Measuring Executive Functions in Young Children from Low- and Middle-Income Countries: An Expert Interview with Michael Willoughby

While there has been increased research on how developmental changes in executive functions (EF) during early childhood contribute to children’s school readiness, most of it has focused on children in high-income countries. Much remains unknown about the impact of executive functions on the performance of children in low- and middle-income countries (LMICs).

Michael Willoughby, a Fellow in Education at RTI International, recently developed and evaluated a digital assessment tool of executive functions for use with preschool-aged children in LMICs. The tool, EF Touch, is administered using Tangerine, RTI’s open source software used to conduct off-line data collection and assessments on low-cost Android tablets. RTI received a digital innovation award for its use of this tool.

We sat down with Michael to learn more about EF Touch, how it is used, and what implications it has for broader research in the field.

What are executive functions, and why is this aspect of child development so important?

Executive functions are cognitive skills that facilitate problem solving, learning, and goal-directed activities. I intentionally use the word ‘skills’ instead of ‘abilities’ to convey the idea that executive function skills are malleable and can be improved. Executive function skills are interesting because they are higher order cognitive processes that act on other aspects of cognition. As such, they are often referred to as the ‘air traffic control system of the brain.’

A long-standing debate in the research literature involves which specific cognitive skills fall under the rubric of executive functions. In early childhood, there is consensus that executive function skills primarily consist of inhibitory control (the ability to override a dominant or habitual response), working memory (the ability to hold information in mind while processing or updating it), and cognitive flexibility (the ability to shift your attention or strategy in the face of changing information). Inhibitory control, working memory, and cognitive flexibility are three foundational executive function skills that facilitate later cognitive development including problem solving, decision making, and self-monitoring. Executive function skills undergo rapid improvements in early childhood, which is why a lot of EF research focuses on this developmental period.

You mention that EF is malleable and can be improved. What are some things parents or teachers can do to improve EF?

This is an active area of research and is not something that has a quick and easy answer. A number of other areas of children’s development are related to their executive function skills. For example, early language development, especially expressive language, is associated with improved executive function skills in children. Moreover, multiple aspects of the early caregiving environment, especially the experience of consistent, contingent, and sensitive caregiving behaviors, are related to executive function development.

In theory, efforts to enhance children’s early caregiving milieu should benefit their executive function skills — however, these supporting environments will benefit multiple aspects of children’s cognitive and social development and are not specific to EF. There is a growing body of literature on interventions that improve executive function skills. Interventions range from classroom curricula, to adaptive computerized training programs, to physical activity programs, all of which have some evidence for improving executive function skills. If there is a common ingredient among all of these programs, it is that executive function skills benefit from repeated experiences that both engage these cognitive processes and that make developmentally appropriate demands of children.

Tell us about EF Touch — how did it start, and how does it currently work within RTI’s Tangerine platform?

EF Touch is a computerized battery of tasks that were developed to measure executive function skills — including inhibitory control, working memory, and cognitive flexibility — in early childhood. EF Touch represents a collaborative effort with my colleague Clancy Blair. We were fortunate to receive research funding from the National Institute of Child Health and Human Development to develop a set of paper and pencil executive functions tasks.

After an extensive period of task development and testing, we were subsequently funded by the Institute of Educational Sciences in the United States Department of Education to develop a computerized version of tasks, which we referred to as EF Touch, because children made responses to tasks using a touch screen monitor. When I came to RTI, I learned about Tangerine, an electronic data collection software used for learning assessments and designed for use on laptops, tablets and smartphones. RTI made an internal investment to expand the functionality of Tangerine, which enabled us to administer the EF Touch tasks using tablets and the Tangerine software.

EF Touch contains many subtasks, each of which focuses on different areas of executive functions. What about the tasks and subsequent results do you find most exciting? Why?

First off, I wouldn’t say that I find any task more exciting than the others. Each task is an imperfect indicator of skill. In our research, we have emphasized the creation of a composite measure which indexes children’s performance across multiple tasks. Take the Bubbles task, for example, which is a reaction time task that serves as a control variable. In this task, bubbles appear on the screen and children are told to pop them as fast as they can. We’re interested in how quickly the kids will touch them once they first appear. In Tangerine, we can record this reaction time to the millisecond. We find that kids who are faster at that task actually do better on all of the EF tasks. This is backed up by the research, where reaction time is often used as an indicator of developmental progress. But what is the relevance of EF to broader learning outcomes? A recent paper showed a correlation between age and improvement in EF; if you don’t measure all tasks you can misattribute EF to other developmental growth. It’s unclear how early you can differentiate specific skills — some aren’t differentiated skill sets until kids are much older. This is why we like to focus on a battery-wide score rather than scores on individual tasks.

You recently piloted EF Touch in Kenya within RTI’s Tayari program, funded by the Children’s Investment Fund Foundation (CIFF). What were you expecting to find, and were these hypotheses confirmed?

We did a pilot in May [2017], which was mainly a proof of concept study. The primary goal of this work was to test the feasibility and usability of the Tangerine version of EF Touch tasks, including a new data collection platform (Android tablets versus Windows laptops), new setting (developing country context), and expanded language offering (tasks were administered in English or Swahili).

In general, children enjoyed and were able to complete most of the tasks, enumerators felt good about quality of the data that they collected, and individual differences in children’s performance on the tasks was evident. Children’s performance on the tasks showed the expected associations with age (older children outperformed younger children) and behavior (enumerator ratings of children’s attention-related behaviors were associated with improved task performance). Overall, it was a successful first experience.

Now that you have piloted EF Touch, what will you do next?

Based on the success of the May pilot, the Tayari team elected to include a subset of the EF Touch tasks in the Tayari midline assessment that occurred in October [2017]. All of the children in pre-primary classrooms, approximately 1,200 children, were given the opportunity to complete EF Touch tasks. Once we analyze this data, we’ll begin to ask systematic questions about how performance on EF Touch tasks is related to children’s performance on more traditional measures of school readiness, including pre-academic achievement and social-emotional functioning. Most of what is known about these associations is derived from studies in high-income countries. Our data analysis will be one of the largest efforts to date to consider these issues in a developing country context. In addition, we also plan to conduct formal psychometric tests of EF Touch tasks, including testing whether the tasks worked equally well (in a measurement sense) when administered in English and Swahili. This is important work that is essential for ensuring that any inferences that we draw from our data are not an artifact of differences in test performance for children from different language backgrounds.

What is next for EF Touch, and what other research do you think is necessary moving forward?

Our immediate focus involves data analysis and publishing the results from the Tayari program data collection. However, we are actively seeking out opportunities to partner with other organizations and projects that have an interest in measuring EF skills in early childhood. We want to learn the best way to facilitate and support external partners in their use of these tasks. As we think about new work and new applications, we also hope to connect to potential funders who have an interest in early childhood development and EF.

Children in LMICs have a unique set of experiences that likely influence the development of executive function skills for better and for worse. These children and the diversity of their experiences aren’t fully represented in the existing research literature, which is something that we’d like to address. Ultimately, early childhood educators around the world are interested in knowing whether executive function skills contribute to school success and, if so, whether it’s worth making an investment to improve these skills. We hope to play a central role in answering those questions.