occur with infant sustained attention but may play a key supportive role (Yu et al., 2019). As such good joint attention skills may increase the protective effect of sustained attention or reduce the reported negative effect of frequent exogenous attention allocation on later development. On the other hand, research has shown that individual susceptibility to environmental cues in early life (i.e., environmental sensitivity—Bahrick et al., 2018; Moyano et al., 2023) can make individuals more malleable to both its positive and negative influences, essentially making them more responsive to both risk and protective factors (Greven et al., 2019). As such exogenous attention allocation—framed as an index of environmental sensitivity i.e., heightened openness to the environment—may act as a mediator in word learning enhancing or reducing the reported positive effects of joint attention engagement.

Thus, the goals of the present study were: (a) to investigate associations between 12-month exogenous attention allocation (to social auditory stimuli), joint attention initiation—measured in real-life social settings—and language development at 24 months assessing both vocabulary dimension and composition; (b) to explore the presence of interactive/modulating effects between 12-month infants' exogenous attention allocation and initiation of joint attention in predicting later language development.

The present study aimed at exploring how early forms of attention at 12 months (i.e., exogenous attention allocation and joint attention initiation) might modulate individual differences in 24 months language development (i.e., vocabulary) in typically developing children.

Higher vocabulary size at 24 months—and specifically more advanced vocabulary composition as signaled by the percentage of predicates—was negatively associated with the amount of time infants' exogenous attention allocation was attracted by external social auditory stimuli during the interaction with their primary caregiver. These findings are consistent with studies previously conducted in more experimental settings (Salley et al., 2013; Testa et al., 2023) highlighting negative associations between distractibility and language development. Moreover, the results can add to the current literature on the importance of early attention allocation in impacting later cognitive development at different levels (Hendry et al., 2019). Although orienting to new external stimuli is certainly an important skill, supporting environmental exploration and learning, it can be hypothesized that the role of this ability could be modulated by different contexts (e.g., interactive vs. individual context) and in relation to different developmental outcomes (e.g., socio-cognitive vs. communicative development). Moreover, if we frame exogenous attention allocation as an individual feature potentially related to the concept of environmental sensitivity (Greven et al., 2019), the present results could suggest that higher levels of sensitivity to the environment may expose infants to higher distractibility and be linked to less optimal language outcomes already at 24 months.

Furthermore, interactive effects between exogenous attention allocation and joint attention initiation emerged. On one hand, infants that exhibited less joint attention initiation bids showed a stronger association between such distractibility and language development compared to counterparts that produced higher levels of joint attention initiation. As such, joint attention initiation ability appears to emerge as a potential protective factor in the previously presented developmental association between attention and language. A potential explaining hypothesis for this finding may be that the child's ability to actively bring the adults' attention to a distracting event (i.e., by initiating joint attention episodes) could attenuate the negative effects of distractibility by integrating the distracting event in the interaction. If confirmed, these findings could stress the importance of assessing more complex and interactive effects in early development cascades and the relevant active role of the child on his/her own cognitive development (Thelen, 1995). On the other hand, only infants with higher exogenous attention allocation tended to experience positive associations between joint attention initiation and later language. A potential explaining hypothesis for this finding, could be that being more open and responsive to the environment can increase the positive effect of joint attention initiation episodes—for instance creating more opportunities for concurrent maternal language input—on later early vocabulary acquisition. If confirmed this finding could support evidence on the moderating effects of environmental sensitivity on other risk and protective factors (Greven et al., 2019).

The present study has limitations. Firstly, the sample is relatively small and homogeneous, recruited in an urban middle to high-income area, thus the findings require further replication in larger and more diverse cohorts, ideally including also families from different socio-economic and cultural backgrounds. Nonetheless, consistent with the available literature in Italian speaking children (Caselli et al., 1999; D'Odorico et al., 2001), bigger vocabulary size at 24 months featured a larger predominance of predicates. This suggests that—despite the sample size of our study might be relatively small—the representativeness of language data is preserved. Second, despite its longitudinal nature, the present study is observational and the relationships among variables can only be considered in in terms of associations rather than predictions. Furthermore, the study is focused on the link between early attentional abilities and language; we acknowledge the complex and multi-faceted mechanisms underlying communicative and linguistic development and that other factors may intervene and further contribute to the described association framework. Finally, whereas the 12-month measures have been collected observing mother-infant interactive behaviors, vocabulary was tested via parental questionnaire. The CDI is a widely used tool in language development research; notwithstanding, a combination of parent-reported and direct testing would have allowed a broader assessment of language abilities.

The present study aimed at testing early attention abilities in an interactive social context and assessing not only direct links but also interaction effects between these early abilities and later language. The results of the present study could thus be relevant in expanding our knowledge of early developmental cascades in language acquisition and in supporting the current literature on potential targets for screening and intervention.