Gaze Behavior Research Articles

ACE-DNV: Automatic classification of gaze events in dynamic natural viewing

Eye movements offer valuable insights for clinical interventions, diagnostics, and understanding visual perception. The process usually involves recording a participant’s eye movements and analyzing them in terms of various gaze events. Manual identification of these events is extremely time-consuming. Although the field has seen the development of automatic event detection and classification methods, these methods have primarily focused on distinguishing events when participants remain stationary. With increasing interest in studying gaze behavior in freely moving participants, such as during daily activities like walking, new methods are required to automatically classify events in data collected under unrestricted conditions. Existing methods often rely on additional information from depth cameras or inertial measurement units (IMUs), which are not typically integrated into mobile eye trackers. To address this challenge, we present a framework for classifying gaze events based solely on eye-movement signals and scene video footage. Our approach, the Automatic Classification of gaze Events in Dynamic and Natural Viewing (ACE-DNV), analyzes eye movements in terms of velocity and direction and leverages visual odometry to capture head and body motion. Additionally, ACE-DNV assesses changes in image content surrounding the point of gaze. We evaluate the performance of ACE-DNV using a publicly available dataset and showcased its ability to discriminate between gaze fixation, gaze pursuit, gaze following, and gaze shifting (saccade) events. ACE-DNV exhibited comparable performance to previous methods, while eliminating the necessity for additional devices such as IMUs and depth cameras. In summary, ACE-DNV simplifies the automatic classification of gaze events in natural and dynamic environments. The source code is accessible at https://github.com/arnejad/ACE-DNV.

Conducting head-mounted eye-tracking research with young children with autism and children with increased likelihood of later autism diagnosis

BackgroundOver the past years, researchers have been using head-mounted eye-tracking systems to study young children’s gaze behaviors in everyday activities through which children learn about the world. This method has great potential to further our understanding of how millisecond-level gaze behaviors create multisensory experiences and fluctuate around social environments. While this line of work can yield insight into early perceptual experiences and potential learning mechanisms, the majority of the work is exclusively conducted with typically-developing children. Sensory sensitivities, social-communication difficulties, and challenging behaviors (e.g., disruption, elopement) are common among children with developmental disorders, and they may represent potential methodological challenges for collecting high-quality data.ResultsIn this paper, we describe our research practices of using head-mounted eye trackers with 41 autistic children and 17 children with increased likelihood of later autism diagnosis without auditory or visual impairments, including those who are minimally or nonspeaking and/or have intellectual disabilities. The success rate in gathering data among children with autism was 92.68%. 3 of 41 children failed to complete the play-session, resulting in an 86.36% success rate among 1–4-year-olds and a 100.00% success rate among 5–8-year-olds. 1 of 17 children with increased likelihood of later autism diagnosis failed to complete the play-session, resulting in a success rate of 94.11%. There were numerous “challenging” behaviors relevant to the method. The most common challenging behaviors included taking the eye-tracking device off, elopement, and becoming distressed. Overall, among children with autism, 88.8% of 1–4-year-olds and 29.4% of 5–8-year-olds exhibited at least one challenging behavior.ConclusionsResearch capitalizing on this methodology has the potential to reveal early, socially-mediated gaze behaviors that are relevant for autism screening, diagnosis, and intervention purposes. We hope that our efforts in documenting our study methodology will help researchers and clinicians effectively study early naturally-occuring gaze behaviors of children during non-experimental contexts across the spectrum and other developmental disabilities using head-mounted eye-tracking. Ultimately, such applications may increase the generalizability of results, better reflect the diversity of individual characteristics, and offer new ways in which this method can contribute to the field.

Gaze Behavior and Cognitive Performance on Tasks of Multiple Object Tracking and Multiple Identity Tracking by Handball Players and Non-Athletes.

Multiple object tracking (MOT) and multiple identity tracking (MIT) each measure the ability to track moving objects visually. While prior investigators have mainly compared athletes and non-athletes on MOT, MIT more closely resembles dynamic real-life environments. Here we compared the performance and gaze behavior of handball players with non-athletes on both MOT and MIT. Since previous researchers have shown that MOT and MIT engage different eye movement strategies, we had participants track 3-5 targets among 10 moving objects. In MOT, the objects were identical, while in MIT they differed in shape and color. Although we observed no group differences for tracking accuracy, the eye movements of athletes were more target-oriented than those of non-athletes. We concluded that tasks and stimuli intended by researchers to demonstrate that athletes' show better object tracking than non-athletes should be specific to the athletes' type of sport and should use more perception-action coupled measures. An implication of this conclusion is that the differences in object tracking skills between athletes and non-athletes is highly specific to the skills demanded by the athletes' sport.

A review of driver gaze estimation and application in gaze behavior understanding

Driver gaze plays a key role in different gaze-based applications, such as driver attentiveness detection, visual distraction detection, gaze behavior understanding, and building driver assistance system. The main objective of this study is to perform a comprehensive summary of driver gaze fundamentals, methods to estimate driver gaze using machine learning (ML) based technique, and its applications in real world driving scenarios. We first discuss the fundamentals related to driver gaze, involving head-mounted and remote setup based gaze estimation and the terminologies used for driver gaze behavior understanding. Next, we list out the existing benchmark driver gaze datasets, highlighting the collection methodology, and the equipment used for such data collection. This is followed by a discussion of the algorithms used for driver gaze estimation, which primarily involves traditional machine learning and deep learning (DL) based techniques. The estimated driver gaze is then used for understanding gaze behavior while maneuvering through intersections, on-ramps, off-ramps, lane changing, determining the effect of roadside advertising structures and also for developing driver gaze based applications such as maneuver prediction, driver inattention, and distraction detection systems, etc. Finally, we discuss the limitations in the existing literature, challenges, and future scope in driver gaze estimation and gaze-based applications.

Correlation Between Gaze Behaviors and Social Communication Skills of Young Autistic Children: A Meta-Analysis of Eye-Tracking Studies.

This meta-analysis examined correlations between eye-tracking measures of gaze behaviors manifested during dynamic salient social stimuli and behavioral assessment measures of social communication skills of young autistic children. We employed a multilevel model with random effects to perform three separate meta-analyses for correlation between social communication skills and (a) all gaze behaviors, (b) gaze duration, and (c) gaze transition. Subsequently, we performed meta-regression to assess the role of four moderators, including age, continuum of naturalness of stimuli, gaze metric, and area of interest, on correlation effect sizes that were heterogeneous at the population level. A total of 111 correlation coefficients from 17 studies for 1132 young autistic children or children with high-likelihood for autism (Mage range = 6-95months) were included in this meta-analysis. The correlation effect sizes for all three meta-analyses were significant, supporting the relation between improved gaze behaviors and better social communication skills. In addition, age, gaze metric, and area of interest were significant moderators. This suggests the importance of identifying meaningful gaze behaviors related to social communication skills and the increasingly influential role of gaze behaviors in shaping social communication skills as young autistic children progress through the early childhood stage. The continuum of naturalness of stimuli, however, was revealed to trend towards having a significant moderating effect. Lastly, it is important to note the evidence of potential publication bias. Our findings are discussed in the context of early identification and intervention and unraveling the complex nature of autism.

The distracting effect of mobile phones on visual attention in consumer decision-making: a gaze behavior perspective

ABSTRACT Visual attention as a highly complex system is considered a scarce source. This study investigates the distraction effect (specifically, mobile phone usage unrelated to purchasing decisions) on visual attention during consumers’ decision-making processes of the Mobile native generation. An eye-tracking experiment utilizing 2-AFC tasks was used for the product choice procedure. The findings revealed no significant differences in gaze behavior in distracting and non-distracting conditions. Interestingly, the results indicate the significant effect of consumer distraction on the relationship between visual attention and product preference. A stronger tendency for a value-based gaze and attentional bias effect under distraction conditions was evidenced. The results of this research highlight the importance of further developing theories of attention in consumer decision-making in terms of the need to re-evaluate the effects of factors on visual attention in distraction-rich environments, particularly regarding mobile technology.

Situational Awareness Assessment of Drivers Boosted by Eye-Tracking Metrics: A Literature Review

The conceptual framework for assessing the situational awareness (SA) of drivers consists of three hierarchical levels: perception of the elements of the environment, comprehension of the elements, and decision-making in the near future. A common challenge in evaluating SA is the determination of the available subjective and objective techniques and their selection and integration into methodologies. Among the objective techniques, eye tracking is commonly used, considering the influence of gaze behavior on driving. This review is presented as an innovative approach to the subject matter, introducing physiological metrics based on eye tracking and investigating their application in assessing the SA of drivers. In addition, experiments and methodologies that revealed patterns at the three levels of SA were identified. For this purpose, databases were searched, and 38 papers were considered. Articles were clustered according to prevalent themes such as eye-tracking metrics, eye-tracking devices, experiment design, and the relationship between SA and eye-tracking. This review summarizes the main metrics and key findings for each article and reveals a wide relationship between the eye-tracking metrics and SA. The influence of appropriately calibrated equipment, refined data collection protocols, and adequate selection of the eye-tracking metrics was examined. Further reviews are needed to systematically collect more evidence.

Sustained looking at faces at 5months of age is associated with socio-communicative skills in the second year of life.

Efficiently processing information from faces in infancy is foundational for nonverbal communication. We studied individual differences in 5-month-old infants' (N=517) sustained attention to faces and preference for emotional faces. We assessed the contribution of genetic and environmental influences to individual differences in these gaze behaviors, and the association between these traits and other concurrent and later phenotypes. We found an association between the mean duration of looking at a face (before looking away from it) at 5months and socio-communicative abilities at 14months (β=0.17, 95% CI: 0.08; 0.26, p<0.001). Sustained attention to faces predicted socio-communicative abilities over and above variance captured by mean fixation duration. We also found a statistically significant but weak tendency to prefer looking at smiling faces (relative to neutral faces), but no indication that variability in this behavior was explained by genetic effects. Moderate heritability was found for sustained attention to faces (A=0.23, CI: 0.06; 0.38), while shared environmental influences were non-significant for both phenotypes. These findings suggest that sustained looking at individual faces before looking away is a developmentally significant 'social attention' phenotype in infancy, characterized by moderate heritability and a specific relation to later socio-communicative abilities.

Designing for Trust: How Human-Machine Interface Can Shape the Future of Urban Air Mobility

This study investigates the influence of human-machine interface (HMI) design on passenger trust in autonomous electric vertical take-off and landing (eVTOL) vehicles. An immersive simulator-based experiment was conducted with 34 participants, exposing them to four HMI conditions: baseline, movement, hazard detection, and full information condition. As related measures of passenger trust, we collected self-reported measures including trust, perceived safety, perceived reliability, and intention to use. In addition, physiological measures including gaze behavior, electrodermal activity, and heart rate were collected. The results indicated that movement and hazard detection information improved passenger trust, suggesting that HMI design could play a crucial role in enhancing the acceptance of autonomous eVTOLs. In addition, gaze behavior showed a stronger relationship with self-reported trust than other physiological measures. The findings underscore the importance of HMI design in fostering passenger trust, which is critical to the success of urban air mobility.

Visual search patterns during exploration of naturalistic scenes are driven by saliency cues in individuals with cerebral visual impairment

We investigated the relative influence of image salience and image semantics during the visual search of naturalistic scenes, comparing performance in individuals with cerebral visual impairment (CVI) and controls with neurotypical development. Participants searched for a prompted target presented as either an image or text cue. Success rate and reaction time were collected, and gaze behavior was recorded with an eye tracker. A receiver operating characteristic (ROC) analysis compared the distribution of individual gaze landings based on predictions of image salience (using Graph-Based Visual Saliency) and image semantics (using Global Vectors for Word Representations combined with Linguistic Analysis of Semantic Salience) models. CVI participants were less likely and were slower in finding the target. Their visual search behavior was also associated with a larger visual search area and greater number of fixations. ROC scores were also lower in CVI compared to controls for both model predictions. Furthermore, search strategies in the CVI group were not affected by cue type, although search times and accuracy showed a significant correlation with verbal IQ scores for text-cued searches. These results suggest that visual search patterns in CVI are driven mainly by image salience and provide further characterization of higher-order processing deficits observed in this population.

Gaze behavior is influenced by task demands rather than football players creativity

Introduction&#x0D; The “most creative” football players seem to make more fixations of shorter durations than “least creative” players, indicating a broader attentional focus (Roca et al., 2018, 2021). In these studies, however, players are classified as “creative” based on their ability to name many, diverse and unusual solutions rather than to execute these options and they were only provided with a third-person viewing perspective when making decisions. In our study, funded by a FIFA scholarship, players experienced a first-person viewing perspective and when executing their most functional solution in virtual-reality game scenarios. Contrasting the idea of a distinct gaze strategy generally used by creative players, we expected that task demands would better explain gaze behavior than creativity.&#x0D; Methods&#x0D; We developed an immersive, 360° virtual reality (VR) setup using the Unreal engine to animate real football scenes and a 6 x 12 m cave-like laboratory with wall and floor projections. In five “mother scenes”, the location of teammates off the ball was systematically manipulated leading to 80 trials; each being occluded 1 s after the virtual ball was at the participants’ foot. We examined the originality, decision-quality, and creativity of the participants’ initial response; response time and gaze behavior (i.e., average number of fixations, the time interval between fixations, and the distance between fixations to examine the breadth of fixations). We expected that distances would be large when scanning the environment and small shortly before and after the ball must be played. Based on expert ratings of players’ executed options, we ranked participants to a more (n = 7) and less creative group (n = 7). Groups were significantly different in overall creativity (p &lt; .0001), originality (p &lt; .0001) and quality (p &lt; .001).&#x0D; Results&#x0D; The results show that groups did not differ in response times (p = 0.34), mean number of fixations (p = .48) or mean fixation duration (p = .68). The fixation distance in the entire trial was significantly higher than shortly before (p &lt; 0.01) or shortly after (p &lt; 0.01) the (virtual) ball is received, with no difference between more and less creative players. There was a significant difference in fixation distances for the 5 “mother scenes” (p &lt; 0.01).&#x0D; Discussion/Conclusion&#x0D; All players, not only the creative ones, have a wide fixation pattern before the ball is played to them, presumably to identify the most functional passing option. They then concentrate on the reception of the ball with a narrow fixation pattern. This pattern remains also in the pass execution phase. Creative players seem to better select and execute the best (and most creative) option rather than showing a distinct gaze behavior. Overall, our results suggest that players should be able to adjust their gaze pattern to the task demands to foster creative actions.&#x0D; References&#x0D; Roca, A., Ford, P. R., &amp; Memmert, D. (2018). Creative decision making and visual search behavior in skilled soccer players. PLoS One, 13(7), Article e0199381. https://doi.org/10.1371/journal.pone.0199381&#x0D; Roca, A., Ford, P. R., &amp; Memmert, D. (2021). Perceptual-cognitive processes underlying creative expert performance in soccer. Psychological Research, 85(3), 1146–1155. https://doi.org/10.1007/s00426-020-01320-5

Does fiducial marker visibility impact task performance and information processing in novice and low-time pilots?

Invisible fiducial markers are introduced for localization of Areas Of Interest (AOIs) in mobile eye tracking applications. Fiducial markers are made invisible through the use of film passing Infra-Red (IR) light while blocking the visible spectrum. An IR light source is used to illuminate the markers which are then detected by an IR-sensitive camera, but which are imperceptible by the human eye. We provide the first empirical study that demonstrates such invisible markers are not distracting to a given task, as demonstrated in a flight simulator where distraction of visible and invisible markers are compared between experienced and novice pilots. Fixation frequency and subjective distraction scores showed that visible markers disrupted natural gaze behavior, particularly in novice pilots. Our findings show that invisible markers should be used when there is a need for them to remain inconspicuous.

How body postures affect gaze control in scene viewing under specific task conditions

Gaze movements during visual exploration of natural scenes are typically investigated with the static picture viewing paradigm in the laboratory. While this paradigm is attractive for its highly controlled conditions, limitations in the generalizability of the resulting findings to more natural viewing behavior have been raised frequently. Here, we address the combined influences of body posture and viewing task on gaze behavior with the static picture viewing paradigm under free viewing as a baseline condition. We recorded gaze data using mobile eye tracking during postural manipulations in scene viewing. Specifically, in Experiment 1, we compared gaze behavior during head-supported sitting and quiet standing under two task conditions. We found that task affects temporal and spatial gaze parameters, while posture produces no effects on temporal and small effects on spatial parameters. In Experiment 2, we further investigated body posture by introducing four conditions (sitting with chin rest, head-free sitting, quiet standing, standing on an unstable platform). Again, we found no effects on temporal and small effects on spatial gaze parameters. In our experiments, gaze behavior is largely unaffected by body posture, while task conditions readily produce effects. We conclude that results from static picture viewing may allow predictions of gaze statistics under more natural viewing conditions, however, viewing tasks should be chosen carefully because of their potential effects on gaze characteristics.

Effects of Exercise Habits and Gender on Sports e-Learning Behavior: Evidence from an Eye-Tracking Study.

In the post-epidemic era, an increasing number of individuals were accustomed to learning sports and physical activity knowledge online for fitness and health demands. However, most previous studies have examined the influence of e-learning materials and resources on learners and have neglected intrinsic factors such as experience and physiological characteristics. Therefore, we conducted a study to investigate the effect of exercise habits and gender on sports e-learning behavior via eye-tracking technology. We recruited a sample of 60 undergraduate students (mean age = 19.6) from a university in Nanjing, China. They were randomly assigned into 4 groups based on 2 genders × 2 exercise habits. Their gaze behavior was collected by an eye-tracking device during the experiment. The cognitive Load Test and Learning Effect Test were conducted at the end of the individual experiment. (1) Compared to the non-exercise habit group, the exercise habit group had a higher fixation count (P<0.05), a shorter average fixation duration (P<0.05), a smaller average pupil diameter (P<0.05), and a lower subjective cognitive load (P<0.05) and better learning outcome (P<0.05). (2) Male participants showed a greater tendency to process information from the video area of interest (AOIs), and had lower subjective cognitive load (P < 0.05) and better learning outcomes (P < 0.05). (3) There was no interaction effect between exercise habits and gender for any of the indicators (P > 0.05). Our results indicate that exercise habits effectively enhance sports e-learning outcomes and reduce cognitive load. The exercise habits group showed significant improvements in fixation counts, average fixation duration, and average pupil diameter. Furthermore, male subjects exhibited superior learning outcomes, experienced lower cognitive load, and demonstrated greater attentiveness to dynamic visual information. These conclusions are expected to improve sports e-learning success and address educational inequality.

Gaze in cats (Felis catus) and dogs (Canis lupus familiaris).

Within human-animal dyadic interactions, dog-human gaze has been identified as the crux of several important visual behaviors, such as looking back, gaze-following, and participation in an oxytocin feedback loop. It has been posited that this gaze behavior may have been motivated and sustained by cooperative relationships between dogs (Canis lupus familiaris) and humans (e.g., hunting, service roles), however, to investigate why gaze evolved, a comparison to a domesticated species that lacks a protracted history of cooperative companionship is needed: the domestic cat (Felis catus). In this study, we compare the gaze duration to owners of cats and dogs in a community science setting. We replicated previous gaze studies with dogs, wolves (Nagasawa et al., 2015), and dingoes (Johnston et al., 2017), requesting owners to sit with their pets for 5 min and interact as they normally would. Cats and dogs gazed at their owners for similar durations, but durations of petting and physical contact were significantly lower with cats. Gaze correlated significantly with vocalizations in dogs; however, no other correlations were significant. Dogs gazed less in our community science setting than dogs tested previously in-lab (Nagasawa et al., 2015). Ultimately, cats resemble dogs in their general gaze patterns, but not in most interactions with their owner. Future research should aim to include feral cats or wild cat species to shed light on gaze behavior development in the genus, while more community science work can identify the behaviors that shift for dogs between familiar and unfamiliar environments. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Role of Body Gaze, Interpersonal Sexual Objectification and Sensation Seeking in Sexual Self-Concept in Heterosexual Young Adults in Pakistan

Past research has shown that body gaze behavior is marked for being a central feature of sexual objectification and that it may be associated with the sexual self-concept and tendency for sensation seeking among both men and women. The purpose of the present study was to assess the role of body gaze, interpersonal sexual objectification and sensation seeking in the sexual self-concept of heterosexual young adults in Pakistan. A cross-sectional research design was used to assess the aforementioned predictive associations with a sample of 455 individuals comprising 250 males and 205 females selected through purposive sampling. Data analysis was done using Pearson product moment correlation, multiple regression using dummy coding and MANOVA. The findings showed that body gaze behavior toward the other gender significantly predicted interpersonal sexual objectification, higher levels of sexual desire and sexual anxiety along with a higher orientation toward sensation seeking. Moreover, men scored high on body gaze behavior, sexual desire, sexual anxiety and sensation seeking. These results resonate with the existing body of research on sexual objectification, sexual self-concept and sensation seeking and how cultural contexts may contribute toward the above associations.

Effect of the Visual Illusion on Stepping-Over Action and Its Association with Gaze Behavior.

An adequate foot clearance height while stepping over an obstacle is important for safety in daily life. In the present study, we examined whether visual illusions affect foot clearance during a stepping-over action, and whether this is further influenced by gaze behavior. Twelve participants stepped over an obstacle placed four meters away under conditions of three different obstacle characteristics: white, horizontal, or vertical lines. We measured the participants' foot clearances during the step-over action and their gaze behavior during the approaching phase. Participants stepped significantly higher over the obstacles in the vertical lines (illusion) condition. The duration of gaze fixation on the obstacle positively correlated with increased foot clearance in the vertical condition, suggesting that the effect of the visual illusion on foot clearance was enhanced by prolonged gaze fixation. Conversely, prolonged fixation negatively correlated with foot clearance in the white (control) condition, implying that a cautious perception of an obstacle may contribute to efficient stepping-over action.

Interacting with autistic virtual characters: intrapersonal synchrony of nonverbal behavior affects participants' perception.

Temporal coordination of communicative behavior is not only located between but also within interaction partners (e.g., gaze and gestures). This intrapersonal synchrony (IaPS) is assumed to constitute interpersonal alignment. Studies show systematic variations in IaPS in individuals with autism, which may affect the degree of interpersonal temporal coordination. In the current study, we reversed the approach and mapped the measured nonverbal behavior of interactants with and without ASD from a previous study onto virtual characters to study the effects of the differential IaPS on observers (N = 68), both with and without ASD (crossed design). During a communication task with both characters, who indicated targets with gaze and delayed pointing gestures, we measured response times, gaze behavior, and post hoc impression formation. Results show that character behavior indicative of ASD resulted in overall enlarged decoding times in observers and this effect was even pronounced in observers with ASD. A classification of observer's gaze types indicated differentiated decoding strategies. Whereas non-autistic observers presented with a rather consistent eyes-focused strategy associated with efficient and fast responses, observers with ASD presented with highly variable decoding strategies. In contrast to communication efficiency, impression formation was not influenced by IaPS.The results underline the importance of timing differences in both production and perception processes during multimodal nonverbal communication in interactants with and without ASD. In essence, the current findings locate the manifestation of reduced reciprocity in autism not merely in the person, but in the interactional dynamics of dyads.

How language influences spatial thinking, categorization of motion events, and gaze behavior: a cross-linguistic comparison

Abstract According to Talmy, in verb-framed languages (e.g., French), the core schema of an event (Path) is lexicalized, leaving the co-event (Manner) in the periphery of the sentence or optional; in satellite-framed languages (e.g., English), the core schema is jointly expressed with the co-event in construals that lexicalize Manner and express Path peripherally. Some studies suggest that such differences are only surface differences that cannot influence the cognitive processing of events, while others support that they can constrain both verbal and non-verbal processing. This study investigates whether such typological differences, together with other factors, influence visual processing and decision-making. English and French participants were tested in three eye-tracking tasks involving varied Manner–Path configurations and language to different degrees. Participants had to process a target motion event and choose the variant that looked most like the target (non-verbal categorization), then describe the events (production), and perform a similarity judgment after hearing a target sentence (verbal categorization). The results show massive cross-linguistic differences in production and additional partial language effects in visualization and similarity judgment patterns – highly dependent on the salience and nature of events and the degree of language involvement. The findings support a non-modular approach to language–thought relations and a fine-grained vision of the classic lexicalization/conflation theory.

The Effect of Eye Contact in Multi-Party Conversations with Virtual Humans and Mitigating the Mona Lisa Effect

The demand for kiosk systems with embodied conversational agents has increased with the development of artificial intelligence. There have been attempts to utilize non-verbal cues, particularly virtual human (VH) eye contact, to enable human-like interaction. Eye contact with VHs can affect satisfaction with the system and the perception of VHs. However, when rendered in 2D kiosks, the gaze direction of a VH can be incorrectly perceived, due to a lack of stereo cues. A user study was conducted to examine the effects of the gaze behavior of VHs in multi-party conversations in a 2D display setting. The results showed that looking at actual speakers affects the perceived interpersonal skills, social presence, attention, co-presence, and competence in conversations with VHs. In a second study, the gaze perception was further examined with consideration of the Mona Lisa effect, which can lead users to believe that a VH rendered on a 2D display is gazing at them, regardless of the actual direction, within a narrow range. We also proposed the camera rotation angle fine tuning (CRAFT) method to enhance the users’ perceptual accuracy regarding the direction of the VH’s gaze.The results showed that the perceptual accuracy for the VH gaze decreased in a narrow range and that CRAFT could increase the perceptual accuracy.