Related Experiment Video
Updated: Mar 7, 2026
How to Obtain Reliable Visual Event-related Potentials in Newborns
Published on: October 24, 2019
Visual sensory determinants of EEG activity in neonates receiving intensive care
Christina Boman-Markaki1, Neelum Mistry2, Kimberley Whitehead3
1Research Division of Digital Health and Applied Technology Assessment (DHATA), King's College London, London, United Kingdom.
Insights
Infant brain activity (EEG) was largely unaffected by eye shields, suggesting visual cortex activity in sick newborns may be internally generated rather than driven by external visual input.
Area of Science:
- Neonatal neurology
- Neurophysiology
- Developmental neuroscience
Background:
- Cortical activity is vital for early neural organization, particularly in vulnerable infants in intensive care.
- Sensory determinants of cortical activity in neonates are not well understood.
- Eye shields are sometimes used for acutely unwell infants, presenting a natural experiment.
Purpose of the Study:
- To test if reducing external visual input with eye shields attenuates occipital cortical activity in neonates.
- To investigate the sensory determinants of cortical activity in vulnerable infants.
Main Methods:
- Analyzed 9-channel video-electroencephalography (EEG) in 18 sleeping neonates (14 with hypoxic-ischemic encephalopathy).
- Compared EEG power spectral density with and without eye shields, matched for vigilance, medications, and time.
- Cross-referenced EEG changes with pupillary light response to assess visual pathway integrity.
Main Results:
- Gamma frequency EEG power was slightly lower with eye shields, but not statistically significant after multiple comparisons.
- No significant association was found between EEG responsivity to eye shields and pupillary light response.
- Occipital EEG activity showed minimal change when external visual input was reduced.
Conclusions:
- EEG activity in this sample of acutely unwell infants was insensitive to reduced visual input.
- The majority of visual cortical activity in these infants may be endogenously generated.
- Eye shield interventions are unlikely to modulate visual cortical activity in this population.
Introduction:
Cortical activity is crucial for neural organisation during early life, especially in vulnerable infants who have suffered neurological compromise and/or require intensive care. However, its sensory determinants are poorly understood. In hospital, acutely unwell infants sometimes wear eye shields. We used this natural experiment to test the hypothesis that reduction of external visual inputs will attenuate occipital cortical activity.
Methods:
We analysed 9-channels video-EEG from 18 sleeping neonates (median gestational age 39 weeks (IQR 36-40); 14/18 had mild-moderate hypoxic-ischemic encephalopathy). We compared power spectral density of EEG between eye shield vs. no eye shield, matched by vigilance state, any neuroactive medications, and within four hours (total 31 test occasions). We cross-referenced these data against power spectral changes of EEG to another visual stimulus: pupillary examination with a torch light (18 neonates, 7 overlapping with eye shield analysis: total 29 neonates had visual sensory data analysed).
Results:
Gamma (>30 Hz) but not slower frequency power was slightly lower on average during the eye shield condition - specifically at the occipital region - but this did not survive adjustment for multiple comparisons. There was no association between infants' EEG responsivity to the eye shield and torch light stimuli.
Conclusion:
EEG activity was insensitive to reduction of external visual inputs in this sample of acutely unwell infants, even when integrity of their retina-visual cortex pathway could be confirmed by response to torch light. In this population, the majority of visual cortical activity may be endogenously generated, and therefore not modulated by an eye shield intervention.
