Elsevier

Brain and Development

Volume 43, Issue 2, February 2021, Pages 186-191
Brain and Development

Original article
Do low birth weight infants not see eyes? Face recognition in infancy

https://doi.org/10.1016/j.braindev.2020.09.002Get rights and content

Abstract

Background

Progress in neonatal medicine has dramatically improved the survival rate of preterm births, but the evidence suggests that these low-birth weight infants (LBWIs) go on to develop pervasive development disorders and attention deficit hyperactivity disorder (ADHD) at greater rates than the general population. Children with neurodevelopmental disorders are known to suffer from deficits in visual cognition, such as in face perception and attentional functions, the characteristics of which already manifest in early infancy.

Purpose

This study aimed to investigate visual cognition in LBWIs during infancy.

Subjects

20 LBWIs and 20 normal-birth-weight infants (NBWIs: control) of age 9–10 months (corrected age was used for LBWIs).

Method

Children were held seated in front of an eye tracking system by a parent, and presented with facial photos as visual stimuli. During the familiarization phase, the child was presented with two images of the same human face (familiarization stimulus) on the left and right side of a display screen (5 × 10 s trials). Next, during the test phase, the child was presented with the same image on one side of the screen, and a photo of a different person’s face (novel stimulus) on the other (2 × 5 s trials). Gaze behavior was assessed in terms of the total time spent looking at either facial stimulus, and specifically at the eyes of the stimuli, as well as the number of attentional shifts between stimuli, and novelty preference.

Results/Discussion

LBWIs spent significant less time looking at facial stimuli overall, and less time at the eye region, than NBWIs. These findings seem to evidence developmental differences in functions related to visual cognition.

Introduction

While recent progress in neonatal medicine has markedly increased the survival rate of low-birth-weight infants (LBWIs), survivors reportedly develop a host of neurological and developmental conditions, such as autism spectrum disorder (ASD) and attention deficit hyperactivity disorder (ADHD), at higher rates than their normal-birth-weight infants (NBWIs) [1], [2]. These include, characterized by social communication difficulties and limited, stereotypic patterns of behavior, and, characterized by inattention and hyperactivity/impulsivity. These behaviors are believed to originate in dysfunctional processing of visual information in domains such as face perception and attentional functions.

Studies have demonstrated that individuals with ASD have trouble recognizing faces and understanding facial expressions [3], [4], [5], [6], [7], and that autistic children exhibit such deficits in facial recognition skills even before verbal communication difficulties, stereotypy, and other characteristic features of the disorder become apparent [7]. The dysfunctional recognition of facial expressions in infants with ASD has been attributed to their utilization of a “face scanning” technique different from that employed by normally developing infants [8]. One study reported that autistic infants tend not to look into the eyes of facial stimuli at two years of age [9]; another has claimed they give preferential attention to the mouth over the eyes [10]. Similarly, Klin and colleagues have observed associations between gaze patterns and sociality [11].

The idea that impaired executive function lies at the root of the core symptoms of ADHD is nothing new, having been purported long ago by Barkley et al. [12]. Executive functions are a collection of processes used to filter and sort important information, guiding and implementing behaviors towards achieving some goal. One subset of these, crucial to normal cognition, are the attentional functions, defined as a subset of processes that ensure selectivity in conscious activity by judging the relative importance of competing information, establishing and following an accurate and organized plan of action, and continuously regulating this sequence in the course of its execution [13], [14].

Face perception and attentional processes have been demonstrated to be already active in infancy, with Haxby et al. reporting the same for core neurological systems for processing visual information related to faces: notably, constantly changing factors such as facial expressions and eye and mouth movements [15]. Regions involved in attention such as the dorsolateral prefrontal, anterior cingulate, and orbitofrontal cortices begin to function around six months after birth, while higher-level attentional functions (e.g. sustained and selective attention) start to engage around the same period [16]. Children’s performance on attention-related tasks in infancy is known to relate with their later executive functioning: Cuevas et al. reported that infants who more efficiently processed information about stimuli at five months of age exhibited superior executive functioning throughout early childhood [17].

Eye trackers are a type of recently developed device, utilized in a variety of fields, which can be used to monitor a subject’s gaze easily and highly accurately. Infrared light is shined from the device into a subject’s eyes, and their line of sight determined by sensor based on its reflection off their corneas. Eye trackers have made it possible to assess visual cognition in an objective way: in fact, multiple studies have reported their usefulness in detecting neurodevelopment disorders in infancy [18], [19], [20]. Konishi et al.’s research on visual cognitive functions in both full- and pre-term infants at three to four months of age has found that at the corrected age of four months, premature infants focus on different areas of the face from full-term infants, evidencing developmental differences in facial recognition [21], [22]. One might expect that such differences between pre-term and full-term children grow more and more obvious as they grow and develop, yet few studies to date have actually compared the eye movements and gaze patterns of these two groups in a detailed way [23].

Using gaze trackers to identify aspects of visual cognition in infancy typical of preterm infants, and determining how these patterns relate to subsequent manifestation of neurodevelopmental disorders, would help to inform the development of early discovery and early intervention approaches this at-risk population.

This paper details the findings of our investigation of attention and facial perception in LBWIs during infancy, in which the eye movements of LBWIs and NBWIs of corrected age 9–10 months towards images of faces were measured and analyzed in the context of a novelty preference paradigm.

Section snippets

Subjects

LBWIs seen at the Pediatrics Department of the Kagawa University Hospital were enrolled between December 2016 and September 2017, and tested at the corrected age of 9–10 months. NBWIs (control group) were recruited between April 2017 and April 2018, from children aged 9–10 months brought in for health checkups in the town of Shodoshima, Kagawa Prefecture. This study was approved by the Ethics Committee of the Kagawa University School of Medicine (accession no. 2018-025). The study overview was

Subjects

We obtained consent from 20 LBWIs and 22 NBWIs. Two infants of NBWIs were not able to look at the screen for intense crying. Data was analyzed for 20 preterm LBWIs (male:female ratio: 10:10, birth weight: 505–2087 g, gestation length: 23.4–36.9 weeks) and 20 full-term NBWIs (9:11, 2500–4000 g, 37.0–41.6 weeks). All participating children were first examined by a pediatrician, and confirmed to lack neurological abnormalities. LBWIs and NBWIs had a mean post menstrual age (PMA) of 569 ± 12.4 and

Discussion

Our findings demonstrate that preterm and full-term infants react differently to facial stimuli in early infancy. First, the gaze time of LBWIs was significantly shorter than a term infant in familialization phase. Also, the eye gaze rate in the familialization phase was significantly low in LBWIs. Whereas, as for the gaze time and the eye gaze rate in the test phase, there were no significant differences in both groups. In addition, LBWIs shifted their attention between the facial images less

Conclusion

LBWIs spent significantly less time looking at facial stimuli, as well as the eye region specifically (in terms of both absolute time and as proportion of total gaze time), in a novelty preference task in late infancy. These observations seem to reflect differences in attentional processes and facial cognition between preterm and full-term infants at an early age. Reluctance to look in the eyes is a characteristic also seen in autistic children; this trait may predict future development of ASD.

Conflict of interest disclosures

The authors declare no competing interests.

References (25)

  • A. Klin et al.

    A normed study of face recognition in autism and related disorders

    J Autism Dev Disord

    (1999)
  • K. Chawarska et al.

    Impairments in monkey and human face recognition in 2-year-old toddlers with Autism Spectrum Disorder and Developmental Delay

    Dev Sci

    (2007)
  • View full text