Eye-tracking-based assessment of cognitive function in low-resource settings

Linda Forssman1, Per Ashorn1,2, Ulla Ashorn1

  • 1Tampere Center for Child Health Research, School of Medicine, University of Tampere, Tampere, Finland.

Insights

Infrared eye tracking shows promise for assessing infant neurodevelopment, even in low-resource settings. This automated method achieved high completion rates and replicated cognitive patterns in infants across Finland and Malawi.

Area of Science:

  • Developmental neuroscience
  • Infant cognitive assessment
  • Human-computer interaction

Background:

  • Infant neurocognitive development is vulnerable to socioeconomic factors like poverty and malnutrition.
  • Effective infant neurodevelopmental assessment requires tools usable early in life, across diverse settings.
  • Current assessment methods may lack objectivity or broad applicability.

Purpose of the Study:

  • To evaluate infrared eye tracking as a novel, automated infant assessment technique.
  • To assess visual-orienting, sequence-learning, and facial expression attention in 9-month-old infants.
  • To compare feasibility in high-resource (Finland) and low-resource (Malawi) settings.

Main Methods:

  • Infrared eye tracking piloted in Finland (N=39) and field-tested in Malawi (N=40).
  • Assessed visual-orienting, sequence-learning, and attention to facial expressions.
  • Measured parent perception, infant test completion, and valid trial percentages.

Main Results:

  • High parental acceptance (95% Finland, 92% Malawi) and infant test completion (95% Finland, 90% Malawi).
  • Satisfactory valid trial percentages (69-85% Finland, 68-73% Malawi).
  • Eye tracking showed higher completion rates than observational tests in Malawi; cognitive patterns replicated, though with slower processing in Malawian infants.

Conclusions:

  • Eye tracking is feasible for infant development assessment in low-resource settings, evidenced by high completion rates and replicated cognitive patterns.
  • The technique demonstrates potential for objective, automated infant assessment across diverse environments.
  • Further research is required on test-retest stability and predictive validity in low-income contexts.
Abstract

Related Concept Videos