Abstract
Objectives
This study aimed to analyse if children with untreated or treated caries (restorations/missing teeth) are perceived differently compared to children with healthy teeth and to explore possible differences in the perception by laypersons and dental experts.
Methods
Eye movements of female experts (n = 20) and laypersons (n = 18) were recorded by eye-tracking while paired images (neutral expression/teeth not visible; emotional expression/smiling, teeth visible) of children with healthy teeth, with visible untreated or treated caries (restorations/missing teeth as a consequence of caries treatment), each n = 13, were presented. First fixation, total fixation time and number of fixations on the areas of interest (eyes, nose, mouth) in the first two seconds of presentation were determined. Furthermore, the images were rated regarding arousal, valence and attractivity. Statistical analysis was performed using Mann-Whitney-U- and Kruskal-Wallis-tests (α = 0.05).
Results
Generally, laypersons spent more time exploring and fixating the eye region than the mouth, while dental experts more often first percept and longer and more often fixated the mouth region, especially in images with emotional expression. Dental experts, but not laypersons, were significantly longer fixating the mouth of children with untreated caries than the mouth of children with healthy teeth in images with emotional expression. When evaluating images with emotional expression, both dental experts and laypersons rated children with healthy teeth to be more attractive, pleasant and calm than children with untreated or treated caries.
Conclusions
Children with visible treated and untreated caries were differently perceived by laypersons and dental experts than children with healthy teeth.
1
Introduction
Caries in primary teeth, especially early childhood caries, is a common disease which negatively impacts the quality of life of children and their parents/caretakers by affecting functional, psychological and social well-being. Among other effects on social life, visible caries is assumed to impair social interactions, for instance as children might be teased or bullied because of esthetic or phonetic problems .
Several studies have shown that beside visible dentofacial anomalies and missing teeth , visible caries may also affect observers subjective ratings of peoples’ facial attractiveness and provoke negative social judgements. When presenting computer-modified photographs of adults with or without caries lesions of the anterior teeth, subjects with caries were rated to be less socially and intellectually competent . Karunakaran et al. showed that the presence of visible caries resulted in lower rating of attractiveness, but did not affect social judgements .
So far, no information is available on how children with untreated caries are perceived by their social environment. It is known that children from early infancy to six years of age are generally judged to be more likeable or attractive than adult faces, most probably as the so-called “baby schema” evokes positive reactions of adult raters . Children attractiveness influences the social interaction with adults, e.g. attractive children were perceived as more competent and more pleasant .
As children faces are more positively perceived than adult faces, it is questionable whether the presence of visible caries affects the perception and, thus the overall attractiveness compared to children with healthy teeth. Therefore, the present study aimed to analyse if children with visible untreated or treated caries (e.g. stainless steel crowns, missing teeth) are perceived differently compared to children with healthy teeth and to determine possible differences in the perception by laypersons and dental experts.
Recording the movements of participants’ eyes allows to analyse participants’ attentional focus and cognitive information processing while perceiving visual stimuli . Therefore, the eye tracking technology was used in the present study to precisely determine involuntary eye movements and fixations to understand the initial stage of perception of children with untreated or treated caries to children with healthy teeth. Moreover, image rating was used as a subjective measurement to evaluate attractivity, valence and arousal.
The null hypotheses for both eye tracking and image rating data were 1) that children with visible untreated or treated caries were not differently perceived and 2) that perceptions of laypersons and dental experts were not significantly different.
2
Materials and methods
The study was approved by the local ethics committee (36/2/16) and previously registered at ClinicalTrials.gov (NCT02899273).
2.1
Participants
Female participants with or without professional dental experience were recruited from the University Medical Center in Göttingen (dental students, experts) and the Institute of Psychology in Hildesheim (psychology students, laypersons), Germany. Inclusion criteria were female undergraduate students (for dental students: in the clinical curriculum), >18 years , right hander , without glasses, without diseases or medication altering perception or neurophysiological stimulus processing . Dental students had to have performed at least one patient-treatment course with focus on Operative Dentistry including Pediatric Dentistry.
As this was the first study analysing the perception of ECC by eye-tracking, the sample size could not be calculated on the basis of prior studies. A total of 38 students (18 laypersons, mean age: 21.4 ± 2.6 years; 20 experts, mean age: 23.9 ± 2.1 years) volunteered to participate. The participants were initially informed that this study aims to assess the visual perception of childrens’ faces and gave written informed consent. After the experiment, the participants were debriefed and fully informed about the aim of the research project.
2.2
Patient images
Images were obtained from patients of the Department of Preventive Dentistry, Periodontology and Cariology, University Medical Center Göttingen, Germany. Children (aged 2–10 years) with healthy teeth, visible untreated caries (maxillary front teeth with or without primary molars or maxillary and mandibular front teeth with or without primary molars) or treated caries (restorations, e.g. stainless steel crowns, fillings and missing teeth) were asked to participate in the study. Exclusion criteria were pronounced skeletal and soft tissue changes. Written consent was obtained from parents/legal guardians of the children before enrollment in the study.
Two frontal portraits (neutral facial expression/teeth not visisble and emotional facial expression (smiling)/teeth visible) were taken from 124 children using a digital reflex camera (exposure time: 1/200s, aperture: f/8.0, sensitivity: ISO 200, Canon EOS 500D, Canon Inc., Japan) with macro lens (Canon EF 100 mm f/2.8 Macro USM, Canon Inc., Japan) and flashlight (Canon Speedlite 580EX II, Canon Inc, Japan) under standardized conditions. The initial resolution of the images amounted to 4752 pixels x 3168 pixels. To reduce background information, hair, neck, clothes and other peripheral features were cropped (Adobe Photoshop, version 2015.0.0, Adobe Systems Inc., USA). The interpupillary lines were horizontally aligned. All images were standardized to 400 pixels x 600 pixels (Adobe InDesign, version 2015.0.0, Adobe Systems Inc., USA) and centrally fitted into a grey background (RGB color: 128, 128, 128).
Images with neutral expression were assessed by 26 dentists (6 male, 20 female, mean age: 33.5 ± 11.4 years) with regard to arousal, valence and attractivity to exclude images that induce severe emotions (e.g. child is perceived as very unattractive), which might influence the eye-tracking experiment . Only images which were rated neutral by the examiners were included in the study (Kruskal-Wallis test followed by Nemenyi post-hoc tests, α = 0.001). Then, classification of the remaining images into the three groups was done by 15 dentists (5 male, 10 female, mean age: 31.0 ± 9.7 years) using images with visible teeth. Images were excluded, if the inter-observer agreement among the examiners was below 60%. Finally, image pairs (neutral and emotional expression) of 39 children were included in the present study: healthy teeth (n = 13, mean age: 6.4 ± 2.3 years), visible caries (n = 13, mean age: 6.0 ± 2.4) and treated caries (n = 13, 6.9 ± 1.8). Examples of images with neutral and emotional expression are presented in Fig. 1 .
2.3
Recording of eye movements
Eye movements were recorded using a high speed eye tracker (iView X™ Hi-Speed 1250 Hz Systems, SensoMotoric Instruments, Teltow, Germany) and an analysis software (iViewX 2.7 built 13, SensoMotoric Instruments, Teltow, Germany). The system uses an infrared camera and was ajusted to capture the monocular left eye. Images of the eye‘s position were collected at 1250 Hz and a spatial resolution of 0.25° to 5° visual arc. The eye tracking system was calibrated before each participant using a nine-point-calibration matrix.
To eliminate ambient noise, to avoid distortions of the corneal reflex, and to standardize light conditions, eye-tracking measurements were performed in a noise-cancelling room (4 m x 2 m x 2.5 m) at 50 cd/m 2 luminous intensity. Each participant was asked to sit in an upright position with a distance of 100 cm from the screen (Quato Intelli Proof 240, 240 inch TFT-monitor, 1920 × 1200 pixel, 60 Hz, 12 ms S/W, Quatographic Technology, Germany). The head was stabilized using chin and forehead supports to eliminate head movements.
Stimulus presentation was controlled by an iMax 24“(IntelCore™ 2 Duo CPU, 3.06 GHz, 4 GB RAM; system software: Mac OS X, Version 10.5.8., Apple Inc 1983–2009), using Psych Toolbox 3 (Brainard, 1997; Pelli, 1997; Kleiner et al., 2007) for MatLab, version 7.8.0. (R2009a, The MathWorks Inc, 1984–2009; Natick, MA, USA). Images with neutral and emotional expression were paired and presented side by side with a visual angle of 7.11°. Each pair of images was shown for 3 s, followed by a grey screen with a fixation cross for 1.5 s . The paired images were presented in a random order. To eliminate any biases between right and left sides, the pictures were presented twice with changed positions . Thus, in total 78 paired images were presented.
2.4
Image rating
In the second part of the study, the participants were asked to rate the printed images (colored, size Din A4, neutral and emotional expression) on a 9-point scale (Self-Assessment Manikin, ) for arousal (calm to excited), valence (pleasant to unpleasant) and attractivity (unattractive to attractive). The rating was allowed without time-limit.
2.5
Outcome parameters and statistical analysis
The eyes, nose and mouth were each defined as an area of interest (AOI) according to anthropometric landmarks ( Fig. 1 ). The analyses were done accordingly to previous studies using the SMI BeGaze analysis software (version 3.4.27, SensoMotoroic Instruments, Teltow, Germany). This analysis software allowed for an automatic detection and exclusion of blinks, and for the extraction of saccade-like events (SLE) and fixations using a velocity-based, high-speed event detection algorithm. Saccades were defined as SLE, when peak velocity exceeded a threshold of 80°/s in a time window comprising 20–80% of the duration of a SLE. Additionally, a saccade duration of at least 25 ms, followed by a fixation duration of at least 100 ms were required. Accordingly, fixations were defined as the time between the end of one SLE and the beginning of the next one .
The first fixation and the total fixation time and the number of fixations on the AOIs in the first two seconds were defined as outcome parameters and extracted from the raw data. Comparisons between laypersons and dental experts were statistically analysed by Mann-Whitney-U-Tests folllowed by Bonferroni correction. Comparison between children with healthy teeth, with untreated caries and with treated caries were performed by Kruskal-Wallis-tests and post-hoc tests (p < 0.05).
Valence, arousal and attractivity were statistically analysed by Mann-Whitney-U-Tests folllowed by Bonferroni correction (for comparison between laypersons and dental experts) and Kruskal-Wallis-tests and post hoc tests (for comparison between children with healthy teeth, with untreated caries and with treated caries), p < 0.05.
2
Materials and methods
The study was approved by the local ethics committee (36/2/16) and previously registered at ClinicalTrials.gov (NCT02899273).
2.1
Participants
Female participants with or without professional dental experience were recruited from the University Medical Center in Göttingen (dental students, experts) and the Institute of Psychology in Hildesheim (psychology students, laypersons), Germany. Inclusion criteria were female undergraduate students (for dental students: in the clinical curriculum), >18 years , right hander , without glasses, without diseases or medication altering perception or neurophysiological stimulus processing . Dental students had to have performed at least one patient-treatment course with focus on Operative Dentistry including Pediatric Dentistry.
As this was the first study analysing the perception of ECC by eye-tracking, the sample size could not be calculated on the basis of prior studies. A total of 38 students (18 laypersons, mean age: 21.4 ± 2.6 years; 20 experts, mean age: 23.9 ± 2.1 years) volunteered to participate. The participants were initially informed that this study aims to assess the visual perception of childrens’ faces and gave written informed consent. After the experiment, the participants were debriefed and fully informed about the aim of the research project.
2.2
Patient images
Images were obtained from patients of the Department of Preventive Dentistry, Periodontology and Cariology, University Medical Center Göttingen, Germany. Children (aged 2–10 years) with healthy teeth, visible untreated caries (maxillary front teeth with or without primary molars or maxillary and mandibular front teeth with or without primary molars) or treated caries (restorations, e.g. stainless steel crowns, fillings and missing teeth) were asked to participate in the study. Exclusion criteria were pronounced skeletal and soft tissue changes. Written consent was obtained from parents/legal guardians of the children before enrollment in the study.
Two frontal portraits (neutral facial expression/teeth not visisble and emotional facial expression (smiling)/teeth visible) were taken from 124 children using a digital reflex camera (exposure time: 1/200s, aperture: f/8.0, sensitivity: ISO 200, Canon EOS 500D, Canon Inc., Japan) with macro lens (Canon EF 100 mm f/2.8 Macro USM, Canon Inc., Japan) and flashlight (Canon Speedlite 580EX II, Canon Inc, Japan) under standardized conditions. The initial resolution of the images amounted to 4752 pixels x 3168 pixels. To reduce background information, hair, neck, clothes and other peripheral features were cropped (Adobe Photoshop, version 2015.0.0, Adobe Systems Inc., USA). The interpupillary lines were horizontally aligned. All images were standardized to 400 pixels x 600 pixels (Adobe InDesign, version 2015.0.0, Adobe Systems Inc., USA) and centrally fitted into a grey background (RGB color: 128, 128, 128).
Images with neutral expression were assessed by 26 dentists (6 male, 20 female, mean age: 33.5 ± 11.4 years) with regard to arousal, valence and attractivity to exclude images that induce severe emotions (e.g. child is perceived as very unattractive), which might influence the eye-tracking experiment . Only images which were rated neutral by the examiners were included in the study (Kruskal-Wallis test followed by Nemenyi post-hoc tests, α = 0.001). Then, classification of the remaining images into the three groups was done by 15 dentists (5 male, 10 female, mean age: 31.0 ± 9.7 years) using images with visible teeth. Images were excluded, if the inter-observer agreement among the examiners was below 60%. Finally, image pairs (neutral and emotional expression) of 39 children were included in the present study: healthy teeth (n = 13, mean age: 6.4 ± 2.3 years), visible caries (n = 13, mean age: 6.0 ± 2.4) and treated caries (n = 13, 6.9 ± 1.8). Examples of images with neutral and emotional expression are presented in Fig. 1 .
2.3
Recording of eye movements
Eye movements were recorded using a high speed eye tracker (iView X™ Hi-Speed 1250 Hz Systems, SensoMotoric Instruments, Teltow, Germany) and an analysis software (iViewX 2.7 built 13, SensoMotoric Instruments, Teltow, Germany). The system uses an infrared camera and was ajusted to capture the monocular left eye. Images of the eye‘s position were collected at 1250 Hz and a spatial resolution of 0.25° to 5° visual arc. The eye tracking system was calibrated before each participant using a nine-point-calibration matrix.
To eliminate ambient noise, to avoid distortions of the corneal reflex, and to standardize light conditions, eye-tracking measurements were performed in a noise-cancelling room (4 m x 2 m x 2.5 m) at 50 cd/m 2 luminous intensity. Each participant was asked to sit in an upright position with a distance of 100 cm from the screen (Quato Intelli Proof 240, 240 inch TFT-monitor, 1920 × 1200 pixel, 60 Hz, 12 ms S/W, Quatographic Technology, Germany). The head was stabilized using chin and forehead supports to eliminate head movements.
Stimulus presentation was controlled by an iMax 24“(IntelCore™ 2 Duo CPU, 3.06 GHz, 4 GB RAM; system software: Mac OS X, Version 10.5.8., Apple Inc 1983–2009), using Psych Toolbox 3 (Brainard, 1997; Pelli, 1997; Kleiner et al., 2007) for MatLab, version 7.8.0. (R2009a, The MathWorks Inc, 1984–2009; Natick, MA, USA). Images with neutral and emotional expression were paired and presented side by side with a visual angle of 7.11°. Each pair of images was shown for 3 s, followed by a grey screen with a fixation cross for 1.5 s . The paired images were presented in a random order. To eliminate any biases between right and left sides, the pictures were presented twice with changed positions . Thus, in total 78 paired images were presented.
2.4
Image rating
In the second part of the study, the participants were asked to rate the printed images (colored, size Din A4, neutral and emotional expression) on a 9-point scale (Self-Assessment Manikin, ) for arousal (calm to excited), valence (pleasant to unpleasant) and attractivity (unattractive to attractive). The rating was allowed without time-limit.
2.5
Outcome parameters and statistical analysis
The eyes, nose and mouth were each defined as an area of interest (AOI) according to anthropometric landmarks ( Fig. 1 ). The analyses were done accordingly to previous studies using the SMI BeGaze analysis software (version 3.4.27, SensoMotoroic Instruments, Teltow, Germany). This analysis software allowed for an automatic detection and exclusion of blinks, and for the extraction of saccade-like events (SLE) and fixations using a velocity-based, high-speed event detection algorithm. Saccades were defined as SLE, when peak velocity exceeded a threshold of 80°/s in a time window comprising 20–80% of the duration of a SLE. Additionally, a saccade duration of at least 25 ms, followed by a fixation duration of at least 100 ms were required. Accordingly, fixations were defined as the time between the end of one SLE and the beginning of the next one .
The first fixation and the total fixation time and the number of fixations on the AOIs in the first two seconds were defined as outcome parameters and extracted from the raw data. Comparisons between laypersons and dental experts were statistically analysed by Mann-Whitney-U-Tests folllowed by Bonferroni correction. Comparison between children with healthy teeth, with untreated caries and with treated caries were performed by Kruskal-Wallis-tests and post-hoc tests (p < 0.05).
Valence, arousal and attractivity were statistically analysed by Mann-Whitney-U-Tests folllowed by Bonferroni correction (for comparison between laypersons and dental experts) and Kruskal-Wallis-tests and post hoc tests (for comparison between children with healthy teeth, with untreated caries and with treated caries), p < 0.05.
Stay updated, free dental videos. Join our Telegram channel
VIDEdental - Online dental courses
