A Part to Play: How Caregivers Shape Infants' Visual Exploration and Alter Their Neural Mechanisms of Attention

Project Information

InterPlay

Grant agreement ID: 891535

DOI

10.3030/891535

Project closed

EC signature date 23 April 2020

Start date 1 July 2020

End date 26 July 2023

Funded under

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Total cost

€ 184 590,72

EU contribution

€ 184 590,72

Coordinated by

THE PROVOST, FELLOWS, FOUNDATION SCHOLARS & THE OTHER MEMBERS OF BOARD, OF THE COLLEGE OF THE HOLY & UNDIVIDED TRINITY OF QUEEN ELIZABETH NEAR DUBLIN
Ireland

Periodic Reporting for period 1 - InterPlay (A Part to Play: How Caregivers Shape Infants' Visual Exploration and Alter Their Neural Mechanisms of Attention)

Reporting period: 2020-07-01 to 2022-06-30

Summary of the context and overall objectives of the project

Infants are curious learners. They face the daunting task of learning to understand the world around them. Healthy infants born at term (that is, between 37 and 42 weeks of pregnancy) seem to attend to objects and people in their environment in a way that is beneficial for learning. For example, researchers have shown that infants prefer to look at sequences according to “the Goldilocks effect”: Infants prefer those sequences that are neither too boring nor too complex, but rather exactly right (Kidd, et al., 2012). Infants born prematurely (that is, before 37 weeks of pregnancy) allocate their visual attention differently from infants born at term. It has been hypothesized that preterm infants do not allocate their attention in a way that is beneficial for learning, which could possibly explain why some preterm infants later experience developmental problems or delays. It is therefore essential to understand how learning via visual attention happens in order to grasp the impact of deviating attention allocation on development. The MSCA-IF project “InterPlay” set out to increase our understanding of how infants learn by focusing on the way they decide where to look.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

Due to the COVID-19 pandemic, the project had to be adjusted to the circumstances by conducting most experiments online and developing the method of online testing on small and large scale.

Based on the Goldilocks effect described in earlier research, four online studies measuring infants’ visual exploration and attention allocation have been conducted. Instead of focusing on the complexity of the sequence, the studies investigated infants preferences for various levels of visual complexity. For example, do infants prefer simple drawings of an animal over real-life photos? And how do their preferences change with age in the first year of life? More than 150 datasets have been acquired and are in the process of being analysed. The results are expected to increase our understanding of how infants learn from observing their environment. They will furthermore help developmental scientists to create studies that are engaging for infants by better knowing what infants like to see.

Originally, the InterPlay project foresaw the inclusion of a caregiver-infant interaction component. The plan was to investigate to what extent caregivers guide or maintain infants' visual attention and whether these attention guiding or maintaining behaviors could have a beneficial influence on infants’ learning. Due to the COVID-19 pandemic, this research could not be carried out as planned. Instead of bringing infants and caregivers into the lab, an online study has been set up. This study allows caregivers and infants to do a small interaction game at home while being recorded on video. The study makes use of a novel method to acquire rich data that aims to closely mimic the laboratory setup. The data collection for this study is still in progress. Our pilot data is looking promising for wider adoption of this new method both for scientific purposes as well as for clinical or educative applications.

Triggered by the experiences and based on the expertise acquired with the aforementioned online studies, the still growing "ManyBabies-AtHome" initiative (an international, multi-lab collaborative project that aims to improve online developmental science experiments and make the methods globally accessible) has been started out of InterPlay. Online testing in developmental science is usually done in North America or Europe and often targets participants from these continents. The method has the potential to reach a much more diverse population, though. ManyBabies-AtHome is overcoming the hurdles associated with studying larger, more diverse samples of infants via browser-based experiments. A second goal is to provide a large, multicultural, annotated dataset that can be used for training more accurate automatic gaze coding algorithms for developmental science. The first study that has been set up within the consortium is testing infants’ preference for visual stimuli and attention allocation. Differently from earlier studies on this topic, this study will acquire a dataset from infants across the world, allowing for cross-cultural comparisons as well as in-lab versus online comparisons. Once the first study has been shown successful, the consortium plans to conduct various other studies.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

The InterPlay project has gone beyond the state-of-the-art both in content and in method. The previously described Goldilocks effect has been shown only for sequences. In InterPlay, other types of complexity have been explored to gain better understanding on how infants select the stimulus from which they want to learn. The four studies carried out in the project are only the beginning of a large range of studies that will have to be conducted to obtain insight in the many factors that could be influencing infants behavior. Furthermore, InterPlay has made substantial contributions to the method of online testing in developmental science. A new way of testing caregiver-infant interaction has been developed and is currently being used to collect data. Moreover, the ManyBabies-AtHome project (that was started from within InterPlay) is the only large scale, international consortium with the specific goal of developing and improving the method of testing infants and children online and making it accessible worldwide. This method allows researchers to acquire diverse samples whilst relieving the burden of requiring a lot of monetary or material resources. Once accessible to the larger scientific community globally, this method thus has the potential to enable researchers from less wealthy countries or institutions to acquire many interesting datasets and answer important research questions.

The research carried out in the InterPlay project has impact on clinical applications as well, as it may facilitate the development of clinical interventions for infants born prematurely. 4-16% of infants are born prematurely, an estimated 13.4 million infants per year worldwide. Infants who are born prematurely are more likely than infants born at term to have later developmental, cognitive or socio-emotional problems. However, clinicians currently have no way of predicting which infants will develop such problems. This means that clinicians have to adopt a “wait-and-see” approach until potential problems have gotten severe enough to be reliably detected. The InterPlay project may be able to help in various ways. First, by understanding how infants typically allocate their visual attention, it becomes possible to detect individual deviations from the typical behavior and closely monitor these infants over time to intervene early and adequately. Second, the development of the online testing method can be expanded to also be used by clinicians to do remote consultations or interventions. This means clinicians can offer help to a wider range of people. Moreover, the InterPlay project has been carried out in close collaboration with the EU-funded FOUNDCOG project (ERC Advanced, led by Prof. Rhodri Cusack), in which the goal is to identify early neural markers for developmental problems in infants born prematurely and at term. This puts us in a position to potentially identify neural or behavioral markers for developmental problems by looking at the data acquired in these projects jointly.

The fellow, whilst 22 weeks pregnant, presenting their Nature Machine Intelligence published work.

The fellow hosting a virtual ManyBabies-AtHome consortium meeting during the first lockdown