Elsevier

Neuropsychologia

Volume 84, April 2016, Pages 89-104
Neuropsychologia

Longitudinal study of preterm and full-term infants: High-density EEG analyses of cortical activity in response to visual motion

https://doi.org/10.1016/j.neuropsychologia.2016.02.001Get rights and content

Highlights

  • With acquired mobility, infants depend on accurate pick-up of visual information

  • Development of visual perception in (pre)term infants studied with high-density EEG

  • Changes in brain electrical activities were observed in VEP and induced EEG

  • Infants rely on perception of structured optic flow to move around efficiently

  • Preterm infants show poorer responses to visual flow compared to full-term infants

Abstract

Electroencephalogram (EEG) was used to investigate brain electrical activity of full-term and preterm infants at 4 and 12 months of age as a functional response mechanism to structured optic flow and random visual motion. EEG data were recorded with an array of 128-channel sensors. Visual evoked potentials (VEPs) and temporal spectral evolution (TSE, time-dependent amplitude changes) were analysed. VEP results showed a significant improvement in full-term infants' latencies with age for forwards and reversed optic flow but not random visual motion. Full-term infants at 12 months significantly differentiated between the motion conditions, with the shortest latency observed for forwards optic flow and the longest latency for random visual motion, while preterm infants did not improve their latencies with age, nor were they able to differentiate between the motion conditions at 12 months. Differences in induced activities were also observed where comparisons between TSEs of the motion conditions and a static non-flow pattern showed desynchronised theta-band activity in both full-term and preterm infants, with synchronised alpha-beta band activity observed only in the full-term infants at 12 months. Full-term infants at 12 months with a substantial amount of self-produced locomotor experience and neural maturation coupled with faster oscillating cell assemblies, rely on the perception of structured optic flow to move around efficiently in the environment. The poorer responses in the preterm infants could be related to impairment of the dorsal visual stream specialized in the processing of visual motion.

Introduction

Most daily activities including social interactions and cognitive skills such as object categorisation, reading, and navigating through the environment depend on perception and correct interpretation of visual information. Considering the relevance of these abilities to everyday life, it is important to understand the developmental processes underlying how infants learn to make use of relevant visual information for perception.

During daily activities, it is important to recognise the landmarks in the environment and their egocentric positions in defining how relative an object's position is to an observer and vice versa (Wall and Smith, 2008, Warren, 1976). The concept where major images emanate from a central point in a structured form within the angle of sight when there is a change of scenery either by locomotion in any direction or as an object approaches an observer is referred to as radial optic flow (Gibson, 1979). The visual motion perception that is achieved by this change is crucial for adjusting posture (Vaina and Rushton, 2000), perceiving time-to-contact (Kayed and Van der Meer, 2009), avoiding obstacles (Turano et al., 2005, Wilkie and Wann, 2003), and reaching a target efficiently by determining heading direction (Lappe et al., 1999). The relation between optic flow perception and action has often been studied in terms of motor activities (e.g., Bruggeman et al., 2007). It has been shown that perception of optic flow plays a major role in the control of walking speed and direction (Bruggeman and Warren, 2010, Lamontagne et al., 2007, Vilhelmsen et al., 2015, Warren et al., 2001). The estimation of speed in order to effectively reach a target is achieved by integrating inputs from the visual system with locomotor variables (Saleem et al., 2013). Improvement in visual perception by actual motor learning through action-to-perception transfer has been reported in adults (Beets et al., 2010). Young infants, on the other hand, are not as efficient as adults in detecting optic flow (Gilmore et al., 2007, Gilmore et al., 2004, Van der Meer et al., 2008).

Rudimentary perception of optic flow patterns appears within the first weeks and months after birth (Gilmore et al., 2004), with adults able to readily distinguish between expansion/contraction, rotational, and translational (horizontal/vertical) forms of motion (Gilmore et al., 2007). Behaviourally, infants younger than a month of age exhibit defensive responses such as backward head tilting and frequent eye blinking in reaction to structured motion, with even neonates as young as 3 days old observed to exhibit responses through backward head movements when exposed to backwards flow stimuli (Jouen et al., 2000, Shirai and Yamaguchi, 2010). These defensive behavioural responses in very young infants, however, have been interpreted to be the result of multimodal integrative and cooperative processes in which visual, vestibular, and proprioceptive senses are involved rather than a direct consequence of motion perception (Jouen et al., 2000). Further studies using electroencephalogram (EEG) have shown that infants younger than 6–8 weeks do not discriminate between motion directions and do not smoothly pursue small moving objects but such abilities improve rapidly around 6–14 weeks after birth (Rosander et al., 2007). These motion patterns show unique developmental trends with age in infants (Shirai and Yamaguchi, 2010). Here, perception of translational motion emerges at approximately two months of age whereas perception of radial motion and its cortical responses develop rapidly afterwards from just before 3 months until approximately 4 months of age (Shirai and Yamaguchi, 2010). However, this fast perceptual development starts decelerating after the first few months but approaches adult level around 4 years of age (Kaufmann, 1995).

Various brain studies in humans and other primates have investigated the cerebral networks specialized for perception of visuo-spatial information (for review, see Shirai and Yamaguchi (2010)). In particular, perception of movement control including self- and object-motion is processed via the dorsal visual stream (see Creem and Proffitt (2001)). Neurons within the medial temporal (MT/V5+) complex are generally sensitive to radial motion processing, with the dorsal medial superior temporal (dMST) area involved specifically in optic flow processing (Creem and Proffitt, 2001, Duffy and Wurtz, 1991, Dukelow et al., 2001, Fukushima, 2008, Greenlee, 2000, Imura et al., 2008, Itoh et al., 2005, Newsome and Pare, 1988, Riecansky, 2004, Smith et al., 2006, Tohyama and Fukushima, 2005, Wall et al., 2008).

The human brain is an organised dynamic network of interconnected neurons and associated synapses that work together such that dysfunctions within the network can have adverse effects on behavioural patterns. Magnetic resonance imaging (MRI) studies show that being born preterm causes differential brain development that leads to abnormalities in the microstructure of tissues and in cerebral morphology (Counsell and Boardman, 2005, Counsell et al., 2006). Premature birth with its continuously increasing incidence (Martin et al., 2013, Martin et al., 2003) has therefore been given considerable attention over the years. Some of the dysfunctions of preterm birth have been related to cognitive and behavioural impairments (Aarnoudse-Moens et al., 2009, Bhutta et al., 2002, De Jong et al., 2012, Delobel-Ayoub et al., 2009, Johnson, 2007, Salt and Redshaw, 2006) including deficits in visual perception and other neurodevelopmental disorders (see De Jong et al. (2012)).

Some studies have reported deficit in motion and form perception in premature children that was either a consequence of being born preterm (MacKay et al., 2005, Taylor et al., 2009) or a result of the accompanying periventricular leukomalacia (PVL) (Downie et al., 2003, Jakobson et al., 2006). Using behavioural experiments, further studies have shed more light on the deficit in motion and form perception where differences between global motion (optic flow) perception and form perception have been found (Guzzetta et al., 2009). They reported that preterm children with and without brain damage appeared to perform worse than full-term controls in global motion perception, with PVL preterm infants performing worse on form perception than preterm infants without PVL (Guzzetta et al., 2009). These observations suggested impairment of the dorsal stream during visual processing in preterm children with and without brain damage. In four cohort studies in which several assessment techniques were used and the correlations between the results and MRI indicators of brain abnormalities were analysed, it was also pointed out that the impairment of the visual dorsal stream is a possible cause for the cluster of developmental problems seen in children born prematurely (Atkinson and Braddick, 2007). These findings are further supported by other studies where global motion, global form, and biological motion (combination of cues for form and motion) perception were compared in preterm children (e.g., Taylor et al., 2009). Global motion perception deficit was observed more densely than global form deficit, and difficulties in biological motion perception were also present (cf., Atkinson et al., 2008a, Atkinson et al., 2008b). Further studies (e.g., Van Braeckel et al., 2010) that investigated the motor skills of preterm children of 7–11 years report that they perform less accurately or slower in movement pointing tasks, implying less efficient elementary visuomotor processing and impaired functioning of the dorsal visual stream. Other studies with the aim to help detect early brain damage in premature infants have further formulated specific markers of visual cortical function in the first postnatal months for both normally and atypically developing infants (Atkinson et al., 1992, Atkinson et al., 2008a, Atkinson et al., 2008b, Braddick et al., 1992, Braddick et al., 2005). Such studies showed that the severity of detected brain injury in preterm infants was correlated with impaired orientation-reversal visual event-related potentials and the cortical control of visual attention through fixation shifting. These studies also showed preterm infants to display worse performances on given cortical tests compared to full-term infants.

During visual perception tasks, EEG primarily records electrical activities of pyramidal neurons with high temporal resolution (in the millisecond scale) that permits the study of the neuronal basis of motion perception and the functional specialisations of cortical structures (Agyei et al., 2015, Dukelow et al., 2001, Sakkalis et al., 2008, Tucker, 1993). Motion-sensitive visual evoked potential (VEP) waveforms in EEG are thought to originate from V5/MT+ in occipital and parietal visual areas (Kobayashi et al., 2004, Wall et al., 2008), and are characterized by a negative N2 peak component that occurs around 270–290 ms post stimulus in 11–12-month-old infants (Agyei et al., 2015) and around 130–200 ms in adults (Fielder et al., 1983, Heinrich et al., 2005). Previous studies reported the N2 visual component in infants to be longer in latency and higher in amplitude compared to adults (Van der Meer et al., 2008), with the longer N2 latencies improving with infants' age (Agyei et al., 2015). Other studies of early visual N1, P1 and P2 components (e.g., Fielder et al., 1983, Hammarrenger et al., 2007) have shown similar results where latencies were observed to decrease as children became older.

In addition to the use of VEPs in EEG studies, investigation of neural activities in the time-frequency domain has allowed for the study of cognitive and perceptual functions through observations of the natural frequencies in EEG (Agyei et al., 2015, Basar et al., 1999, Formaggio et al., 2010, Ganzetti and Mantini, 2013, Salari et al., 2012). Analyses in the time-frequency domain permit the extraction of event-related frequency changes in EEG that cannot be observed in VEPs by using averaging techniques. These frequency changes are observed as event-related (de)synchronisation (ERD/ERS) oscillatory activities that represent decreases/increases in amplitude/power of specific frequency bands (Pfurtscheller and Lopes da Silva, 1999), an indication of decrease/increase in synchrony of the underlying neuronal populations (Hoechstetter et al., 2004, Pfurtscheller et al., 1994). Based on spectral profiles within specific frequency bands, different classes of oscillations have been distinguished over the years: delta-band (1–4 Hz), theta-band (4–7 Hz), alpha-band (7–13 Hz), beta-band (13–30 Hz), and gamma-band (30–150 Hz), with these rhythms shown to reflect neurophysiological processes that manifest functionally different roles (Buzsaki and Draguhn, 2004, Engel and Fries, 2010, Ganzetti and Mantini, 2013, Saby and Marshall, 2012). For example, while delta-band oscillations play a role in processes such as signal detection, event-related oscillations in the theta-band have been reported to play an important role in cognitive processes (Başar et al., 2000, Freunberger et al., 2011, Gruber and Müller, 2006, Khader et al., 2010, Klimesch et al., 1996). Event-related synchronisations in the alpha-band may reflect control of inhibition and cortical processing (Doppelmayr et al., 2005, Doppelmayr et al., 1998, Freunberger et al., 2011, Klimesch, 1999, Klimesch et al., 2007, Sauseng and Klimesch, 2008). Further studies with visual stimuli (e.g., Wróbel, 2000) have linked beta oscillatory responses to multisensory stimulation and the shifting of neural systems to a state of attention that ultimately allows for gamma synchronization and the utilisation of the resulting information for perception (Herrmann et al., 2010, Herrmann et al., 2004). Particularly in infants, low-frequency oscillations especially in the theta-alpha range undergo systematic development from early childhood to adulthood (Stroganova and Orekhova, 2007, Stroganova et al., 1999). These low-frequency rhythms in infants have been attributed to general signs of immaturity (Orekhova et al., 2006, Stroganova and Orekhova, 2007). Recent infant studies have implicated low-amplitude desynchronised theta-band activities over visual areas when motion stimuli were compared with static stimuli (Agyei et al., 2015, Van der Meer et al., 2008), with further observation of high-amplitude activities especially in alpha-band frequency observed in one-year-old infants (Agyei et al., 2015). This transition of EEG spectral power/amplitudes from lower to higher frequencies during development is considered a sign of maturation in various psychophysiological studies (Hudspeth and Pribram, 1992, Stroganova et al., 1999).

The present study explores the development of the visuo-cognitive systems, especially visual motion perception during early infancy. The study presents information that contributes to the understanding of the normal development of visual motion perception in infants and the developmental impairments associated with motion perception in preterm birth. Using EEG longitudinal data, the paper compares preterm and full-term infants at the ages of 4 months and 12 months through a combination of VEP and time-frequency analyses. VEPs were assumed to represent responses of cortical neurons to changes in afferent activity (Brecelj, 2003, Van der Meer et al., 2008), while event-related time-frequency responses represent interactions of local cortical neurons that control the frequency components of an ongoing EEG (Pfurtscheller and Lopes da Silva, 1999). It was expected that perception of optic flow would rapidly improve with increasing age especially in the full-term infants, with no such marked improvement in perception of random visual motion, and that low-frequency decrease in amplitudes would generally be observed.

Section snippets

Participants

Twenty full-term and preterm normally developing infants were recruited for the study through birth announcements, contact with parents and/or through the Neonatal Intensive Care Unit at St. Olav's University Hospital in Trondheim. The full-term infants had mean gestational age of 39.9 weeks (SD=0.9, median=39.6, range=38.4–41) and mean birth weight of 3693 g (SD=628, median=3500, range=3085–5120). The preterm infants (moderate to very preterm) were born at mean gestational age of 31.4 weeks

VEP responses

For each group and each testing session, four posterior electrodes that showed the highest N2 amplitude values in the forwards optic flow condition of the grand average VEPs (Fig. 3) were chosen for further analysis (cf., Agyei et al., 2015, Van der Meer et al., 2008). The four grand average electrodes for full-term infants at 4 months were PO4, PO3, POz, and O1, and those at 12 months were PO4, PO8, Oz, and O2. The corresponding electrodes for preterm infants at 4 months were POz, PO4, Oz, and

Discussion

The present longitudinal study investigated the perception of visual motion information during early infancy, and the differences in visual motion perception between normally developing full-term and preterm infants. High-density EEG was used to observe cortical electrical activity as a perceptual response to structured radial optic flow (forwards and reversed) and to random visual motion. Motion specific N2 components of the VEP waveforms for 4- and 12-month-old full-term and preterm infants

Acknowledgements

This project was partly made possible by the Norwegian ExtraFoundation for Health and Rehabilitation. The authors are grateful to Broen van Besien and Eirik Paulsen for programming the visual stimuli, and to the parents and their infants for taking part in the study. They are also grateful to Belde Mutaf for her contribution during her Master's thesis, and to Magnus Holth and Kenneth Vilhelmsen for their help with testing, as well as the neonatologist at St. Olav's University Hospital, Ragnhild

References (164)

  • F. De Pasquale et al.

    A cortical core for dynamic integration of functional networks in the resting human brain

    Neuron

    (2012)
  • M. Doppelmayr et al.

    Intelligence related differences in EEG-bandpower

    Neurosci. Lett.

    (2005)
  • T.C. Ferree et al.

    Scalp electrode impedance, infection risk, and EEG data quality

    Clin. Neurophysiol.

    (2001)
  • E. Formaggio et al.

    Brain oscillatory activity during motor imagery in EEG-fMRI coregistration

    Magn. Reson. Imaging

    (2010)
  • R. Freunberger et al.

    Brain oscillatory correlates of working memory constraints

    Brain Res.

    (2011)
  • P. Fries

    A mechanism for cognitive dynamics: neuronal communication through neuronal coherence

    Trends Cogn. Sci.

    (2005)
  • P. Fries et al.

    Finding gamma

    Neuron

    (2008)
  • T. Fujioka et al.

    Comparison of artifact correction methods for infant EEG applied to extraction of event-related potential signals

    Clin. Neurophysiol.

    (2011)
  • K. Fukushima

    Extraction of visual motion and optic flow

    Neural Netw.

    (2008)
  • M. Ganzetti et al.

    Functional connectivity and oscillatory neuronal activity in the resting human brain

    Neuroscience

    (2013)
  • C.D. Gilbert et al.

    Brain states: top-down influences in sensory processing

    Neuron

    (2007)
  • M.W. Greenlee

    Human cortical areas underlying the perception of optic flow: brain imaging studies

    Int. Rev. Neurobiol.

    (2000)
  • P.G. Grieve et al.

    Spatial correlation of the infant and adult electroencephalogram

    Clin. Neurophysiol.

    (2003)
  • T. Gruber et al.

    Oscillatory brain activity in the human EEG during indirect and direct memory tasks

    Brain Res.

    (2006)
  • S.P. Heinrich et al.

    Pattern specificity of human visual motion processing

    Vis. Res.

    (2005)
  • C.S. Herrmann et al.

    Human gamma-band activity: a review on cognitive and behavioral correlates and network models

    Neurosci. Biobehav. Rev.

    (2010)
  • C.S. Herrmann et al.

    Cognitive functions of gamma-band activity: memory match and utilization

    Trends Cogn. Sci.

    (2004)
  • N. Hoogenboom et al.

    Localizing human visual gamma-band activity in frequency, time and space

    Neuroimage

    (2006)
  • W.J. Hudspeth et al.

    Psychophysiological indices of cerebral maturation

    Int. J. Psychophysiol.

    (1992)
  • L.S. Jakobson et al.

    Motion-defined form processing in extremely premature children

    Neuropsychologia

    (2006)
  • S. Johnson

    Cognitive and behavioural outcomes following very preterm birth

    Semin. Fetal Neonatal Med.

    (2007)
  • F. Jouen et al.

    Optic flow sensitivity in neonates

    Infant Behav. Dev.

    (2000)
  • G.B. Keller et al.

    Sensorimotor mismatch signals in primary visual cortex of the behaving mouse

    Neuron

    (2012)
  • P.H. Khader et al.

    Theta and alpha oscillations during working-memory maintenance predict successful long-term memory encoding

    Neurosci. Lett.

    (2010)
  • P. Klaver et al.

    Neurodevelopment of the visual system in typically developing children

    Prog. Brain Res.

    (2011)
  • W. Klimesch

    EEG alpha and theta oscillations reflect cognitive and memory performance: a review and analysis

    Brain Res. Rev.

    (1999)
  • W. Klimesch et al.

    EEG alpha oscillations : the inhibition – timing hypothesis

    Brain Res. Rev.

    (2007)
  • Y. Kobayashi et al.

    Perception of apparent motion in depth: a high-density electrical mapping study in humans

    Neurosci. Lett.

    (2004)
  • M. Lappe et al.

    Perception of self-motion from visual flow

    Trends Cogn. Sci.

    (1999)
  • L. Lejeune et al.

    Responsiveness to terrestrial optic flow in infancy: does locomotor experience play a role?

    Hum. Mov. Sci.

    (2006)
  • C.S.H. Aarnoudse-Moens et al.

    Meta-analysis of neurobehavioral outcomes in very preterm and/or very low birth weight children

    Pediatrics

    (2009)
  • S.B. Agyei et al.

    Longitudinal study of perception of structured optic flow and random visual motion in infants using high-density EEG

    Dev. Sci.

    (2015)
  • S.B. Agyei et al.

    Development of visual motion perception for prospective control: brain and behavioral studies in infants

    Front. Psychol.

    (2016)
  • Atkinson, J., Braddick, O., 2007. Visual and visuocognitive development in children born very prematurely. In: von...
  • J. Atkinson et al.

    Cortical vision, MRI and developmental outcome in preterm infants

    Arch. Dis. Child. Fetal Neonatal Ed.

    (2008)
  • J. Atkinson et al.

    High-density VEP measures of global form and motion processing in infants born very preterm

    J. Vis.

    (2008)
  • J. Atkinson et al.

    Changes in infants' ability to switch visual attention in the first three months of life

    Perception

    (1992)
  • N. Auestad et al.

    Growth and development in term infants fed long-chain polyunsaturated fatty acids: a double-masked, randomized, parallel, prospective, multivariate study

    Pediatrics

    (2001)
  • T. Auranen

    Nonparametric Statistical Analysis of Time-Frequency Representation of Magnetoencephalographic Data

    (2002)
  • E. Basar et al.

    Oscillatory brain theory: a new trend in neuroscience

    IEEE Eng. Med. Biol. Mag.

    (1999)
  • Cited by (24)

    • Interweaving social and manipulative development in early infancy: Some direction for infant caregiving

      2021, Infant Behavior and Development
      Citation Excerpt :

      Libertus and Violi (2016) reported the development of posture skills may alter infant's language learning. Self-produced locomotion experience influences infants’ visual perception of optic flow (Agyei, van der Weel, & van der Meer, 2016). Walle and Campos (2014) examined the effects of walking acquisition on social development and parent's perception of their child.

    • 14 challenges and their solutions for conducting social neuroscience and longitudinal EEG research with infants

      2020, Infant Behavior and Development
      Citation Excerpt :

      However, when applied to infants, the artefact exclusion approach may entail considerable data loss (Fujioka et al., 2011), and there is an increasing interest in correction procedures that permit the accurate identification and removal of artefacts from infant EEG without significantly compromising the integrity of underlying neural activity (Fujioka et al., 2011; Gwin, Gramann, Makeig, & Ferris, 2010; He, Hotson, & Trainor, 2007). Yet, while various methods can ameliorate the problem of data loss due to artefact rejection, and they have been used in a number of research studies as a cleaning strategy for data pre-processing (Agyei, van der Weel, & van der Meer, 2016; Corrigall & Trainor, 2014; Folland, Butler, Payne, & Trainor, 2014; Slugocki & Trainor, 2014; Trainor, Lee, & Bosnyak, 2011), these strategies still rely on the successful classification of portions of the EEG signal as artefactual, which is itself a non-trivial task. In order to assess whether the artefactual activity is likely to inflate or reduce the effect that researchers are testing, systematic control analyses should be carried out on how specific movement artefacts distort infant EEG data.

    View all citing articles on Scopus
    View full text