Parents of Children with Cleft Lip Exhibit Heightened Visual Attention to the Perioral Area

Abstract

Background: 

Following high-quality surgical repair, children born with a cleft lip anomaly may still display lasting visual differences. We exposed control adults and parents of affected children to images of children with cleft deformity and compared their visual tracking patterns. The protocol investigated whether parental exposure to secondary cleft deformity heightens or diminishes visual attraction to this type of structural facial variation.

Method: 

Twenty participants (10 control adults, 10 parents of affected children) assessed 40 colored images of children's faces while their eye movements were tracked. Twenty-four control images and 16 repaired cleft lip images were displayed to observers. Nine bilateral facial aesthetic zones were considered as regions of interest. Percentage of time visually fixating within each region, and statistical differences in fixation duration percentage between the two participant groups and across the bilateral regions of interest were analyzed.

Results: 

While both groups of observers directed more visual attention to the nasal and oral regions of the cleft images than control images, parents of children with cleft lip spent significantly more time fixating on these areas (25% and 24% of the time, respectively) than did unaffected adults (14.6% and 19.3%; P < 0.001).

Conclusions: 

These results demonstrate that parents of cleft lip children exhibit heightened attention to this type of facial difference relative to the naive observer. These findings highlight that observer profile can meaningfully influence the perception of a facial deformity. Awareness of this information may enhance communication between surgeon and parents of an affected child by providing added insight into parental perspective.

Takeaways

Question: Are parents of children with cleft lip deformity sensitized or de-sensitized to cleft deformity?

Findings: When observing a combination of normal and affected faces, parents of children with cleft lip exhibit a heightened attention to this type of facial difference relative to the naive observer.

Meaning: An ideal perspective for the cleft surgeon would be to see a child’s face simultaneously through both the eyes of an expert clinician and those of the parents, to facilitate a treatment plan that best incorporates those unique frames of reference. The findings here provide added insight for the treating surgeon when managing cleft lip deformity.

INTRODUCTION

When observing a face, individuals focus primarily on central discriminating features such as the eyes, nose, and mouth.¹ Faces that are disfigured in some way are visually perceived differently than unaffected faces. This difference exists because structural outliers generally attract an observer’s visual attention. Cleft lip (CL) with or without cleft palate is one of the most common congenital facial deformities.^2,3 After surgical repair of CL, the secondary deformity can vary from barely detectable to significant, and has been shown to be associated with psychosocial ramifications such as low self-esteem.⁴ Eye-tracking research has confirmed that the attention of naive observers is drawn toward areas of the face distorted by congenital or acquired forms of facial difference, including CL.^5,6 These eye-tracking studies have tended to exclude as observers those who are affected by—or who possess some sort of direct personal history with—CL. In terms of the perception and attitudes of parents of children with CL, relatively little has been written.^7,8 In the current study, we endeavored to determine whether parental exposure to secondary cleft deformity heightens (sensitizes) or diminishes (desensitizes) visual attraction to these structural variations.

For healthcare providers involved in the management of CL (or other facial differences), there is inherent value in understanding how patients, their family, and naive observers instinctively perceive facial difference. While a legitimate surgical treatment goal should be to achieve anatomic symmetry and landmark alignment resulting in a face that is eye-tracked normally, an important question to consider is who is the observer, and what is the impact of familiarity on observational patterns. Beyond the academic value of this work, elucidating the perceptual disposition of parents may better inform shared clinical decision-making between provider, patient, and parents.

MATERIALS AND METHODS

Study Participants

This investigation was approved by the Qatar Biomedical Research Institute institutional review board (QBRI-IRB-2020-02-010). In this study, two distinct population groups consisting of 10 participants each were considered. The sample size in our study (20 participants) was determined based on previously published eye-tracking studies with a similar scope of work.^9,10 The first group (N = 10) were adults without personal or familial history of CL, and were referred to as the control group, denoted as P− (nonparents). The second group (N = 10) were parents of children with CL, denoted as P+. All participants were at least 18 years of age and reported 20/40 vision or better (lens correction permitted). P+ subjects were parents recruited from the clinical practice of the senior author (M.A.S.) and have children who had undergone prior CL repair. Only one parent per affected child was recruited. The P− participants were recruited from the general community.

Image Material

Fifty-six colored images of children’s faces were collected to construct two slideshow presentations. Of the 56 images, 32 experimental photographs were of children with repaired CL and 24 were of children without any facial deformity. The experimental images were then divided into two groups of 16. Each group was evenly distributed in terms of left and right CL images, with eight images for each category and with equal gender distribution. The 24 control images represented unaffected children’s faces and were obtained using a facial generator employing a generative adversarial neural network (Style GAN II) that produces highly realistic facial images.¹¹ All images were sized initially to 450 × 580, and were then cropped to eliminate hair, ears, and other peripheral features. Two slideshows consisting of 40 images were constructed to each contain 16 cleft images and 24 noncleft control images. One of the two slideshows was randomly selected and presented to each participant.

Eye-tracking Protocol

Image stimuli were presented to participants on a 14-inch laptop monitor for 6 seconds in randomized order. Successive images were also randomly positioned on the screen (top, bottom, right, or left) to prevent habituation. The slideshow was presented using iMotions Platform (iMotions A/S, Copenhagen, Denmark) and eye movement details were recorded, stored, and retrieved using the same software package. Observers were situated 60 cm from the eye-tracking sensor with their heads held stationary in a chinrest. The participants were kept naive to the particular nature of the images and were informed that the research aim was to study human perception of faces. Each experiment commenced with a calibration procedure. Eighteen regions of interest (ROIs) were outlined onto each facial image post-hoc after the collection of the eye-tracking data using a freehand drawing tool within the iMotions software tool (Fig. 1). A Tobii Pro X3-120 screen-based eye tracker (Tobii AB, Stockholm, Sweden) with temporal resolution of 8 ms, sampling rate of 120 Hz, gaze accuracy of 0.4 to 0.5 degrees, and gaze precision of 0.24 to 0.34 degrees was used to record the participants’ eye movements.

Fig. 1.: For each facial image, eighteen facial regions of interest (ROIs) were assigned and analyzed post hoc after the eye-tracking protocol. The ROIs were labeled as forehead (1,2); periorbital (3,4); glabellar (5,6); infraorbital (7,8); lateral nasal sidewall (9,10); mid-cheek (11,12); nasal tip, nares, and columella (13,14); upper lip (15,16); lower lip, chin, mandible (17,18). Image accessed from internet, licensed under the Creative Commons Attribution-Share Alike 3.0 Unported license.¹²

Data Analysis

The primary measure of evaluation in this study was fixation duration (FD). FD reflects the duration of time an observer’s eyes fixate within a particular region of interest. The FD measures were normalized by dividing the FD of a particular ROI by the total FD spent in all ROIs generating fixation duration percentage (FD%). Heatmaps were constructed in order to demonstrate the overall patterns of visual fixation across the two groups of observers (P+ and P−). Mann-Whitney U test was applied to the data set considering FD% across all ROIs and comparing P+ and P− observer groups. All statistical tests were performed using an open-source statistics program (JASP, https://jasp-stats.org), and a P value of 0.05 was assigned for significance.

RESULTS

Our eye-tracking analysis (Figs. 2 and 3) demonstrated that the periorbital region was predictably the most attractive target of fixation for all observers whether examining cleft images (P− 25.6%; P+ 37.1%) or control images (P− 31.4%; P+ 41.5%). Our data also revealed that all participants preferentially fixated on the nasal and perioral regions of CL images relative to control images. However, notable differences were detected in the visual fixation patterns of parent (P+) versus nonparent (P−) observers. Parents of children with CL fixated 25% of the time on the upper lip, and 24% within the nasal tip/nares/columellar region, whereas control adults fixated on those same regions 14.6% and 19.3% of the time (P < 0.001 and P = 0.026, respectively, Fig. 2). Figure 4 shows heatmap representations for the two groups’ observation of CL images and control images. The heatmaps reflect the visual tendency of P+ subjects to more closely attend to the perioral lookzones compared with their control peers when viewing a repaired CL image.

Fig. 2.: Parents’ versus naive observers’ visual fixation of bilateral ROIs when viewing repaired cleft lip (CL) images. P−, nonparental controls; P+, parents of patients with CL.

Fig. 3.: Parents’ versus naive observers’ visual fixation of bilateral ROIs when viewing control images. P− (nonparental controls); P+ (parents of patients with CL).

Fig. 4.: Heatmaps representing the gaze patterns of control adult observers (P−) and parents of children with CL (P+) when viewing an image of a patient with a repaired CL (A, B) and a control image (C, D). Image (A, B) accessed from internet, (https://www.flickr.com/photos/interplast/429787606)¹³, licensed under https://creativecommons.org/licenses/by-nc-nd/2.0/, [Accessibility verified June 25, 2022]. Image (C,D) generated using the Style GAN II facial generator.¹¹

DISCUSSION

This study employed eye-tracking technology to compare observers’ visual processing of children’s faces with or without a secondary CL deformity. The particular focus of our work was to investigate whether intimate familiarity with a category of facial deformity sensitizes or desensitizes one to similar structural facial anomalies. A number of prior studies have assessed the gaze patterns of observers encountering images depicting CL.^5,6,14–17 However, none to our knowledge have specifically compared as observers parents of cleft patients versus adult controls. Moreover, most of the prior reports were limited by some combination of considering only black and white images, a small sample size of experimental or control stimuli, cleft images depicting only adult patients, and/or a limited number of facial ROIs, which would presumably reduce the sensitivity of gaze pattern assessment.

In the current study, we examined the responses to 40 colored facial images of children aged 4 to 7 years, 16 of whom exhibited a repaired unilateral CL, in addition to 24 control images. We considered gaze patterns vis-a-vis 18 facial aesthetic ROIs that were hand-drawn post-hoc onto all stimuli images. Our investigation confirmed prior findings that observers in general are distracted away from the usually predominant ROI of the periorbital area towards areas considered as facial structural outliers. In fact, our data align remarkably well with data that we collected previously that compared the gaze patterns of observers with or without any personal history of facial deformity (loosely defined as “any type of notable facial blemish affecting themselves or someone they know closely in their lives”) responding to cleft and control images.¹⁵ In our prior work we found a 1.47 times relative increase in the FD% of the upper lip ROI in observers with any personal history of facial deformity compared with control observers. In the current report we show that parents of children with a repaired CL had a 1.71 times relative increase in FD% in the upper lip region of interest. We also measured here a 1.24 relative increase in FD% on the part of parents with respect to the nasal tip/nares/columella ROI, whereas in our prior study there was essentially no difference detected by adults with a less well-defined personal history of facial deformity.

Taken together, these findings suggest that parents of children with CL are particularly primed to identify subtle distortions about the lip and nose, more so than would the casual observer. This is consistent with the conclusion of Dindaroglu et al, who showed that orthodontists were more attracted to the perioral region of images of repaired CL than were laypeople.

Another aspect to consider is that facial appearance has been shown to exert a powerful influence over the instantaneous attribution of personality characteristics.^18,19 While van Schijndel et al were unable to detect a significant difference in observer personality ratings for nine adult cleft images (raw versus digitally repaired),⁵ others have found a distinct detrimental effect of facial and dental deformity on character appraisal.^20,21 In prior work, even natural aging changes of the face have been shown to be associated with a reduction in positively valenced character attributes.²² To our knowledge, there are no preceding investigations that consider the interplay between observers’ gaze pattern, personal history of the facial change in question, and social attribution. These parameters, and how they interrelate, are important considerations for the surgeon managing a child with a cleft. Both the expressed and the latent attitudes that may underlie a parent’s decision-making should be factored into any careful preoperative discussion. Although two-dimensional anthropometric evaluation of facial landmark symmetry, expert ratings, and patient self-reporting of outcomes provide meaningful information, a deeper understanding of observer perception of cleft faces promises a more nuanced approach to clinical care in the future. An ideal perspective for the cleft surgeon would be to see a child’s face simultaneously through both the eyes of an expert clinician and those of his/her parents, to facilitate a treatment plan that best incorporates those unique frames of reference.

CONCLUSIONS

This study used eye tracking methodology to compare the gaze patterns of naive observers to those of parents of children with repaired CL when viewing control and cleft-affected facial images. We show that the intimate familiarity of parents with the target cleft lesion subconsciously primes them, resulting in their heightened attraction during the early phase of visual processing to the affected areas of the face. These findings highlight the fact that observer background/profile can meaningfully influence the perception of a facial deformity. Awareness of this information may refine the level of communication between surgeon and parents of a child with a facial deformity.