Elsevier

NeuroImage

Volume 118, September 2015, Pages 576-583
NeuroImage

Theta- and alpha-band EEG activity in response to eye gaze cues in early infancy

https://doi.org/10.1016/j.neuroimage.2015.06.042Get rights and content

Highlights

  • Alpha desynchronized more to object-directed eye gaze in 4- and 9-month-old infants.

  • Theta synchronized more to object-averted eye gaze in 5-month-old infants.

  • Theta and alpha mark developmental changes in eye gaze processing in infants.

Abstract

In order to elucidate the development of how infants use eye gaze as a referential cue, we investigated theta and alpha oscillations in response to object-directed and object-averted eye gaze in infants aged 2, 4, 5, and 9 months. At 2 months of age, no difference between conditions was found. In 4- and 9-month-olds, alpha-band activity desynchronized more in response to faces looking at objects compared to faces looking away from objects. Theta activity in 5-month-old infants differed between conditions with more theta synchronization for object-averted eye gaze. Whereas alpha desynchronization might reflect mechanisms of early social object learning, theta is proposed to imply activity in the executive attention network. The interplay between alpha and theta activity represents developmental changes in both kinds of processes during early infancy.

Introduction

From very early on in life, eye gaze is an important cue influencing infants’ perception and attention. As it helps infants to direct their attention to relevant information in the environment, eye gaze direction, among other social cues (Bertenthal et al., 2014), affects information processing and facilitates social learning (Csibra and Gergely, 2006, Hoehl et al., 2009, Reid and Striano, 2007). Here, we measure oscillatory brain activity in response to eye gaze as a referential cue in early infancy.

Infants show an early sensitivity to eye gaze direction in relation to the location of objects. Nine-month-old infants look longer to object-directed gaze shifts than to non–object-directed gaze shifts (Senju et al., 2008). Even younger infants differentiate between object-directed and object-averted eye gaze: event-related potentials (ERPs) in response to faces looking toward objects were compared to those for faces looking away from objects in 2-, 4- and 5-month-olds (Hoehl et al., 2008, Hoehl et al., 2009). Whereas no effects on the negative central (Nc) component were found in the youngest age group, infants at 4 and 5 months showed a larger amplitude for this component for object-averted gaze. As the Nc component is related to attention (Reynolds and Richards, 2005), it was concluded that infants allocated more attention to faces that looked away from objects because this situation was less expected and more ambiguous to them. Moreover, it was only in the 4- and not in the 5-month-olds that a larger positive slow wave (PSW) was found for object-directed looks. The PSW is related to memory updating processes (Nelson, 1997, Webb et al., 2005). Thus, eye gaze may have facilitated building memory representations for cued objects. In the aforementioned cross-sectional approach, the studies by Hoehl et al., 2008, Hoehl et al., 2009 highlight developmental changes in the way infants process eye gaze and its relation to objects.

Similar developmental changes have been revealed by behavioral studies. At 3 months of age, infants are already sensitive to triadic interactions (Striano and Stahl, 2005). Their ability to follow gaze shifts of strangers increases between 4 and 6 months (Gredebäck et al., 2010). At the same time, infants’ joint attention skills gradually develop (Striano and Bertin, 2005) and their ability to use social cues to encode new information advances. In a live paradigm measuring looking times, infants at 7 and 9 months but not at 4 and 5 months of age showed enhanced object processing in a joint attention situation (Cleveland et al., 2007, Cleveland and Striano, 2007). Studies that presented similar stimuli on a screen found that infants were already able to use social cues for object learning at 4 months (Hoehl et al., 2014b, Reid and Striano, 2005, Reid et al., 2004, Wahl et al., 2013). These studies compared ERPs and looking times in response to objects that were previously cued by another person’s eye gaze and/or head turn with objects that were not cued. Cued objects were processed more efficiently whereas uncued objects were more novel to infants when they were presented to the infant a second time. This was reflected in enhanced amplitudes of either the PSW or the Nc as well as in longer looking times to previously uncued objects. Eye gaze cues guided infant attention and thereby facilitated object learning. The age discrepancy between live and video-based studies may be due to the different types of paradigms and dependent variables. A video-based presentation condenses information on a small screen and this may help infants to focus on the stimuli. The setting in a live paradigm is more complex as infants are interacting with a real person who, inevitably, covers more space. Furthermore, the dependent variable in the live studies was the overt behavior of the infant, whereas video-based studies mostly applied ERPs and/or eye tracking.

The aforementioned studies show developmental changes in the way infants make use of social cues. One possible mechanism behind these changes is how infants are able to control their attention. At 4 months of age, infants supposedly react to eye gaze cues due to an automatic shift of attention (Hoehl et al., 2014b, Moore and Corkum, 1998). During the following two months, it has been proposed that an attention network starts to monitor and integrate infants’ own and others’ gaze direction and behavior. Between 7 and 9 months, infants are able to internally control their shifts of attention (Mundy and Newell, 2007, Petersen and Posner, 2012).

As the results of studies investigating the use of social cues differ depending on the paradigm used, the current study makes use of the same paradigm for all age groups in a cross-sectional design with infants aged 2, 4, 5, and 9 months. As in the study by Hoehl et al. (2008), infants saw static images of faces either looking toward or away from an object while their EEG was measured. So far, the neural processing of eye gaze–object relations in infancy has only been investigated using ERPs. In the current study, we analyze oscillatory changes to further clarify underlying neural mechanisms of how social information is processed.

Based on the literature, the alpha- and the theta-band are likely to be sensitive to eye gaze-object relations: theta-band activity in adults lies between 4 and about 7 Hz (Klimesch, 1999, Saby and Marshall, 2012). Theta in infants, that we refer to in the current study, is primarily defined between 3 and 6 Hz and the frequency range does not seem to change between 4 and 12 months (Saby and Marshall, 2012, Stroganova and Orekhova, 2007). Theta synchronization may imply activity of the frontal cortex including an attention network involved in executive and voluntary control of attention as it has been proposed by Posner and Petersen (Bazhenova et al., 2007, Orekhova et al., 1999, Petersen and Posner, 2012, Posner and Petersen, 1990). It has been suggested that this attention system emerges at around 4–6 months and allows infants to monitor the relation between their own and others’ gaze direction and goal-directed behavior (Mundy and Newell, 2007). Frontal theta activity decreases with age. This decrease is proposed to reflect maturation processes in the attention system as the system gets increasingly effective (Orekhova et al., 1999). If theta activity implies executive control of attention, it would be expected to vary with developmental changes in response to social cues. Therefore, we expect to find no differences between conditions in theta synchronization in the 2- and 4-month-olds as the executive attention network should not be developed yet. In 9-month-olds, the network should have matured and be more efficiently functioning (Orekhova et al., 1999). As theta decreases in older infants, we expect little or no difference in theta synchronization between conditions. Changes in theta activity may reflect the development of this system which occurs at around 5–6 months of age and we therefore expect theta effects specifically in this age group.

Alpha desynchronization in adults has been related to attentional mechanisms that actively suppress distracting information to focus on relevant input (Ward, 2003). In a live triadic joint attention interaction, Lachat et al. (2012) reported attenuated alpha signal power (11–13 Hz) in adult participants that jointly attended to the same stimulus. This result was interpreted as reflecting higher arousal induced by mutual attentiveness. Hoehl et al. (2014a) recently showed similar effects in 9-month-old infants in a live paradigm. Here, alpha (5–7 Hz) desynchronized in response to novel objects only when these objects were presented in a joint attention situation (Hoehl et al., 2014a), indicating that alpha-band activity varied depending on the social context in which stimuli were perceived. Alpha desynchronization was therefore suggested to relate to early social learning processes in infants (Hoehl et al., 2014a). Enhanced alpha desynchronization may indicate that attention is focused on the relevant object (here an object that is cued by eye gaze). Thereby it could enable or at least facilitate object learning in such situations. Similar processes might already take place at 4 months as infants differentiate between eye gazes toward and away from objects and build stronger memory representations for cued objects (Hoehl et al., 2008, Hoehl et al., 2014b, Reid and Striano, 2005, Reid et al., 2004, Wahl et al., 2013). In the current study, eye gaze that is directed toward an object identifies it as an object that is of high relevance for the infant. Thus, we expect desynchronization to occur in response to object-directed gaze starting at 4 months of age in the alpha-band frequency range 4–10 Hz, which is the typical range for alpha in infants (Marshall et al., 2002, Stroganova et al., 1999).

The current study investigates oscillatory brain activity in response to object-directed and object-averted eye gaze for synchronization in the theta range and for desynchronization in the alpha range. By studying 2-, 4-, 5-, and 9- month-old infants with the same paradigm, we expect to gain insights into how the processing of social cues develops and how attentional and social information processes change in early infancy (Cleveland et al., 2007, Cleveland and Striano, 2007, Striano and Stahl, 2005).

Section snippets

Participants

The final sample consisted of 58 (32 female) 2-, 4-, 5-, and 9-month-old infants born full term (37–41 weeks) and within the normal range for birth weight (see Table 1 for detailed information about age, sex, and the number of trials included in the final analyses separately for each age group).

Another 79 infants were tested but excluded from the final sample due to fussiness (17) or failure to reach the minimum criterion of 10 artifact-free trials per condition (62). This inclusion criterion

Theta

No significant differences between conditions were found for the 2-, 4-, and 9-month-olds, all ps > .431. However, the object-averted condition and the object-directed condition differed significantly in the 5-month-olds, t (15) =  3.50, p = .012. Theta synchronized more in the object-averted compared to the object-directed condition. Theta activity in both conditions did not differ from baseline, all ps > .195. See Fig. 2 and Table 2 for means and standard errors.

Alpha

Whereas alpha activity in the

Discussion

In order to investigate developmental changes in neural mechanisms underlying the processing of eye gaze-object relations in early infancy, we presented infants (2, 4, 5, and 9 months old) with faces that were either looking away from or toward objects while EEG was measured. Differences between conditions in the theta and the alpha frequency bands were investigated for each age group. In line with studies showing that 4  8-month-old infants differentiate between object-directed and

Acknowledgments

We are grateful to the children and parents who participated. This work was supported by a grant from the Deutsche Forschungsgemeinschaft (DFG) [grant number HO 4342/2-2] awarded to Stefanie Hoehl and a stipend from the Studienstiftung des deutschen Volkes awarded to Christine Michel. Eugenio Parise and Vincent Reid are researchers in the International Centre for Language and Communicative Development (LuCiD) at Lancaster University. The support of the Economic and Social Research Council [

References (45)

  • T.A. Stroganova et al.

    Externally and internally controlled attention in infants: An EEG study

    Int. J. Psychophysiol.

    (1998)
  • T.A. Stroganova et al.

    EEG alpha rhythm in infants

    Clin. Neurophysiol.

    (1999)
  • L.M. Ward

    Synchronous neural oscillations and cognitive processes

    Trends Cogn. Sci.

    (2003)
  • B.I. Bertenthal et al.

    When do infants begin to follow a point?

    Dev. Psychol.

    (2014)
  • A. Cleveland et al.

    Joint attention and object learning in 5- and 7-month-old infants

    Infant Child Dev.

    (2007)
  • G. Csibra et al.

    Social learning and social cognition: The case for pedagogy

  • G. Csibra et al.

    Gamma oscillations and object processing in the infant brain

    Science

    (2000)
  • C.K. Friesen et al.

    Attentional effects of counterpredictive gaze and arrow cues

    J. Exp. Psychol. Hum. Percept. Perform.

    (2004)
  • A. Frischen et al.

    Gaze cueing of attention: Visual attention, social cognition, and individual differences

    Psychol. Bull.

    (2007)
  • G. Gredebäck et al.

    The development of joint visual attention: A longitudinal study of gaze following during interactions with mothers and strangers

    Dev. Sci.

    (2010)
  • S. Hoehl et al.

    What are you looking at? Infants' neural processing of an adult's object-directed eye gaze

    Dev. Sci.

    (2008)
  • S. Hoehl et al.

    Looking at eye gaze processing and its neural correlates in infancy—Implications for social development and autism spectrum disorder

    Child Dev.

    (2009)
  • Cited by (24)

    • Decoding functional brain networks through graph measures in infancy: The case of emotional faces.

      2022, Biological Psychology
      Citation Excerpt :

      The alpha band has been mostly investigated on its premotor and parietal configurations, encompassing infants’ mu rhythm desynchronization recorded in response to the execution and perception of others’ actions (Quadrelli, Geangu & Turati 2019b; Saby & Marshall, 2012; Southgate, Johnson, Osborne & Csibra, 2009), or facial emotional expressions (Quadrelli, Roberti, Turati & Craighero, 2019c; Quadrelli, Roberti, Polver, Bulf, & Turati, 2021; Rayson, Bonaiuto, Ferrari & Murray, 2016; Rayson, Bonaiuto, Ferrari & Murray, 2017). The theta band also appears well characterized during infancy: it has been mostly linked to social stimulation, such as the expression of both positive and negative emotions (Bazhenova, Stroganova, Doussard-Roosevelt, Posikera & Porges, 2007; Saby & Marshall, 2012; Michel et al., 2015), and to attentional modulation (Orekhova, Stroganova, & Posikera, 1999). In the latter case, theta activity was reportedly greater during attentional anticipatory periods, a result consistent with adult literature, in which frontal theta activity has been associated with more demanding tasks (Saby & Marshall, 2012).

    • Attention: a prerequisite for learning

      2022, International Encyclopedia of Education: Fourth Edition
    • Dynamic modulation of frontal theta power predicts cognitive ability in infancy

      2020, Developmental Cognitive Neuroscience
      Citation Excerpt :

      Participants were only included if they provided at least five artefact-free segments per first and second 30 s of this period. Inclusion criteria for both indices are in line with previous work (Michel et al., 2015; Elsabbagh et al., 2009; Southgate et al., 2008). Power values were averaged across artefact-free segments and electrodes within a fronto-central topographical group (see Figure S1) and within each of the first and second 30 s of the video.

    View all citing articles on Scopus
    View full text