Child Pedestrians’ Perception of External Car Display: Effects of Communication Style and Visualization Type on Trust and Perceived Safety

1. Introduction

Simple traffic cues can confuse child pedestrians in interpreting their meaning. Cars’ lights indicating their intention to turn left or right can be easily understood by adults but may not be clear to children. Children also find it challenging to estimate the speed and gap distance of a distant car, making them vulnerable road users who are at higher risk than other commonly known vulnerable road users such as pedestrians, cyclists, and motorcyclists (Holländer, Colley, Rukzio & Butz, 2021; WHO, 20 June 2022). As a result, child pedestrians rely heavily on adult guidance and drivers’ cues such as eye contact and hand gestures when making decisions to cross the road (Deb, Carruth, Fuad, Stanley & Frey, 2020; Gitelman, Levi, Carmel, Korchatov & Hakkert, 2019; Koekemoer, Van Gesselleen, Van Niekerk, Govender & Van As, 2017; Morrongiello & Barton, 2009; Rosenbloom, Ben-Eliyahu & Nemrodov, 2008; Schwebel et al., 2018).

However, with the advent of autonomous vehicles (AVs) in which drivers become passengers, pedestrians will no longer be able to rely on driver cues. This could be particularly deadly for children. Researchers have identified the potential for external human-machine interfaces (eHMIs) to replace driver cues and provide more information to pedestrians in FAVs (Dey et al., 2020a; Holländer et al., 2019a; Rouchitsas & Alm, 2019). They have developed various designs for vehicle sound (Mahadevan, Somanath & Sharlin, 2018), car lights (Ackermans, Dey, Ruijten, Cuijpers & Pfleging, 2020; De Clercq, Dietrich, Núñez Velasco, De Winter & Happee, 2019; Dey, Habibovic, Pfleging, Martens & Terken, 2020; Dey, Martens, Wang, Ros & Terken, 2018; Li, Dikmen, Hussein, Wang & Burns, 2018), displays (Chang, Toda, Igarashi, Miyata & Kobayashi, 2018; Chang, Toda, Sakamoto & Igarashi, 2017; De Clercq et al., 2019; Deb, Strawderman, & Carruth 2018; Holländer et al., 2019a; Stadler, Cornet, Novaes Theoto & Frenkler, 2019), smartphone haptics (Mahadevan et al., 2018), and road projections (Chang et al., 2018; Fridman et al., 2017), and found that these can increase the acceptability and usability of AVs, and enhance pedestrians’ confidence, trust, and perceived safety when making crossing decisions (Chang et al., 2017; De Clercq et al., 2019; Deb et al., 2018; Holländer et al., 2019a; Rouchitsas & Alm, 2019; Stadler et al., 2019). Despite these new findings, previous studies have mostly focused on adult pedestrians, and research on child pedestrians is still lacking.

The aim of this study was to examine child pedestrians’ characteristics in interpreting external car display (ECD) designs, a type of eHMI on AVs, with a focus on their trust and perceived safety. We categorized ECD design concepts into two communication styles (advisory and informative) and three visualization types (text, sign, and symbol), and we investigated child pedestrians’ trust, perceived safety, and preferences. This was done using a combination of quantitative analysis through questionnaires and qualitative data analysis through in-depth interviews, providing a more comprehensive understanding.

The contributions of this study are as follows: First, our findings contribute to a better understanding of the impact of communication styles and their visualization types on pedestrians’ perceptions and crossing decisions, and provide a new perspective for designing ECD concepts for future studies. Second, the insights into child pedestrians’ perceptions of ECDs can inform researchers and practitioners in developing more effective ECD design concepts for AV-pedestrian communication. Given the scarcity of research on child pedestrians’ perceptions and interactions with AVs, understanding their needs and concerns can provide valuable information for developing guidelines and recommendations for ECD design concepts in AVs.

2. Related Works

2. 1. External Human-Machine Interfaces of Fully Autonomous Vehicles

In the context of AVs, eHMIs are projected to play a crucial role in not only substituting traditional driver cues, but also in enhancing pedestrian safety when making crossing decisions. Mahadevan et al. (2018) mixed visual, auditory, and physical cues in their eHMI prototypes, incorporating a speaker and smartphone haptic (Mahadevan et al., 2018). The speaker, mounted on the vehicle, played the voice of “stopping” or “starting,” while the smartphone held by participants vibrated to signal that it was safe to cross.

On the other hand, visual cues are the most frequently used eHMIs, ranging from car lights to road projections (Fridman et al., 2017). Among the various visual types of eHMIs, ECDs have received significant attention and have been verified to be effective in prior studies (Rouchitsas & Alm, 2019). ECDs are typically located at the front of the vehicle, such as the front grill or window, and aim to provide information about the vehicle to pedestrians. The information conveyed through ECDs varies, ranging from symbolic graphic images (e.g., a hand and walking pedestrian) to specific textual messages (see Figure 1).

Figure 1

ECD design concepts from prior works

Similar to traditional traffic signals, ECDs that provide visual instructions to pedestrians have the potential to reduce uncertainty and enhance trust and perceived safety. Results from prior studies have indicated that compared to conditions without ECDs, the use of ECDs leads to a more positive effect on trust and perceived safety (Deb et al., 2018; Holländer et al., 2019a; Löcken, Golling & Riener, 2019). However, further research is necessary to examine the design components of ECDs in order to enhance their effectiveness.

One aspect of ECD design that requires further investigation is the visualization type. If the different forms of representation such as text, signs, or symbols in ECD are confusing, they may reduce pedestrians’ perceived safety and increase the time taken to make crossing decisions (Deb et al., 2020; Deb et al., 2018). Additionally, the appropriate communication style to be used in ECDs is also a subject of debate. There are many prior studies that have utilized advisory messages (e.g., “walk”), but it has been suggested that informative messages (e.g., “braking”) may be more appropriate for enhancing safety and avoiding excessive assumptions or misleading pedestrians (Bazilinskyy et al., 2019; Cefkin et al., 2019; Habibovic et al., 2018; Liu et al., 2020; Rasouli & Tsotsos, 2019; Schaefer et al., 2016; She et al., 2021).

Although previous studies have validated the overall effectiveness of ECD designs, they have generally overlooked the specific design components that comprise these systems. In light of these considerations, it is essential to investigate the impact of the various components of ECD design, including visualization type and communication style, on pedestrian trust and perceived safety.

3. Categorization of ECD Design Concepts

Before studying child pedestrians’ characteristics in crossing decisions, we investigated ECD design concepts presented in prior studies. Some studies have classified the design concepts for ECDs according to their components, yet their focus is either too specific or too broad to be used in our study (Bazilinskyy et al., 2019; Löcken et al., 2019; Otherson et al., 2018). Otherson et al. (2018) who focused specifically on visualization style divided their design concepts into two types (i.e., lighting elements and symbols) and two dynamics (i.e., static and dynamic) (Otherson et al., 2018).

On the other hand, Löcken et al. (2019) who covered broad design components classified 28 design concepts presented in prior works using two axes—communication categories (i.e., visual, combinations of acoustic and visual, simulation of human behavior, and smart infrastructure) and complexity ranges (i.e., a simple cue, detailed cues, multiple modes of communication, and mimicking human behavior). Despite these efforts to examine various ECD design concepts, a more investigation into the components offered by ECD designs is necessary to gain a better understanding of the specific characteristics of ECDs that influence the perception and behavior of pedestrians (Bazilinskyy et al., 2019; Rouchitsas & Alm, 2019; She et al., 2021).

In this study, the design concepts of ECDs were categorized into two components: visualization type and communication style. With regards to visualization type, various ECD design concepts have been presented, and they can be grouped into three categories: texts (e.g., “Walk” or “Don’t walk”), signs (e.g., arrows), and symbols (e.g., walking man and anthropomorphic smile) (see Figure 2) (Ackermans et al., 2020; Chang et al., 2018; Chang et al., 2017; Deb et al., 2018; Fridman et al., 2017; Holländer et al., 2019a; Mahadevan et al., 2018; Stadler et al., 2019). Communication style categorization includes “advisory” which refers to messages that suggest appropriate actions to pedestrians and “informative” which delivers the status of vehicles (Li et al., 2018; Rasouli & Tsotsos, 2019; She et al., 2021). For instance, when transmitting the message “yielding,” De Clercq et al. (2019) utilized a textual message with an advisory style (e.g., “walk”), whereas Deb et al. (2018) used an informative style (e.g., “braking”).

Figure 2

Outcome of categorization of ECD design concepts presented in prior studies

4. Methods

Based on the literature review, we designed an experiment to test the following two research questions:

RQ 1: How do different communication styles (advisory and informative) in ECD design concepts influence/affect child pedestrians’ trust and perceived safety?
　
RQ 2: How do different visualization types (text, sign, and symbol) in ECD design concepts influence/affect child pedestrians’ trust and perceived safety?

4. 1. Study Design

To investigate the influence of ECD design concepts on child pedestrians’ trust and perceived safety, we designed a within-subject study with 2 (communication styles: advisory and informative) × 3 (visualization types: text, sign and symbol) experiments. Six design concepts were presented in videos and randomly shown to the participants. In the videos, there was a black car with a display presenting design concepts in the first lane of the road, and the car approached the crosswalk (see Figure 3). Physical prototypes or virtual reality (VR) were also considered but excluded in the end due to safety issues for the children participants (Kiefer, Pincus, Richardson, & Myer, 2017; Rouchitsas & Alm, 2019; Salanger, Lewis, Vallier, McDermott & Dergan, 2020).

Figure 3

Captured images of advisory-text concept stimuli simulated with 3D videos (left: the first scene of an approaching car; right: the second scene of a stopped car).

Design concepts for our experiments were created from two axes: communication style and visualization type. The communication style in our study was categorized into advisory and informative, as outlined by Rasouli and Tsotsos(2019), with an emphasis on elucidating “how to communicate the message” from the pedestrian perspective. Concerning visualization, three distinct types(text, sign and symbol) were selected to examine information based on Langer’s (1942) definition, where a sign serves as a “proxy for their objects,” and a symbol represents “vehicles for the conception of objects.” While acknowledging that signs and symbols may not be entirely mutually exclusive, our exploration was to delve deeper into these distinctions; specifically, sign “announce” their object, event, or situation, whereas symbols “lead him to conceive” their objects or meaning (Langer, 1942).

All the design concepts were created in a blue-green color with a black background for legibility and consistency (Faas, Stange & Baumann, 2021). We used the text type stimuli translated in Korean, reflecting the cultural context. The images of the six design concepts are shown in Figure 4.

Figure 4

Six design concepts for the ECDs used in the experiment

4. 3. Procedures

The sample size was calculated using G*power software. The recommended total sample size was 28 to achieve a statistical power of 95% for the 1 groups and 6 measurements with an anticipated effect size of 0.25 and a probability level of 0.05 (Faul et al., 2007; Faul et al., 2009). A total of 38 children(25 boys, 13 girls) were recruited through posts in online communities. The participants were elementary school students with a range of ages from 10 to 12 years old (M=10.73, SD = 1.12) who could read Korean messages used in the study, and their and parental consents were obtained prior to study. All participants received a $20 reward after completing the 40-minute experiment.

All the procedures of this study were conducted online. The participants entered the online Zoom meeting room on the scheduled date and time, and each session lasted about 40 minutes. Their parents or tutors were asked to stay freely with the participants but not to intervene during the session. At the beginning of the experiment, we explained the purpose and procedure of the study and asked for permission for the video recording. No participants refused to participate or be recorded.

This study had three parts and the summary of the procedure is shown in Figure 5. This procedure was reviewed and approved by the Institutional Review Board(IRB) of Yonesi University.

Figure 5

Summary of the procedure

4. 4. Data Analysis

The quantitative data were analyzed using IBM SPSS Statistics version 26.0. Of the 38 participants in the experiment, data for 37 were analyzed, and the answers of one who could not concentrate on the experiment were excluded. In the first step, we tested the distribution of our data using the Kolmogorov-Smirnov test, which supported the assumption of normality for both trust (p = .06) and perceived safety (p = .08). To test the main effects of communication style and visualization type, 2 (communication style: advisory vs. informative) × 3 (visualization types: text vs. sign vs. symbol) repeated-measures analyses of variances (ANOVAs) were employed (Table 1 and 2). The interview data were qualitatively analyzed using thematic analysis (Braun & Clarke, 2006) based on our interests in trust and perceived safety. By using open coding procedures, we discovered clusters of similar topics and organized them inductively according to thematic relationships. By reiterating this process until there were no further disagreements, we finalized two main themes with three and four sub-themes each. The themes are shown in Table 3.

Table 1

Descriptive statistics

Table 2

Summary of results in analyses of variance

Table 3

The themes identified in relation to trust and perceived safety

5. Results

In this section, we present our findings from the analysis of the quantitative and qualitative data. We follow this with a detailed description of the results of the analysis of the quantitative and qualitative data.

5. 3. Comparisons of Six Design Concepts

In this section, we demonstrate the frequency analysis for comparison questions of trust, perceived safety, and preference for all six design concepts. The results were consistent with the preceding outcomes but provided somewhat different insights since we allowed multiple responses. A summary of the frequency analysis results of the comparison questions is shown in Figure 6.

Figure 6

Frequency analysis results of the comparison questions

5. 3. 1. Comparison of Trust

Regarding the multitude of responses dealing with the trustworthiness of the design concepts, 25 participants (37.3%) selected informative–text, 17 selected informative–sign (25.4%), 10 selected advisory–text (14.9%), 7 selected advisory–symbol (10.4%), 5 selected informative–symbol (7.5%), and 3 selected advisory–sign (4.5%) to be the most trustworthy. The quantitative results did not show significant differences in the preceding sections; however, the comparison questions that allowed multiple responses showed that the informative communication was often considered trustworthy (Figure 6, left). One of the related themes that can explain this is the “affirmation of predictability.” The participants tended to trust their crossing decisions, which they made by predicting the car’s behavior.

“I’ll choose the second one because this word means it is completely ‘stopped’ literally.” (P20, informative–text)
“It shows [the car is] gradually slowing down in numbers, so I can be certain it will stop.” (P36, informative–sign)

The other related theme is “uncertainty about the environment.” Some participants were concerned about sudden changes in their surroundings, such as other moving cars in the next lane, although advisory communication may still appear in the focus car. In this case, they could trust informative messages that were more trustworthy because they displayed the current status of each vehicle.

“I’m worried about the truck in the next lane. I don’t know whether the truck will keep moving even though the car said ‘cross now.’” (P31, advisory–text)
“How can I believe the message (cross now)? There is a truck in the next lane.” (P4, advisory–text)

5. 3. 2. Comparison of Perceived Safety

Regarding responses to the safety afforded by the design concepts, 29 participants (27.9%) chose advisory–text, 24 chose advisory–symbol (23.1%), 24 chose informative–text (23.1%), 13 chose informative–sign (12.5%), 9 chose advisory–sign (8.7%), and 5 chose informative–symbol (4.8%) as the safest models (Figure 6, middle). The advisory communication style and text message were most often chosen as best at ensuring safety, which is consistent with the results in the preceding section, indicating that the participants felt safe when complying with the instructions, knowing the following cues, and having confidence when interpreting the ECD design. This is related to the themes “affirmation about the instruction,” “uncertainty about the subsequent cues,” and “uncertainty about the interpretation.”

“I’ll select the sixth one because the symbols are just the same as the traffic light. I can just follow the signals.” (P26, advisory–symbol)
“I do know what will come up after the first text, ‘please wait,’ so I think it is safe.” (P12, advisory–text)

Furthermore, the advisory–symbol was also chosen the most as the design that provided a feeling of safety, unlike the quantitative results in the preceding section that revealed symbols were not considered safe. We interpreted this to be the result of consistency in the participants’ mental models and familiarity with images in traffic lights.

“It makes me feel safe because I’m familiar with these images.” (P35, advisory–symbol)
“I’ve especially seen the sixth one a lot. It is the traffic light.” (P10, advisory–symbol)

5. 3. 3. Comparison of Preferences

Regarding the preferred design concepts, 17 participants chose advisory–text (23.0%), 15 chose advisory–symbol (20.3%), 12 chose informative–symbol (16.2%), 11 chose informative–sign (14.9%), 11 chose advisory–sign (14.9%), and 8 chose informative–text (10.8%) (Figure 6, right). There was no significant difference in terms of distribution. However, there were some notable differences in pattern, unlike the preceding results for trust and perceived safety (Figure 6, left and middle), which showed a higher preference for the informative–symbol type compared to informative–sign and informative–text (Figure 6, right). This can be explained by the participants’ emotional responses to the design concept for ECDs as follows:

“It’s fun. It feels like a human.” (P17, informative–symbol)
“The annoyed face and smiling face are so cute” (P35, informative–symbol).

Overall, the participants’ understanding of the design concepts was consistent with our intentions, but some interpreted specific design concepts slightly differently than expected. For example, the “km/h” sign used in the informative–sign concept was read as the distance between the car and crossroad since some participants were yet to learn what “km/h” means. Nevertheless, it was interesting that the children perceived the decreasing numbers to mean that the car was slowing down. The “pause” design used in the advisory–arrow concept was interpreted to mean the car was in play mode, as some participants correlated the two-stick image with YouTube video-playing functionalities. There was a response that the “hand” sign in the advisory–symbol model was the car saying “hello” to pedestrians.

6. Discussion

The present study revealed that the perceptions of child pedestrians with regards to the affirmations and uncertainty of information components in ECDs have a substantial impact on their level of trust and perceived safety towards AVs. Our findings highlight the importance of considering the themes of affirmation and uncertainty during the design process of ECDs for child pedestrians. This approach could enhance trust in AVs while preventing an excessive level of trust.

7. Conclusion

In this study, we explored the effect of the design of ECDs’ communication style and visualization type on child pedestrians’ trust and perceived safety. The findings indicate that child pedestrians exhibit both affirmations and uncertainties in relation to ECDs, necessitating a dual approach to increasing trust while avoiding overtrust. To address the affirmations, we propose the implementation of advisory communication in conjunction with context-based informative messages. To address the uncertainties, it is recommended to preserve child pedestrians’ existing mental models and to incorporate design elements that evoke positive emotions in the ECD design concepts.

Acknowledgments

This work was supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea (NRF-2019S1A5A2A01045980).

References

• Ackermans, S., Dey, D., Ruijten, P., Cuijpers, R. H., & Pfleging, B. (2020). The effects of explicit intention communication, conspicuous sensors, and pedestrian attitude in interactions with automated vehicles. Paper presented at the Proceedings of the 2020 chi conference on human factors in computing systems. [https://doi.org/10.1145/3313831.3376197]
• Bartneck, C., Kulić, D., Croft, E., & Zoghbi, S. (2009). Measurement instruments for the anthropomorphism, animacy, likeability, perceived intelligence, and perceived safety of robots. International journal of social robotics, 1(1), 71-81. [https://doi.org/10.1007/s12369-008-0001-3]
• Bazilinskyy, P., Dodou, D., & De Winter, J. (2019). Survey on eHMI concepts: The effect of text, color, and perspective. Transportation research part F: traffic psychology and behaviour, 67, 175-194. [https://doi.org/10.1016/j.trf.2019.10.013]
• Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative research in psychology, 3(2), 77-101. [https://doi.org/10.1191/1478088706qp063oa]
• Carmona, J., Guindel, C., Garcia, F., & de la Escalera, A. (2021). ehmi: Review and guidelines for deployment on autonomous vehicles. Sensors, 21(9), 2912. [https://doi.org/10.3390/s21092912]
• Cefkin, M., Zhang, J., Stayton, E., & Vinkhuyzen, E. (2019). Multi-methods research to examine external HMI for highly automated vehicles. In HCI in Mobility, Transport, and Automotive Systems: First International Conference, MobiTAS 2019, Held as Part of the 21st HCI International Conference, HCII 2019, Orlando, FL, USA, July 26-31, 2019, Proceedings 21 (pp. 46-64). Springer International Publishing. [https://doi.org/10.1007/978-3-030-22666-4_4]
• Chang, C. M., Toda, K., Igarashi, T., Miyata, M., & Kobayashi, Y. (2018, September). A video-based study comparing communication modalities between an autonomous car and a pedestrian. In Adjunct Proceedings of the 10th International Conference on Automotive User Interfaces and Interactive Vehicular Applications (pp. 104-109). [https://doi.org/10.1145/3239092.3265950]
• Chang, C. M., Toda, K., Sakamoto, D., & Igarashi, T. (2017, September). Eyes on a Car: an Interface Design for Communication between an Autonomous Car and a Pedestrian. In Proceedings of the 9th international conference on automotive user interfaces and interactive vehicular applications (pp. 65-73). [https://doi.org/10.1145/3122986.3122989]
• Charisi, V., Habibovic, A., Andersson, J., Li, J., & Evers, V. (2017, June). Children's views on identification and intention communication of self-driving vehicles. In Proceedings of the 2017 conference on interaction design and children (pp. 399-404). [https://doi.org/10.1145/3078072.3084300]
• Colley, M., Li, S., & Rukzio, E. (2021, September). Increasing pedestrian safety using external communication of autonomous vehicles for signalling hazards. In Proceedings of the 23rd International Conference on Mobile Human-Computer Interaction (pp. 1-10). [https://doi.org/10.1145/3447526.3472024]
• De Clercq, K., Dietrich, A., Núñez Velasco, J. P., De Winter, J., & Happee, R. (2019). External human-machine interfaces on automated vehicles: effects on pedestrian crossing decisions. Human factors, 61(8), 1353-1370. [https://doi.org/10.1177/0018720819836343]
• Deb, S., Carruth, D. W., Fuad, M., Stanley, L. M., & Frey, D. (2020). Comparison of child and adult pedestrian perspectives of external features on autonomous vehicles using virtual reality experiment. In Advances in Human Factors of Transportation: Proceedings of the AHFE 2019 International Conference on Human Factors in Transportation, July 24-28, 2019, Washington DC, USA 10 (pp. 145-156). Springer International Publishing. [https://doi.org/10.1007/978-3-030-20503-4_13]
• Deb, S., Strawderman, L., Carruth, D. W., DuBien, J., Smith, B., & Garrison, T. M. (2017). Development and validation of a questionnaire to assess pedestrian receptivity toward fully autonomous vehicles. Transportation research part C: emerging technologies, 84, 178-195. [https://doi.org/10.1016/j.trc.2017.08.029]
• Deb, S., Strawderman, L. J., & Carruth, D. W. (2018). Investigating pedestrian suggestions for external features on fully autonomous vehicles: A virtual reality experiment. Transportation research part F: traffic psychology and behaviour, 59, 135-149. [https://doi.org/10.1016/j.trf.2018.08.016]
• Dey, D., Habibovic, A., Löcken, A., Wintersberger, P., Pfleging, B., Riener, A., Martens, M., & Terken, J. (2020a). Taming the eHMI jungle: A classification taxonomy to guide, compare, and assess the design principles of automated vehicles' external human-machine interfaces. Transportation Research Interdisciplinary Perspectives, 7, 100174. [https://doi.org/10.1016/j.trip.2020.100174]
• Dey, D., Habibovic, A., Pfleging, B., Martens, M., & Terken, J. (2020b, April). Color and animation preferences for a light band eHMI in interactions between automated vehicles and pedestrians. In Proceedings of the 2020 CHI conference on human factors in computing systems (pp. 1-13). [https://doi.org/10.1145/3313831.3376325]
• Dey, D., Holländer, K., Berger, M., Eggen, B., Martens, M., Pfleging, B., & Terken, J. (2020, September). Distance-dependent eHMIs for the interaction between automated vehicles and pedestrians. In 12th international conference on automotive user interfaces and interactive vehicular applications (pp. 192-204). [https://doi.org/10.1145/3409120.3410642]
• Dey, D., Martens, M., Wang, C., Ros, F., & Terken, J. (2018, September). Interface concepts for intent communication from autonomous vehicles to vulnerable road users. In Adjunct proceedings of the 10th international conference on automotive user interfaces and interactive vehicular applications (pp. 82-86). [https://doi.org/10.1145/3239092.3265946]
• Dey, D., Matviienko, A., Berger, M., Pfleging, B., Martens, M., & Terken, J. (2021). Communicating the intention of an automated vehicle to pedestrians: The contributions of eHMI and vehicle behavior. it-Information Technology, 63(2), 123-141. [https://doi.org/10.1515/itit-2020-0025]
• Faas, S. M., Stange, V., & Baumann, M. (2021). Self-Driving Vehicles and Pedestrian Interaction: Does an External Human-Machine Interface Mitigate the Threat of a Tinted Windshield or a Distracted Driver? International Journal of Human-Computer Interaction, 37(14), 1364-1374. [https://doi.org/10.1080/10447318.2021.1886483]
• Faul, F., Erdfelder, E., Buchner, A., & Lang, A.-G. (2009). Statistical power analyses using G* Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41(4), 1149-1160. [https://doi.org/10.3758/BRM.41.4.1149]
• Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G* power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39(2), 175-191. [https://doi.org/10.3758/BF03193146]
• Fridman, L., Mehler, B., Xia, L., Yang, Y., Facusse, L. Y., & Reimer, B. (2017). To walk or not to walk: Crowdsourced assessment of external vehicle-to-pedestrian displays. arXiv preprint arXiv:1707.02698.
• Gitelman, V., Levi, S., Carmel, R., Korchatov, A., & Hakkert, S. (2019). Exploring patterns of child pedestrian behaviors at urban intersections. Accident Analysis & Prevention, 122, 36-47. [https://doi.org/10.1016/j.aap.2018.09.031]
• Habibovic, A., Lundgren, V. M., Andersson, J., Klingegård, M., Lagström, T., Sirkka, A., . . . Krupenia, S. (2018). Communicating intent of automated vehicles to pedestrians. Frontiers in psychology, 1336. [https://doi.org/10.3389/fpsyg.2018.01336]
• Hoff, K. A., & Bashir, M. (2015). Trust in automation: Integrating empirical evidence on factors that influence trust. Human factors, 57(3), 407-434. [https://doi.org/10.1177/0018720814547570]
• Holländer, K., Colley, A., Mai, C., Häkkilä, J., Alt, F., & Pfleging, B. (2019a, October). Investigating the influence of external car displays on pedestrians' crossing behavior in virtual reality. In Proceedings of the 21st International Conference on Human-Computer Interaction with Mobile Devices and Services (pp. 1-11). [https://doi.org/10.1145/3338286.3340138]
• Holländer, K., Colley, M., Rukzio, E., & Butz, A. (2021, May). A taxonomy of vulnerable road users for hci based on a systematic literature review. In Proceedings of the 2021 CHI conference on human factors in computing systems (pp. 1-13). [https://doi.org/10.1145/3411764.3445480]
• Holländer, K., Wintersberger, P., & Butz, A. (2019b, September). Overtrust in external cues of automated vehicles: an experimental investigation. In Proceedings of the 11th international conference on automotive user interfaces and interactive vehicular applications (pp. 211-221). [https://doi.org/10.1145/3342197.3344528]
• Jayaraman, S. K., Creech, C., Tilbury, D. M., Yang, X. J., Pradhan, A. K., Tsui, K. M., & Robert Jr, L. P. (2019). Pedestrian trust in automated vehicles: Role of traffic signal and av driving behavior. Frontiers in Robotics and AI, 117. [https://doi.org/10.3389/frobt.2019.00117]
• Kiefer, A. W., Pincus, D., Richardson, M. J., & Myer, G. D. (2017). Virtual reality as a training tool to treat physical inactivity in children. Frontiers in Public Health, 5, 349. [https://doi.org/10.3389/fpubh.2017.00349]
• Koekemoer, K., Van Gesselleen, M., Van Niekerk, A., Govender, R., & Van As, A. B. (2017). Child pedestrian safety knowledge, behaviour and road injury in Cape Town, South Africa. Accident Analysis & Prevention, 99, 202-209. [https://doi.org/10.1016/j.aap.2016.11.020]
• Langer, S. K. (1942). Philosophy in a new key: a study in the symbolism of reason, rite, and art.
• Lee, J. D., & See, K. A. (2004). Trust in automation: Designing for appropriate reliance. Human factors, 46(1), 50-80. [https://doi.org/10.1518/hfes.46.1.50.30392]
• Li, Y., Dikmen, M., Hussein, T. G., Wang, Y., & Burns, C. (2018, September). To cross or not to cross: Urgency-based external warning displays on autonomous vehicles to improve pedestrian crossing safety. In Proceedings of the 10th International Conference on Automotive User Interfaces and Interactive Vehicular Applications (pp. 188-197). [https://doi.org/10.1145/3239060.3239082]
• Liu, Y., Lyu, Y., Böttcher, K., & Rötting, M. (2020). External interface-based autonomous vehicle-to-pedestrian communication in urban traffic: Communication needs and design considerations. International Journal of Human-Computer Interaction, 36(13), 1258-1272. [https://doi.org/10.1080/10447318.2020.1736891]
• Löcken, A., Golling, C., & Riener, A. (2019, September). How should automated vehicles interact with pedestrians? A comparative analysis of interaction concepts in virtual reality. In Proceedings of the 11th international conference on automotive user interfaces and interactive vehicular applications (pp. 262-274). [https://doi.org/10.1145/3342197.3344544]
• Mahadevan, K., Somanath, S., & Sharlin, E. (2018, April). Communicating awareness and intent in autonomous vehicle-pedestrian interaction. In Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems (pp. 1-12). [https://doi.org/10.1145/3173574.3174003]
• Matthews, M., Chowdhary, G., & Kieson, E. (2017). Intent communication between autonomous vehicles and pedestrians. arXiv preprint arXiv:1708.07123.
• Morrongiello, B. A., & Barton, B. K. (2009). Child pedestrian safety: Parental supervision, modeling behaviors, and beliefs about child pedestrian competence. Accident Analysis & Prevention, 41(5), 1040-1046. [https://doi.org/10.1016/j.aap.2009.06.017]
• Muir, B. M. (1987). Trust between humans and machines, and the design of decision aids. International journal of man-machine studies, 27(5-6), 527-539. [https://doi.org/10.1016/S0020-7373(87)80013-5]
• Otherson, I., Conti-Kufner, A. S., Dietrich, A., Maruhn, P., & Bengler, K. (2018). Designing for automated vehicle and pedestrian communication: Perspectives on eHMIs from older and younger persons. Proceedings of the Human Factors and Ergonomics Society Europe, 135-148.
• Raats, K., Fors, V., & Pink, S. (2020). Trusting autonomous vehicles: An interdisciplinary approach. Transportation Research Interdisciplinary Perspectives, 7, 100201. [https://doi.org/10.1016/j.trip.2020.100201]
• Rasouli, A., & Tsotsos, J. K. (2019). Autonomous vehicles that interact with pedestrians: A survey of theory and practice. IEEE transactions on intelligent transportation systems, 21(3), 900-918. [https://doi.org/10.1109/TITS.2019.2901817]
• Rosenbloom, T., Ben-Eliyahu, A., & Nemrodov, D. (2008). Children's crossing behavior with an accompanying adult. Safety science, 46(8), 1248-1254. [https://doi.org/10.1016/j.ssci.2007.07.004]
• Rouchitsas, A., & Alm, H. (2019). External human-machine interfaces for autonomous vehicle-to-pedestrian communication: A review of empirical work. Frontiers in psychology, 2757. [https://doi.org/10.3389/fpsyg.2019.02757]
• Salanger, M., Lewis, D., Vallier, T., McDermott, T., & Dergan, A. (2020). Applying virtual reality to audiovisual speech perception tasks in children. American journal of audiology, 29(2), 244-258. [https://doi.org/10.1044/2020_AJA-19-00004]
• Schaefer, K. E., Chen, J. Y., Szalma, J. L., & Hancock, P. A. (2016). A meta-analysis of factors influencing the development of trust in automation: Implications for understanding autonomy in future systems. Human factors, 58(3), 377-400. [https://doi.org/10.1177/0018720816634228]
• Schwebel, D. C., Wu, Y., Swanson, M., Cheng, P., Ning, P., Cheng, X., Gao, Y., & Hu, G. (2018). Child pedestrian street-crossing behaviors outside a primary school: developing observational methodologies and data from a case study in Changsha, China. Journal of transport & health, 8, 283-288. [https://doi.org/10.1016/j.jth.2018.01.005]
• She, J., Neuhoff, J., & Yuan, Q. (2021). Shaping pedestrians' trust in autonomous vehicles: an effect of communication style, speed information, and adaptive strategy. Journal of Mechanical Design, 143(9). [https://doi.org/10.1115/1.4049866]
• Stadler, S., Cornet, H., Novaes Theoto, T., & Frenkler, F. (2019). A tool, not a toy: using virtual reality to evaluate the communication between autonomous vehicles and pedestrians. In Augmented reality and virtual reality (pp. 203-216): Springer. [https://doi.org/10.1007/978-3-030-06246-0_15]
• WHO. (20 June 2022). Road traffic injuries.