Screen exposure and reduced cognitive abilities, language development, and tactile exploration in very preterm infants aged 12 months—A pilot study

Children are increasingly exposed to digital media from an early age. Screen exposure has been linked to poorer parent–child interactions, and less optimal language, cognitive, motor, and socio-emotional development. However, few studies have explored these relationships specifically in children born preterm. Sixty-four very preterm infants (VPT; <32 weeks gestation) were assessed at 12-months corrected age. Data on media use within families and infant screen exposure were measured using the Family Media Use questionnaire. Psychomotor development, socio-emotional functioning, and tactile exploration abilities were evaluated using the Bayley Scales of Infant Development III (BSID III) and experimental tasks. At 12 months of age, most parents (72%) reported using screens in the presence of their child, with an average daily duration of 1 h. Furthermore, 39% of VPT infants were directly exposed to screens, averaging 4 h per week. No difference was observed in BSID III scores between infants exposed to screens and those who were not. Furthermore, across the entire cohort, no significant association was found between screen exposure and developmental outcomes. However, among infants exposed to screens, greater direct screen exposure was negatively associated with cognitive and language scores (ρ = −0.34, p = 0.048; ρ = −0.56, p = 0.002, respectively) as well as to tactile exploration (ρ = −0.39, p = 0.02), but not to prosocial behavior or joint attention. These associations remained significant after controlling gestational age at birth and parents' socio-economic status. Early direct screen exposure in VPT infants is associated with lower language and cognitive performance, and reduced tactile exploration, highlighting potential developmental implications even within the first year. No associations were observed in socio-emotional skills suggesting that these aspects may evolve differently or become apparent later in development. The effects of screen exposure on the development of infants born VPT have not been extensively studied, and further research is needed to support families in offering alternatives to screen use.

Infants and young children today are growing up in deep digital environments, beginning to interact with various media earlier than ever, and using them increasingly often. Screens are now omnipresent in their daily lives: television often remains on in the background, and parents' smartphones or computers are readily accessible, frequently displaying content that is not age appropriate. This increasing presence raises questions about how digital media may affect early development.

Many studies have examined digital media use among children and adolescents in Europe and worldwide. In the United States, most children use a mobile device daily by the age of 2, with screen time divided almost equally between television and mobile devices; moreover, most children are exposed to television before their first birthday (Kabali et al., 2015). Similar patterns have been observed in Europe. In France, data from the Elfe longitudinal study showed that television was the most commonly used screen among 2-year-olds, with 87% watching it and 68% doing so daily and spending an average of almost 7 h per week on it (Berthomier and Octobre, 2019). In Switzerland, Gillioz et al. (2022), reported that 88% of children had been exposed to screens before the age of 3, spending an average of 5 h per week in front of a screen. The study also found that children's screen exposure was significantly associated with their parents' screen use (Gillioz et al., 2022). More recently, the COVID-19 pandemic further increased children's screen time, as lockdowns and restrictions led many families to rely more heavily on digital media (Ponti, 2023).

In the first year of life infants depend on face-to-face interactions, eye contact, voice, and shared attention with their caregivers to support the development of communication and early language skills. Parental sensitivity to the infant's signals plays a key role in fostering secure attachment and nurturing socio-emotional abilities, which serve as a foundation for later exploration and learning. A growing body of research highlights how screen-based activities can interfere with intra-family social relationships and are often associated to decreased verbal and non-verbal parent-child interactions (Christakis et al., 2009; Kirkorian et al., 2009; Radesky et al., 2024). This raises concerns about a potential impact on broader intra-family social dynamics, especially during early childhood, a sensitive period for the development of brain systems underlying socio-emotional skills (e.g., cortico-limbic circuits) which are highly responsive to environmental input and caregiving experiences (Leisman et al., 2025).

Furthermore, according to the displacement theory (Neuman, 1988) the more time children spend in front of screens, the less time they have for developmentally enriching activities such as interactive play, physical exploration, and social engagement. In line with this theory, screen use may reduce both the quantity and quality of these experiences. Consistent with this hypothesis, a growing body of research has linked early screen exposure to poorer outcomes across several domains (Sticca et al., 2025), including motor skills (Lin et al., 2015), language (Collet et al., 2019; Vandormael et al., 2019; Zimmerman et al., 2007), cognition (Anderson et al., 2017; Lin et al., 2015; Schmidt et al., 2009; Tomopoulos et al., 2010), and social-emotional development (Chonchaiya et al., 2015; Domingues-Montanari, 2017; Hinkley et al., 2018; Lin et al., 2020; Pagani et al., 2013). In addition, Takahashi et al. (2023) found that increased screen time at 1 year is associated with delays in communication and problem-solving abilities by ages two and four. More recently, evidence has also emerged suggesting that screen exposure during the first year of life may be associated with higher autism screening scores (Chakhunashvili et al., 2026; Chakhunashvili and Chakhunashvili, 2025), although the observational nature of these findings precludes causal inference. Together, these findings highlight the need to better understand how screen exposure during infancy may influence children's developmental trajectories.

Such concerns are particularly relevant for children born very preterm (VPT), among whom approximately 30% to 50% experience a range of persistent neurodevelopmental challenges across motor skills, cognition, language, socio-emotional, and behavioral regulation (Pierrat et al., 2021). Although these challenges are partly related to atypical brain maturation, reflected in altered functional organization and structural differences detectable on term-equivalent MRI (Dubois et al., 2008; Inder et al., 2005, 2023), biological vulnerability alone does not fully explain later developmental outcomes. Environmental and social factors also play a major role, and growing evidence suggests that they can shape, and potentially mitigate (Gillioz et al., 2022) the developmental challenges faced by this vulnerable population (Bills et al., 2021). Behavioral and emotional regulation is an area of vulnerability in VPT children, with difficulties emerging early in life. Compared to term-born pears, VPT infants show greater emotional reactivity (Langerock et al., 2013) and, by 2 years of age, are described as less compliant and show poorer attention, social imitation, motivation, and persistence (Spittle et al., 2009). Language is another domain at risk, as its development depends on both intact neural substrates and rich early social interactions. Consistent with this, unfavorable early environments have been associated with poorer cognitive, language, executive, and behavioral outcomes in both preterm and full-term children (Gui et al., 2019; Mangin et al., 2017; Pittet-Metrailler et al., 2019; Ritchie et al., 2015). Despite this well-established vulnerability, the relationship between early screen exposure and developmental outcomes in VPT children remains largely unexplored. To our knowledge, only one study has examined screen use in this population: Vohr et al. (2021) reported that greater screen time at 6–7 years of age was associated with lower cognitive performance, poorer executive functioning, and more behavioral difficulties in children born extremely preterm. Given the increasing use of digital media from infancy, understanding how early screen exposure relates to development during this highly sensitive period is therefore particularly important in VPT infants.

In this study, we hypothesized that screen exposure would be associated with poorer performance in several fundamental domains of child development, including cognitive, motor, and language development, as well as socio-emotional skills—such as prosocial behavior and joint attention—and tactile exploration abilities.

Study population

Upon hospital discharge, all VPT infants (gestational age at birth < 32 weeks) are invited to a regular follow-up program at the Child Development Center of the hospital. Among those, infants delivered at the Geneva University Hospitals, between January 2021 and April 2023 were invited to take part in this study. For this research, exclusion criteria involved major congenital malformations (defined as genetic anomaly, syndrome, or malformation of a major organ system) and insufficient parental proficiency in French to adequately understand the study questionnaires. A total of 147 children were born VPT and survived. Of these, seven were excluded due to the absence of a French-speaking parent, 23 because the family had relocated, four due to severe brain injury, and one due to a genetic syndrome, leaving 112 infants eligible for recruitment. All information about the study was first provided to the parents by phone prior to their appointment at the Child Development Center, and then again in person during the consultation. Parents who agreed to participate provided written informed consent, completed the questionnaires, and their children completed the experimental tasks during the consultation.

Of the 112 eligible participants, 21 parents refused to participate, 15 could not be reached, and 12 did not complete the Family Media Use questionnaire. Thus, a total of 64 infants were included. All of them underwent an assessment of cognitive, language, and motor development, and completed both the identification of prosocial behavior task and the joint attention task (more details in next sections). In addition, 60 children participated in the tactile exploration task (Figure 1).

Perinatal characteristics were extracted from the child's medical records, including sex, degree of prematurity, birth weight, and neonatal complications such as bronchopulmonary dysplasia (BPD), sepsis, and mild to moderate brain injuries. Sociodemographic data were collected following the method described by Largo et al. (1989), based on the mother's level of education and the father's occupation. The resulting scores ranged from 2 to 12, with lower scores indicating higher socio-economic status (SES).

Data collection for this study was approved by the Cantonal Ethics Commission, in accordance with the current version of the World Medical Association's Declaration of Helsinki and the principles and procedures of integrity in scientific research involving human subjects.

Assessment

Family media use questionnaire

Exposure to electronic media at home was assessed with The Family Media Use questionnaire (Gillioz et al., 2022), which was administered in French and included:

• General information about the household, such as type and number of all electronic media present in the household (TV, smartphones, videos/DVDs, tablets, computer...).

• Children's direct screen use, including screen use during meals. Screen time refers to the time spent watching television or using devices (e.g., tablets, smartphones). Parents were asked to estimate the average amount of time their toddler spent using screens on weekdays and weekends, allowing the calculation of the child's total direct screen exposure over a typical week (expressed in hours per week).

Psychomotor assessment

The developmental assessment was conducted by child psychologists, psychomotor therapists, and physiotherapists using the Bayley Scales of Infant Development III (Bayley, 2006), an instrument designed to assess motor, cognitive, and language development in infants and toddlers. These assessments were part of the regular follow-up of children born VPT in Switzerland.

Socio-emotional assessment

Ages & stages questionnaires

Social-Emotional (ASQ ®:SE-2)1: This questionnaire, completed by parents, is a screening tool to identify infants and young children who may need further assessment of their social and emotional development. Each part of the questionnaire assesses self-regulation, compliance, communication, adaptive behaviors, autonomy, affect, and interaction with people. A total score above 48 indicates poor social-emotional development. For the present analysis, the total score was used as the main variable.

In addition, children were presented with experimental tasks designed to assess their socio-emotional skills:

Identification of pro-social behavior (2 tasks)

The first task consisted in the repeated presentation of a brief scenario in which one stuffed animal plays with a ball and accidentally drops it near another stuffed animal [inspired of Hamlin (2013)]. The second animal either returned the ball (pro-social behavior: “helper”) or kept it for itself (anti-social behavior: “hinderer”). After the demonstration, the experimenter placed both puppets (the pro-social and the anti-social character) within the child's reach. The child's behavior was then analyzed by observing which puppet he or she chooses to grasp first: the pro-social or the anti-social puppet. A score of 1 was assigned when the child intentionally chose the helper puppet first, and a score of 0 otherwise. Puppet color and side of presentation were counterbalanced across participants.

The second task was divided into two parts (Greitemeyer, 2022). In the first part, the child observed an interaction between two experimenters. One experimenter built a tower using three small blocks placed on the table. They then attempted to build a second tower but was unable to reach the blocks, which were placed too far away. The second experimenter helped by handling him the blocks, one at a time. In the second part of the task, the same scenario is repeated, but without the presence of the second experimenter. This time, the child was given the opportunity to help by handling over the three blocks to the experimenter who was trying to build the tower. The aim was to observe whether the child engaged in altruistic behavior after having witnessed the model of helpful behavior. The number of blocks the child gave to the experimenter served as the score for this task. The total prosocial behavior score was obtained by adding the scores from the two tasks. One point was awarded for each block intentionally given to the experimenter (range: 0–3). The total prosocial score was calculated as the sum of the scores obtained on the puppet preference task and the cooperative helping task.

Joint attention task

The child was presented with two attractive objects (a bouncing ball and a small remote-controlled mechanical robot), adapted from the Early Social Communication Scales (Mundy et al., 2003). Video recordings were used to count the number of child's gaze alternated between the object and the experimenter, as well as index-finger pointing behaviors. These variables were summed up to obtain a total joint attention score.

Assessment of tactile exploration abilities

Tactile exploration task

In this task, the experimenter presented the child with a variety of objects to explore for a few minutes (Gillioz et al., 2024b). First, the experimenter offered a plastic cube, with each face featuring a distinct interactive element (e.g., a button press, a rotating wheel, etc.). Next, the experimenter offered three different toys: a small wooden rattle, a wrist-worn plush toy and a set of plastic electronic keys. These objects differed in texture and appearance. The exploration was video-coded in order to quantify the duration of active tactile exploration and the number of different exploration strategies (e.g., pressing, shaking, stroking, mouthing, rotating/observing, throwing).

Procedure

All tasks were administered by clinical psychologists from the center or by psychology students who had received specific training beforehand. The assessments took place during the routine follow-up consultation scheduled due to prematurity. The administration of the Bayley Scales required approximately 20–30 min, after which the experimental tasks were proposed, lasting an additional 20 min. One or more breaks were offered to the child whenever needed. The sessions were video recorded to allow for subsequent analyses. Prior to the appointment day, parents received an email with a link to complete the questionnaires online via the Qualtrics platform. If the questionnaires had not been completed before the visit, parents were asked to fill them out either during the consultation or in the following days. Completing the full set of questionnaires generally required 30–40 min.

All experimental sessions were video-recorded and coded offline by the same trained rater using predefined scoring criteria. To minimize any potential bias, the coding of the experimental tasks was completed before the rater had access to the Family Media Use questionnaires; the rater was therefore unaware of the participants' screen time. Since all experimental tasks were coded by a single trained rater, inter-rater reliability was not assessed.

Statistical methods

Participants for whom data on direct screen exposure were missing were excluded from analyses. For all other analyses, the available case data were used. Data from questionnaires, neonatal characteristics extracted from medical records, psychomotor assessments, and performance on experimental tasks were analyzed by using descriptive and bivariate statistics. Quantitative variables were reported using minimum and maximum values, means, and standard deviations. Qualitative variables were expressed as percentages. Children were divided into two groups based on their exposure or non-exposure to screens, as reported in the Family Media Use questionnaire completed by parents. We then examined differences between the two groups in term of psychomotor and socio-emotional development, as well as their performance on the experimental tasks. Group differences in motor, cognitive and language scores, socio-emotional outcomes, and tactile exploration abilities were analyzed using Welch's two sample t -test, appropriate for samples with unequal variances. The associations between screen exposure and children's developmental and behavioral outcomes were further explored in greater detail, first across our entire population and then by focusing exclusively on children who had been exposed to screens. Since most variables were not normally distributed, we conducted non-parametric Spearman's correlation analyses. Partial correlations were performed to adjust for gestational age and parent's socio-economic status given their well-established links to early neurodevelopment. A p-value below 0.05 was considered statistically significant. Spearman correlation coefficients are reported with their 95% confidence intervals, which were estimated using bootstrap resampling (5,000 iterations). Statistical analyses were performed using JASP (JASP-Team, 2025).

Cohort characteristics

The sample consisted of 38 males (59.4%) and 26 females (40.6%). The mean gestational age was 28.9 weeks (SD = 2.18, range = 24–31.7), and the mean birth weight was 1,162 g (SD = 369.16, range = 480–2,060). BPD was diagnosed in 38 infants (59.4%), while 12 infants (18.8%) developed sepsis during the neonatal stay. Two infants presented moderate brain abnormalities: one child had grade 3 intraventricular hemorrhage, and the other had cerebellar hemorrhage. The mean socio-economic status score was 3.9.

Family media use questionnaire

All 64 parents completed the Family Media Use questionnaire. The average number of screens per household was seven [mean (SD) = 7.1 (3); range = 2–17]. The most used screen by parents in their personal use is the smartphone (on average 47.8% of the time), followed by the TV (average 25% of the time), then the computer (average 18% of the time), games consoles (average 3.4% of the time) and finally tablets (average 3.4% of the time). Parents' screen time was calculated separately for work and private use. In their professional activities, parents used screens for an average of 5.9 h a day (SD = 4.1, range = 0–16). In their private time, parents used screens for an average of 5 h a day (SD = 3.1, range = 0.75–14). Most parents (46 out of 64; 72%) reported spending time in front of screens while their child is present in the room, with an average daily duration of 1 h (SD = 1.2, range = 0–5). Additionally, 15 out of 64 parents (23%) stated that the TV is often on in the background, with a daily average of 0.30 h (SD = 0.69, range = 0–4). Finally, 14 out of 64 parents (21.9 %) stated they systematically watched TV during family meals. Of the 64 children, 25 (39%) were directly exposed to screen with an average time of 4.03 h a week (SD 3.80, range 0.5–14).

Psychomotor development

Neurodevelopmental outcomes of all 64 children enrolled in the study were assessed using the Bayley Scales of Infant and Toddler Development (BSID-III). The mean corrected age at the time of the assessment was 12.18 months (SD = 0.32, range = 11.43–13.1 months). Mean cognitive, language and motor scores were within the average for typically developing children (cognitive score = 108.2, SD = 13.1; language score = 97, SD = 13.4; motor score = 94.5, SD = 11.5).

Socio-emotional development

Social-emotional development was evaluated using the ASQ-SE questionnaire completed by parents. Of 64 participants, 53 filled the questionnaire (82.8%). Among these, seven (13.2%) scored above the cutoff of 48, indicating potential socio-emotional difficulties.

Comparison between children directly exposed to screens and those not exposed

First, we aimed to compare children directly exposed to screens with those who were not across the various developmental outcome measures. To do so, we divided our cohort of VPT children into two groups: a screen group, comprising children who had been directly exposed to screens, and a no-screen group, including those with no reported screen exposure. There were no significant differences between the two groups in terms of perinatal and demographic characteristics, and age at assessment.

No statistically significant differences were found between the two groups on any of the three BSID-III scales, nor in social-emotional skills measured by the ASQ-SE. In addition, no significant differences were observed between the two groups in the scores of the different experimental tasks assessing socio-emotional competencies and tactile exploration abilities (Table 1).

Associations between direct screens exposure and developmental outcomes across the entire cohort

To examine whether screen exposure was associated with developmental outcomes at the population level, we conducted Spearman correlation analyses in the full cohort, including both exposed and non-exposed children. Spearman correlation coefficients are reported together with their 95% confidence intervals, which were estimated using bootstrap resampling (5,000 iterations).

In the full cohort, no significant associations were observed between total screen time and cognitive performance [ρ = −0.07, 95% CI (−0.32, 0.18), p = 0.30], language [ρ = −0.20, 95% CI (−0.42, 0.04), p = 0.06], motor development [ρ = 0.05, 95% CI (−0.21, 0.30), p = 0.65], ASQ-SE [ρ = 0.13 95% CI (−0.12, 0.39), p = 0.829], prosocial behavior [ρ = −0.19, 95% CI (−0.42, 0.07), p = 0.08], joint attention [ρ = 0.10, 95% CI (−0.13, 0.33), p = 0.79], or number of strategies during tactile exploration [ρ = −0.12, 95% CI (−0.38, 0.4), p = 0.18]. Results remained similar after adjustment for gestational age and socioeconomic status.

Given the lack of significant correlations within the entire cohort, we then focused on the subgroup of children directly exposed to screens to explore whether variability in screen exposure among these children was associated with developmental outcomes.

Associations between direct screens exposure and developmental outcomes in children directly exposed to screens

To examine the association between direct screen exposure and psychomotor development, Spearman's rank-order correlations were conducted. Screen exposure was significantly negatively correlated with both cognitive and language scores [ρ = −0.34, 95% CI (−0.62, 0.03), p = 0.048; ρ = −0.56, 95% CI (−0.76, −0.21), p = 0.002, respectively]. These associations remained statistically significant after controlling gestational age and parental socio-economic status [ρ = −0.35, 95% CI (−0.63, 0.17), p = 0.036 and ρ = −0.45, 95% CI (−0.73, −0.01), p = 0.02 respectively, Figures 2a, b]. No significant correlation was observed between screen exposure and motor scores [ρ = −0.13, 95% CI (−0.60, 0.31), p = 0.27] even when adjusted for gestational age and SES [ρ = −0.15 CI (−0.63, 0.40), p = 0.25].

Associations between direct screens exposure, socio-emotional capacities, and tactile exploration abilities in children directly exposed to screens

To examine associations between screen exposure, socio-emotional capacities, and tactile exploration abilities, Spearman's correlations were computed between screen time (expressed in hours per week), the total score of the ASQ-SE questionnaire, and performance on the different experimental tasks. No significant correlations were found between screen exposure and the ASQ-SE questionnaire [ρ = 0.021, 95% CI (−0.45, 0.49), p = 0.46], prosocial behavior [ρ = −0.15, 95% CI (−0.63, 0.40), p = 0.22] or joint attention [ρ = 0.04, 95% CI (−0.45, 0.38), p = 0.58]. However, a significant negative correlation was observed between screen exposure and tactile exploration: higher direct screen exposure was associated with fewer strategies used by children to explore different toys [ρ = −0.39, 95% CI (−0.70, 0.05), p = 0.02]. This association remained significant after adjusting for gestational age and parental socio-economic status [ρ = −0.54, 95% CI (−0.08, −0.22), p = 0.002, Figure 3].

Associations between family socio-economic status and children's media exposure

Finally, we wanted to explore the association between family socioeconomic scores (as a reminder, high scores correspond to low socioeconomic status) and screen exposure. No association was found between family socioeconomic scores and screen exposure across the entire cohort [ ρ = 0.18, 95% CI (−0.09, 0.43), p = 0.08].

However, Spearman correlations revealed a positive correlation between socio-economic scores and screen exposure in children directly exposed to screens [ρ = 0.42, 95% CI (0.03, 0.73), p = 0.02, Figure 4].

Understanding how screen exposure relates to early developmental outcomes during the first year of life is increasingly important, particularly in very preterm (VPT) children, a population especially vulnerable to environmental influences on cognitive and behavioral development. To our knowledge, no previous study has explored this question on this vulnerable population at 12 months of age—a critical period during which environment has been shown to strongly influence cognitive and behavioral outcomes.

Home screen equipment and parents use of screen

To better understand the context of screen exposure, we first assessed the availability of screen devices and the frequency of screen exposure of parents and children during the first year of life, using an online questionnaire. Data revealed that screens are widely present in the households of families that participated in our study, with an average of seven devices per household. This finding aligns with results from the Elfe cohort survey, representative of children born in France in 2011 and aged 1 year at the time of the study (Berthomier and Octobre, 2019), and with the online survey conducted in Switzerland by Gillioz et al. (2022), involving 486 parents of children aged between 1 month and 3 years.

Parents in our cohort use screens for an average of almost 6 h per day for professional activities and 5 h per day for personal use. Additionally, 72% of parents reported using screens while their child is present in the room. About a quarter of the parents who participated in our survey reported having the TV on in the background for around half an hour a day, and a quarter said they systematically watched TV during mealtimes.

VPT children's screen exposure

Survivors of VPT birth are at risk for a broad spectrum of developmental disorders including, motor, cognitive, and socio-emotional problems (Barre et al., 2011; Delobel-Ayoub et al., 2009; Spittle et al., 2009). Exploring risk factors for outcomes in preterm children, previous studies by our group have shown that unfavorable environmental conditions have an impact on overall cognitive development (Gui et al., 2019; Pittet-Metrailler et al., 2019) but also on language (Vandormael et al., 2019) and behavior (Bartal et al., 2020). Due to the lack of published studies, we were unable to compare our results with previous research specifically addressing the associations between screen exposure and developmental outcome in premature infants. We will therefore interpret our findings in the context of the broader literature on children in general.

Most children in this study were not directly exposed to television and other devices in the first year of life. By the age of 12 months corrected age, 39% of children were directly exposed to screens, with an average time spent in front of screens of 4 h per week. The screen time of our cohort of VPT infants was higher than that of infants under 1 year of age in the Gillioz et al. (2022) study, which averaged 1.75 h per week. In France, as in the United States, 68% of two-year-olds watch television every day and spend an average of over 2 h a day watching it (Berthomier and Octobre, 2019; Duch et al., 2013). It will therefore be essential to monitor the evolution of screen time in our cohort of VPT children in relation to their age. Indeed, as shown by the Elfe study, which described the progression of screen time among children aged 2 to 5.5 years, screen use increases steadily with age, with an average daily time of 52 min at age 2, 1 h 20 at age 3.5 and 1 h 40 at age 5.5 (Yang et al., 2024).

We observed a positive association between socioeconomic status and screen exposure within the cohort of children directly exposed to screens, indicating that lower socioeconomic status was associated with greater screen exposure. This relationship has been documented in the literature, with several studies showing that children from lower socioeconomic backgrounds are more likely to experience higher levels of screen exposure (Gillioz et al., 2024a; Kucker and Schneider, 2026). Li et al. (2020) in their meta-analysis and systematic review, discussed how family factors, such as socioeconomic status (SES), influence children's screen use and health indicators, highlighting that children from lower SES families tend to have more screen time. This pattern is consistent with findings by Poncet et al. (2022), who showed that parents with a low socio-economic status tend to adhere less to screen-time guidelines for toddlers. Consequently, we corrected for parent socioeconomic status in the following analyses to better isolate the specific effects of active exposure on developmental outcomes.

Screens exposure and psychomotor development

While most studies report associations rather than causal links between screen exposure and developmental disorders, early exposure has been related to less favorable cognitive, language, and emotional outcomes (Madigan et al., 2019, 2020; Pagani et al., 2013; Walsh et al., 2020). Cognitive skills develop rapidly during infancy and are strongly influenced by early environmental experiences. Thus, environmental factors such as screen exposure may also reduce opportunities for active engagement with the environment and interactive play, both of which are crucial for cognitive stimulation and learning.

Given the critical role of the environment in early development and the well-established links between screen exposure and developmental outcomes, we expected to observe differences between the groups with and without screen exposure. However, no significant differences were found between the two groups for BSID-III motor, cognitive, and language scores. The wide variability in screen exposure within the exposed group (ranging from 0.5 to 14 h per week at 12 months) may have limited the sensitivity of the group comparisons to detect potential differences.

To better account for this variability, we subsequently conducted correlation analyses to examine whether the amount of direct screen time was associated with developmental outcomes. We found that the associations between screen time and developmental outcomes varied depending on the level of analysis. Across the entire cohort, no significant association was observed between screen time and cognitive, motor, or language outcomes. Likewise, no association was observed between screen exposure and motor development within the subgroup of screen-exposed infants, either before or after adjustment. This lack of association should be interpreted considering the heterogeneous evidence in the literature regarding association between motor outcomes and screen exposure in early childhood. For instance, Lin et al. (2015) reported that children exposed to lower levels of screen time between 15 and 35 months exhibited better motor skills, suggesting a potential link between screen exposure and reduced opportunities for physical activity. However, such associations may depend on developmental timing and exposure characteristics, which could partly explain why they were not detected in our cohort at 12 months of age. However, within the subgroup of infants directly exposed to screens, screen time was negatively correlated with both cognitive and language scores. Even after adjusting for gestational age at birth and parent's socioeconomic status, the associations remained significant. These results suggest that the observed associations are primarily due to variability in exposure within the screen-exposed group rather than to differences between exposed and non-exposed infants within the entire cohort. These findings are consistent with other studies that found negative associations of screen time with cognitive and language development (Madigan et al., 2019, 2020; Martinot et al., 2021; Yang et al., 2024; Zimmerman et al., 2007). In a longitudinal study of 2,441 participants, Madigan et al. (2019) explored associations between screen time and general development. They showed that children with the most screen exposure at 24, 36, and 60 months of age performed less well on the Ages and Stages Questionnaire, a screening questionnaire completed by the parents which assesses various aspects of psychomotor and communicative development. In addition, analyses also revealed later effects, as greater screen times at 24 and 36 months were significantly associated with poorer performance on developmental screening questionnaire at 36 months and 60 months, respectively. In addition, Fitzpatrick et al. (2025) reported that preschoolers with higher or increasing screen time trajectories tended to show poorer executive functions—including working memory, inhibitory control, and cognitive flexibility. This reduction in enriching experiences could be particularly relevant for VPT infants, who are already vulnerable to neurodevelopmental impairments (Fitzpatrick et al., 2025). Language development is also especially sensitive to early environmental inputs. Given that brain development is largely shaped by early sensory experiences, exposure to language in the infant's environment during the first year after birth is of utmost importance. Indeed, language development is a phenomenon largely modulated by the environment and takes place mainly during dyadic exchanges between child and parent. When children are alone in front of screens, they are at risk of being deprived of this interaction. As demonstrated by Christakis et al. (2009), parents interact less with their infants when the television is on. In fact, having the television on was associated with a significant reduction in parental speech and a decrease in children's vocalizations between the ages of 2 and 48 months. Zimmerman et al. (2007) showed an association between early viewing of baby DVDs/videos and poor language development in infants aged 8 to 16 months. Madigan et al. (2020) carried out a meta-analysis looking for correlations between expressive and receptive language development and multiple aspects of screen use by young children (mean age of 44.4 months.) Negative correlations were found between the development of children's language skills and the total duration of daily exposure to screens (r = −0.14), as well the presence of television on in the background (r = −0.19). They also observed an association between the age of first exposure to screens and the development of the child's language skills (r = 0.17), indicating that later exposure to screens was linked to better language skills in children. More recently, Brushe et al. (2024) found that increased screen time between 12 and 36 months of age was associated with reduced parent–child verbal interactions, a key factor known to support early language acquisition. In their longitudinal study Takahashi et al. (2023), reported that early screen exposure was linked not only to later communication difficulties but also to broader cognitive domains, including early problem-solving abilities, suggesting that its influence may extend beyond language-specific outcomes to more general aspects of early neurodevelopment. These findings are consistent with our results, which highlight associations between screen exposure and emerging cognitive and language skills during the first year of life.

High exposure to screens therefore appears to be negatively associated with language development, particularly when it occurs early in the child's life. In our cohort, while group comparisons did not reveal significant differences, the negative correlations between screen exposure and developmental outcomes further highlight the potential risks of screen exposure, particularly for cognitive and language development. As with previous research, these findings underscore the importance of carefully monitoring screen exposure during early childhood, particularly among vulnerable children such as those born very preterm.

Screens exposure and socio-emotional competencies

Differences in socio-emotional behavior between children born very preterm and those born full term have been reported from as young as 6 weeks of age (Hughes et al., 2002). The findings of one of our previous study suggest that very preterm infants at 12 months of corrected age have an atypical socio-emotional development, characterized by difficulties in regulating emotional and behavioral responses (Langerock et al., 2013). Also, Spittle et al. (2009) found that children born very preterm showed, at the corrected age of 2 years, more dysregulated behavior and poorer social-emotional competences than their term born peers. Furthermore, at 2 years, those infants were classified as showing poorer self-regulation abilities than full term infants in several problem-solving games (Clark et al., 2008). One possible explanation for the difficulties observed in the socio-emotional development of VPT infants is the challenging environment in which they grow up. Premature birth, along with the stress and trauma it often generates, can negatively affect the quality of interactions between parents and infants, which are essential for the child's emotional and behavioral regulation. The premature birth experience can expose both the child and the parents to significant stress, which, as demonstrated by Morisod-Harari et al. (2013), can hinder the establishment of positive parent-infant interactions. This, in turn, can disrupt the development of essential socio-emotional skills. Indeed, a link has been established between post-traumatic stress in mothers following preterm birth and the quality of mother-infant interactions (Forcada-Guex et al., 2011). Furthermore, data collected by McDonald et al. (2018) from participants in the “All Our Babies/Families cohort” suggest that a lack of daily play-based interactions with their parents between the ages of 1 and 2 years is a risk factor for delayed social-emotional development and behavioral problems at age 2, as measured by the Brief-Infant-Toddler Social and Emotional Assessment. Indeed, in the first years of life, face-to-face interactions, eye contact and shared attention with the parent are essential for infants to develop appropriate communication and socio-emotional skills (Ginsburg, K. R., American Academy of Pediatrics Committee on Communications, and American Academy of Pediatrics Committee on Psychosocial Aspects of Child and Family Health, 2007). It is therefore not surprising that a growing body of evidence suggests a negative association between early childhood exposure to screens and social and emotional development (Chonchaiya et al., 2015; Hinkley et al., 2018; Lin et al., 2020). Thus, for VPT infants, whose socio-emotional development is already at risk due to environmental factors, early screen exposure may further hinder the development of essential social and emotional skills, as it reduces opportunities for meaningful parent-child interactions (Hamlin, 2013). Given these findings, we examined whether screen exposure would relate to differences in socio-emotional outcomes in VPT children at 12 months CA. However, no significant differences in social-emotional development, as assessed by parents via the ASQ-SE questionnaire, were observed between screen-exposed and non-exposed children.

Even as toddlers, children tend to prefer individuals who display helpful or prosocial behaviors over those who exhibit antisocial behaviors. They feel concerned by the suffering of others and try to comfort them (Hamlin and Wynn, 2011). We therefore hypothesized that children exposed to screens would have had significantly fewer social experiences in the real world than unexposed children and might therefore be less sensitive to the pro-social behaviors of other individuals. In this sense, their preference for the pro-social puppet would not be as marked as in non-exposed children. Contrary to our expectations, we found no difference between children exposed to screens and those who were not, and no significant correlation between duration of exposure to screens and prosocial behavior score.

When one person voluntarily coordinates his or her attention with that of another, this is called “joint attention” (Braithwaite et al., 2020). Joint attention involves two people paying attention to the same thing, intentionally and for social reasons. Whether pointing or looking, children will learn to use these different gestures to interact and attract the attention of the person with whom they want to communicate. In this way, they can also show their interest in a particular object. Before the age of 9 months, this is done through gaze: there's what we call a sharing of interest between child and parent, with the child looking at the adult, then at the object of interest, then back at the adult, and so on. Next, the child begins to use his hands and point to draw the adult's attention to a particular object (Mundy, 2018). We therefore hypothesized that children exposed to screens should be less accustomed to interacting with the people around them and, consequently, show fewer signs of initiating joint attention and sharing with the adult than non-exposed children. Contrary to our expectations, we found no difference between children exposed to screens and those who were not, and no significant correlation between duration of exposure to screens and the score of joint attention at age 12 months. Interestingly, partially converging evidence has been reported by Gillioz et al. (2025), who found that toddlers not directly exposed to screens initiated significantly more joint attention behaviors, particularly alternating looks, than those directly exposed. However, no significant differences were observed for pointing behaviors, nor as a function of background or total screen exposure time. These findings suggest that early screen exposure may differentially affect specific components of joint attention, with gaze-based behaviors being more sensitive than gesture-based ones. One possible explanation for the lack of detectable effects in our sample is that only few subjects were already pointing at 12 months CA. Several studies indicate that preterm children often exhibit delays in the initiation of pointing—a behavior that reflects the beginning of proto-communicative joint attention. For example, Mateus et al. (2020) found that extremely preterm infants (≤ 28 weeks) scored significantly lower on joint attention response at both 12- and 18-months CA compared to their full-term peers, although no difference was observed in the initiation of joint attention behaviors. Similarly, Sansavini et al. (2019) showed that extremely preterm infants produce significantly fewer pointing gestures at 18 months than full-term controls—a finding linked to subsequent language delay. Thus, the delayed onset of pointing in our sample may have limited our ability to detect associations between screen exposure and measures of joint attention at this early age.

One possible reason why this study did not reveal differences in socio-emotional behavior or in certain experimental tasks between children exposed to screens and those who were not is the very young age of the children in this cohort. These findings underscore the need for longer follow-up periods to determine whether associations between early screen exposure and later socio-emotional development or multisensory skills might emerge over time.

Screens exposure and tactile exploration abilities

Finally, we hypothesized that children exposed to screens, and therefore interacting less regularly with objects or toys, should use tactile exploration strategies (e.g., exerting pressure on the object, lifting it, squeezing it to discover its volume, touching its contours, caressing it to apprehend its texture) that are less appropriate and for a shorter duration of time than unexposed children, since they should be in contact less often with the objects in their environment. In this task, we found no differences between children exposed to screens and those who were not. However, we did find a negative correlation between the duration of screen exposure and the number of strategies used by children to explore different toys, a relationship that remained significant after adjusting for gestational age and parental socio-economic status. These findings are consistent with those of Gillioz et al. (2024b), who showed that greater screen time in children aged 6 to 36 months was associated with reduced tactile exploration skills. Their study also highlighted the importance of alternative sensorimotor activities, such as playing with real objects, in fostering exploratory behaviors that support early cognitive and perceptual development.

Limitations

Several limitations of this pilot study should be acknowledged. First, as in most observational studies, screen exposure was assessed using parent-reported questionnaires, which are subject to recall bias and social desirability bias. Parents may have difficulty accurately estimating the amount of time their children spend in front of screens and may underreport this exposure. In addition, no objective measures (such as direct observation or digital tracking) were used, which may have affected the accuracy of the assessment of screen exposure.

Second, beyond total screen time, our measurement did not consider the qualitative aspects of media use, such as the type of content (educational or entertainment), or contextual factors such as watching programs and sharing content with an adult. Exposure to television or other screens playing in the background was also not included in the main analyses. Furthermore, children's off-screen activities should be taken into account, as they may mitigate the effect of screen use on cognitive, language, and motor development. These unmeasured characteristics may influence developmental outcomes in different ways and should be considered in future research.

Third, although the analyses were adjusted for gestational age and parent's socioeconomic status, the observational design of the study precludes any causal inferences, and the presence of residual confounding factors. In particular, other factors, such as the quality of parent-child interactions, parents' mental health, and neonatal medical complications, may also contribute to child development outcomes and were not fully accounted for.

Fourth, the relatively small sample size of children, due to a high dropout rate and the fact that only a limited number of children were directly exposed to screens, reduces statistical power and increases the risk of Type II error.

Finally, larger-scale longitudinal studies, based on more comprehensive and objective assessments of screen exposure, are needed to confirm these preliminary findings.

Despite these limitations, this pilot study provides new data on the associations between early screen exposure and neurodevelopment in very preterm infants, a particularly vulnerable population that remains understudied.

To our knowledge, this is the first available data on the relationship between exposure to screens and psychomotor, communicative, socio-emotional and tactile competencies, in a cohort of children born very preterm, at the corrected age of 12 months. Evidence was found for negative associations between screen time and cognitive and language scores, as well as between screen time and tactile strategies used by children to explore toys. These associations were still significant after controlling factors influencing neurodevelopment, such as gestational age at birth and parents' socio-economic status.

Interventions aimed at raising parental awareness of the risks linked to screen exposure in the first years of life are essential but currently appear to be insufficient. To truly support healthy child development, it is crucial to offer parental guidance that goes beyond simple awareness, focusing on developing skills that enable them to recognize and respond to the cues given by very young children. In addition, fostering parent-child bonding from an early age through activities that promote meaningful interaction, tailored to each family's unique situation, should be at the center of to any recommendations. Such interventions are essential for enhancing familial interactions and promoting positive child development from birth. Exploring the effect of screens exposure on early childhood development, as well as on early signs of language and social-emotional development, would help legitimize recommendations, guide parents in the appropriate use of screens, and tailor prevention messages to early childhood professionals.