What factors affect digital device viewing behaviour in children?

Paper title: Digital device viewing behaviour in children

Authors: Josh Richards¹, Matt Jaskulski^1,2, Martin Rickert¹, Pete Kollbaum¹

1. Indiana University School of Optometry, Bloomington, Indiana, USA
2. VisionApp Solutions S.L., Murcia, Spain

Date: 21 February 2024

Reference: Richards J, Jaskulski M, Rickert M, Kollbaum P. Digital device viewing behaviour in children. Ophthalmic Physiol Opt. 2024 May;44(3):546-553.

[Link to open access paper]

As the use of digital devices becomes increasingly common among children and teenagers, concerns have arisen regarding their potential influence on the development of myopia.¹ This study investigated how children's habitual viewing behaviours with digital devices might impact myopia development, aiming to measure the distances at which children view screens during various activities to determine their role in myopia onset and progression.

The study involved 38 children, each using their habitual vision correction, who completed five different tasks on a Google Pixel 3 smartphone using MyopiaApp software. These tasks were randomly assigned and included playing a game, watching videos in both well-lit (680 lux) and dimly lit (5.5 lux) environments, and reading text in small (8 pt) and large (16 pt) font sizes. The analysis examined how factors such as task type, whether the child was myopic or non-myopic, age, and arm length affected the median viewing distance, using ANCOVA statistical methods.

The findings showed no significant correlation between arm length and viewing distance across any task. There were also no substantial differences in viewing distance based on task type, refractive status (myopic or not), age, or arm length. Notably, approximately 60% of the variability in viewing distances was attributed to individual differences (fidgeting and trouble sitting still) rather than specific tasks or physical characteristics.

Previously, arm length,^2,3 refractive status⁴ and room illumination⁵ were thought to influence viewing distance. This study indicates that this is not the case and that these factors were poor predictors of a child’s viewing behaviour. Rather than relying on these factors to predict viewing behaviour, clinicians may find this novel MyopiaApp technology helpful in understanding individual children’s smartphone usage and viewing behaviours and give personalised advice and treatment accordingly.

Although the study aimed to minimize bias by keeping children unaware of the specific measurements being taken, the mere presence of study conditions could alter natural behaviour. The setup—sitting alone in a test room—might differ from how children typically interact with devices at home.

While the study includes a broad age range of children from 6 to 17 years, it would be valuable to explore the behaviours of infants and toddlers as well. It is becoming increasingly common for infants to interact with digital devices from their first year of life.^6,7 Given the statistically significant association between early device use and myopia development,⁶ it is important to investigate how viewing behaviours in very young children might influence eye health and how these behaviours evolve with age. Understanding these patterns could provide insights into early interventions and preventive measures for myopia.

Introduction: Habitual viewing behaviour is widely believed to be an important contributing factor to the onset and progression of myopia and may be task dependent. The purpose of this study was to quantify the habitual viewing distance of children performing five different tasks on a smartphone digital device.

Methods: The real-time viewing distance in 38 children with their habitual correction was measured using software (MyopiaApp) on a handheld (Google Pixel 3) device. Five tasks were performed in a randomised sequence: playing a game, watching video in a light (680 lux) and dark (5.5 lux) environment and reading small (8 pt) and large (16 pt) text. ANCOVA statistical analysis was used to evaluate the effect of task, group (myope vs. non-myope) and arm length on the median relative viewing distance.

Results: Arm length was not correlated with viewing distance in any of the tasks, and there was no significant difference in viewing distance between any of the tasks. Specifically, a two-way mixed ANCOVA indicated that task, refractive group (myopic vs. non-myopic), age and arm length, as well as all two-way interactions were not significantly associated with viewing distance. Overall, 60% of the total variance in viewing distance was accounted for by individual differences.

Conclusions: The average handheld viewing distance was similar across a variety of everyday tasks in a representative sample of myopic and emmetropic children. Neither arm length, age nor refractive group were associated with viewing distance in any of the tasks. Importantly, myopic children of a given size did not hold the smartphone digital device at a different distance for any task than their equally sized non-myopic peers. However, both groups exhibited high inter-individual variability in mean viewing distance, indicating some subjects performed all tasks at further distances while other subjects used at nearer distances.