This article was downloaded by: [UOV University of Oviedo] On: 22 October 2014, At: 06:53 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Ergonomics Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/terg20

Effects of competing environmental variables and signage on route-choices in simulated everyday and emergency wayfinding situations a

ab

ab

c

Elisângela Vilar , Francisco Rebelo , Paulo Noriega , Emília Duarte & Christopher B. Mayhorn

d

a

Centre for Architecture, Urban Planning and Design (CIAUD), University of Lisbon, Lisbon, Portugal b

Ergonomics Laboratory, University of Lisbon, Cruz Quebrada - Dafundo, Portugal

c

UNIDCOM/IADE, IADE-U - Instituto de Arte, Design e Empresa - Universitário, Lisbon, Portugal d

Department of Psychology, North Carolina State University, Raleigh, NC, USA Published online: 18 Mar 2014.

To cite this article: Elisângela Vilar, Francisco Rebelo, Paulo Noriega, Emília Duarte & Christopher B. Mayhorn (2014) Effects of competing environmental variables and signage on route-choices in simulated everyday and emergency wayfinding situations, Ergonomics, 57:4, 511-524, DOI: 10.1080/00140139.2014.895054 To link to this article: http://dx.doi.org/10.1080/00140139.2014.895054

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions

Ergonomics, 2014 Vol. 57, No. 4, 511–524, http://dx.doi.org/10.1080/00140139.2014.895054

Effects of competing environmental variables and signage on route-choices in simulated everyday and emergency wayfinding situations Elisaˆngela Vilara*, Francisco Rebeloa,b, Paulo Noriegaa,b, Emı´lia Duartec and Christopher B. Mayhornd a

Centre for Architecture, Urban Planning and Design (CIAUD), University of Lisbon, Lisbon, Portugal; bErgonomics Laboratory, University of Lisbon, Cruz Quebrada - Dafundo, Portugal; cUNIDCOM/IADE, IADE-U - Instituto de Arte, Design e Empresa Universita´rio, Lisbon, Portugal; dDepartment of Psychology, North Carolina State University, Raleigh, NC, USA

Downloaded by [UOV University of Oviedo] at 06:53 22 October 2014

(Received 23 September 2012; accepted 7 February 2014) This study examined the relative influence of environmental variables (corridor width and brightness) and signage (directional and exit signs), when presented in competition, on participants’ route-choices in two situational variables (everyday vs. emergency), during indoor wayfinding in virtual environments. A virtual reality-based methodology was used. Thus, participants attempted to find a room (everyday situation) in a virtual hotel, followed by a fire-related emergency egress (emergency situation). Different behaviours were observed. In the everyday situation, for no-signs condition, participants choose mostly the wider and brighter corridors, suggesting a heavy reliance on the environmental affordances. Conversely, for signs condition, participants mostly complied with signage, suggesting a greater reliance on the signs rather than on the environmental cues. During emergency, without signage, reliance on environmental affordances seems to be affected by the intersection type. In the sign condition, the reliance on environmental affordances that started strong decreases along the egress route. Practitioner Summary: Virtual reality was used to study relative influence of environmental variables and signage, when in competition, on participants’ route-choices in everyday and emergency situations. For everyday no-signs condition, findings suggested a reliance on environmental variables. For emergency sign condition, higher reliance on environmental variables was found for first three intersections. Keywords: wayfinding; route-choice; virtual reality; emergency egress; affordances

1.

Introduction

Wayfinding within complex buildings can be problematic under normal circumstances but can literally be life threatening during emergency situations, such as fires, terrorist attacks and natural disasters. While the role of signage in wayfinding is well known, in both everyday and emergency situations, it is not clear how environmental variables influence the users’ wayfinding behaviour, particularly when both types of information are competing. Such situations are very common because buildings are not always planned considering their actual purpose, such as the case of interventions in historical buildings, renovations and changes in buildings’ use. In these cases, the presence of signage positioned in opposition to the paths that the buildings’ visitors use most can result in low proficiency and/or give rise to safety critical situations since it can cause doubts regarding the route-choice. The study of these critical situations, in which signage points to a direction and people may be attracted to the opposite one by some environmental variables (such as corridor width and brightness), may help to design better signage systems as well as safer buildings. Signage plays an important role during wayfinding because it provides directional information that is explicitly available in the environment (Conroy-Dalton 2001). In fact, a signage system can increase indoor wayfinding performance because it facilitates how people find destinations within a complex building in everyday (e.g. Vilar, Rebelo, and Noriega 2012) and emergency situations (e.g. Mantovani et al. 2001). For both studies, the existence of a signage system placed on the floor increased user’s wayfinding performance when compared with signs placed on the wall and with a situation in which no signs were available. The sole focus on signage is not sufficient to guarantee good wayfinding performance within a complex building. There are environmental variables such as corridor width and brightness that can influence route-choice during wayfinding indoors (Vilar et al. 2013). These environmental variables can be considered implicit information, which somehow informs (or attracts) the users to choose a particular path. O’Neill (1991), for example, found that despite the use of signage, building plan configuration exerts a significant influence on wayfinding performance. Other researchers also found patterns of use associated with environmental features (without considering signage) when investigating route-choice of indoor corridors;

*Corresponding author. Email: [email protected] q 2014 Taylor & Francis

Downloaded by [UOV University of Oviedo] at 06:53 22 October 2014

512

E. Vilar et al.

these findings mainly addressed the effects of light (Taylor and Socov 1974; Vilar et al. 2013), preferred direction (e.g. Robinson 1933; Scharine and McBeath 2002; Xie et al. 2011) and corridor width (Vilar et al. 2012, 2013). Main results found in the literature point out that, while considering two alternative corridors, people prefer to follow brighter paths, turn to the right when both available corridors had the same brightness and width, and follow the wider corridor. Vilar et al. (2013) verified that corridors’ brightness is a stronger factor of attraction than corridors’ width and that a higher percentage of route-choices were made favouring the corridors in which variable width and brightness were together in the same alternative corridor. This relationship between explicit (i.e. signage) and implicit (i.e. environmental variables) directional information is not yet completely understood (Tang, Wu, and Lin 2009), especially regarding the critical situations in which static signs point to a direction which is different from the one that most of the people are attracted to by the environmental variables, and should be useful to architects, designers and safety planners by informing them about people’s unexpected behaviour. According to Bourne and Yaroush (2003), people sometimes fail to respond correctly during evacuation procedures associated with emergencies, and these failures could be related to the influence of the environmental variables during the decision-taking. During an emergency, people find themselves in a stressful situation that requires them to navigate to some safe destination inside or outside the building. To reach safety, Kobes, Oberije´, and Duyvis (2010) argue that people use particular mental processes to carry out several actions that can be characterised by three basic activities: (1) awareness of danger by external stimuli (cue validation), (2) validation of and response to danger indicators (decision-making), and (3) movement to/refuge in a safe place (movement/refuge). The two first activities are considered as the pre-movement phase, and the third as the movement phase. These authors also argue that incident evaluations indicate that most fatalities occur in evacuations with a long pre-movement time, and according to Purser and Bensilum (2001), excessively long pre-movement times are expected in sleeping accommodations such as hotels, hostels and domestic dwellings where occupants are widely distributed. Additionally, Purser and Bensilum (2001) also found that in these type of buildings, when sirens are used as auditory alarms, it is quite common for occupants to ignore them completely, particularly those remote from the fire. This finding might be related to a lack of cue validation activity such that people are unaware of the hazardousness of the situation. Time spent during pre-movement and movement phases could be reduced through the effective manipulation of people’s pattern of navigational flow. In this way, the ability of environmental variables to influence people’s directional decisions inside complex buildings could be helpful in reducing the likelihood of injury or death, mainly in situations in which signs are not very salient (e.g. in the presence of smoke, in cluttered environments) or the individual’s available attention is reduced (e.g. stressful situations, when individuals are occupied with other compelling tasks). Environmental variables could benefit egress when in concordance with the exit signs, or make it more difficult when in opposition to them, because when there is stress that might tie up attention capacity, environmental variables require less cognitive resources to communicate directional information. Support for this pattern would be informative, if found. Virtual reality (VR) has been increasingly used to study behaviour during emergency situations (e.g. Duarte et al. 2013; Gamberini et al. 2003). Gamberini et al. (2003) used VR to examine how people respond during a fire in a public library by manipulating variables such fire intensity and the initial distance to the emergency egress. Their results suggest that participants seemed to recognise a dangerous situation within the context of a simulation and readily produced adaptive responses, thereby indicating that VR is a suitable venue for emergency simulations. In a study that used VR to investigate evacuation times in buildings, Shih, Lin, and Yang (2000) verified that in some situations people followed routes that were different from those indicated by the egress signs and, generally, regardless the presence of signs and smoke, they tried to return the direction they entered the building. This trend was also verified by Xie et al. (2011). Tang, Wu, and Lin (2009) found evidence of the influence of environmental variables over compliance with exit signs. In their study using VR, they reported that when participants were faced with seemingly contradictory information in the form of both an exit sign and an exit door, almost half of the persons choose to proceed through the door rather than follow the directions posted on the sign. In a previous study (Vilar et al. 2013), which had as its objective to verify the influence of environmental variables in directing people while in a simulated emergency situation, we found that the existence of a corridor which was brighter than another one attracted users to the brighter path. It also occurred when a corridor was wider than another; the wider corridor was considered more attractive to the users. The present study expands the previous one to answer two main questions. (1) When facing conflicting directional information (environmental variables vs. signage) during a wayfinding task, will people favour environmental variables or signage? (2) If a pattern exists, does it change according to the situational variables (i.e. everyday vs. emergency situations)? Considering these issues, the current study aims to explore the relative influence of environmental variables (i.e. corridor width and brightness) and signage in directing people during natural movement indoors that vary by level of urgency (everyday situation vs. emergency egress). Two contexts were considered: (1) when trying to find a location within a complex building in a hurry, and (2) when trying to escape from a fire in the building. To conduct the study, a virtual

Ergonomics

513

Downloaded by [UOV University of Oviedo] at 06:53 22 October 2014

building was designed and a VR-based methodology was used to facilitate the manipulation and control of the variables, to allow the exposition of participants to a stressful emergency situation without submitting them to a real hazard, as well as to increase the ecological validity of the study. Given the goals of the current research as well as previous results from Vilar et al. (2013), three hypotheses were formulated: (H1) if no signage is available, regardless of the situations (i.e. everyday – not being late to a meeting or emergency – trying to escape from a fire), people tend to rely on the environmental variables (i.e. corridor width and brightness) to find the way; (H2) when signage and environmental variables are presented in competition (i.e. critical situation), the signage influence on users’ route-decision will be lower in the emergency situation; (H3) from the previous hypothesis, a supplementary one can be deduced, which predicts the existence of an interaction effect such that situational context (i.e. everyday vs. emergency situation) and the presence or absence of contradictory information (i.e. with or without signs) also influence the percentage of choices towards a particular direction. To test these hypotheses, we compared the wayfinding behaviour (i.e. route-choices) using four experimental conditions: no-signs versus signs and everyday versus emergency situations. 2. Method 2.1. Participants Sixty-four university students were randomly assigned to the two groups (i.e. sign and no-sign) as follows: . For the everyday and emergency situations with no-signs: 32 participants were equally distributed with regard to gender. They were aged between 18 and 35 years (mean age ¼ 21.88 years, SD ¼ 3.62). Thirty participants were right-handed and two declared themselves to be left-handed. . For the everyday and emergency situations with signs: 32 participants were equally distributed with regard to gender. They were aged between 18 and 31 years (mean age ¼ 22.31 years, SD ¼ 3.44). Thirty declared themselves, through a questionnaire, to be right-handed and two declared themselves to be left-handed. All participants had normal sight or wore corrective lenses, and no colour vision deficiencies were detected through the Ishihara test (Ishihara 1988). They also reported no physical or mental conditions that would prevent them from participating in a VR simulation. 2.2.

Experimental settings

The VE was projected onto a screen using a stereoscopic projector (i.e. Lightspeed DepthQ 3D) and visualised by the participants through active glasses (i.e. MacNaughton Inc.’s APG6000). The projected image size was 1.72 m (59.78 of horizontal field of view – FOV) by 0.95 m (35.28 of vertical FOV) with an aspect ratio of 16:9. The observation distance (i.e. the distance between the observers’ eyes and the screen) was 1.50 m. A LogitechwAttacke 3 joystick was used as a navigation device. The movement speed inside the virtual environment (VE) gradually increased from stop (0 m/s) to a maximum speed (3 m/s). Wireless headphones (i.e. Sonyw MDRRF800RK) allowed the participants to listen to instrumental ambient music, the wayfinding task instructions given orally by the virtual characters and the sounds of a fire siren and fire. 2.3.

Study design

The influence of the environmental variables (i.e. corridor width and brightness) and signage in directing the participants towards an intended direction (i.e. wayfinding behaviour) was tested in four experimental conditions resulting from to the combination of two factors, the situational context and the signage condition. The study used a mixed within- and betweensubjects design. The within-subjects factor is the situational context that has two categories (i.e. everyday and emergency conditions). The between-subject factor is the signage presence (i.e. sign and no-sign conditions). The dependent variable is the percentage of choices favouring an intended direction (environmental variables direction) in 12 corridor intersections. 2.4. 2.4.1.

Experimental task Critical situation

The critical situation was created taking into account the existence of contradictory information that was manipulated by inserting posted static signs pointing to the opposite direction of the corridors that were the most chosen by participants. The most chosen corridors were selected from a previous study carried out by Vilar et al. (2013) about navigational choice

Downloaded by [UOV University of Oviedo] at 06:53 22 October 2014

514

E. Vilar et al.

preferences associated with environmental variables (i.e. corridor width and brightness). Next, we give details on how such corridors were evaluated during the referred previous study (Vilar et al. 2013). Images of corridor intersections with two alternative paths were randomly presented to the participants using a constant stimuli method combined with a two-forced choice method to collect the participants’ response. Thus, participants had to choose what path, of two available options, they would follow to leave a building in an emergency situation. Two types of corridors configuration were considered (i.e. ‘T-type’ and ‘F-type’), differing in the continuity of the line of sight. In this way, for ‘T-type’ intersection, both alternative corridors have the same line of sight, while for the ‘F-type’ one of the alternative corridors had a more straight line of sight (the front corridor) than the other one (the side corridor). The environmental variables considered were corridor width and corridor brightness, and a total of 57 different situations were generated by the manipulation of these two environmental variables. According to these previous results, people were influenced by environmental variables, preferring to follow wider and brighter corridors, and that brightness is a stronger factor of attraction than width. Vilar et al. (2013) verified that the variables corridor width and brightness, can act as factors of attraction, enhancing corridors affordances and the likelihood of selecting a particular path during the wayfinding process. Thus, corridor width and brightness can be considered environmental affordances (implicit information) that somehow inform the users of what path to choose. For the current study, only 12 corridors intersections (Figure 1) of those 57 different situations studied earlier (Vilar et al. 2013) were selected. Corridor selection followed the criteria: (1) the most chosen corridors considering the available alternative corridors (i.e. left vs. right, front vs. left, and front vs. right) for each situation (i.e. only corridor width, brightness enhanced in the wider corridor and brightness enhanced in the narrower corridor), (2) the intersection type (i.e. ‘T-type’ and ‘F-type’) and (3) the narrower corridor when the difference across the percentage of choices was less or equal to 1%. The factors of attraction (i.e. corridor width, brightness enhanced in wider corridor and brightness), attractor’s direction (i.e. left, right and front) and percentages of choices towards the attractor attained by Vilar et al. (2013) for the 12 selected corridor intersections can be seen in Table 1. 2.4.2.

Virtual environment (VE) – the hotel

For the design of the hotel building used in the present study, 12 corridor intersections previously selected (Table 1) from Vilar et al. (2013) were used. These 12 corridor intersections were mixed and then randomly divided into three groups of four corridors each that comprise three sections of the building floor plan. Each section was designed to have the same travel distance, regardless of participants’ directional choice at each choice point. The virtual hotel was generated based on requirements operationalised during systematic meetings involving experts in ergonomics, architecture, psychology, design and computer engineering. The requirements for the VE development were the following: . represent the context of a hotel and convention centre; . consider a scenario where the participant is going to present a lecture at a conference but he/she is late to the event;

Figure 1.

The 12 selected ‘T-type’ and ‘F-type’ corridor intersections from Vilar et al. (2013).

Ergonomics

515

Table 1. Percentages of choice for the 12 most chosen corridor intersections from Vilar et al. (2013) used as the basis for the design of the virtual building and sign placement. Corridor intersection

Downloaded by [UOV University of Oviedo] at 06:53 22 October 2014

C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12

Variable (attractor) Width Brightness Brightness and Brightness and Brightness Brightness and Brightness and Width Width Brightness Brightness Width

width width width width

Direction

% of choices towards the attractor

Right Front Left Right Left Left Right Right Left Front Right Left

72.05 75.83 87.87 89.58 81.67 91.25 89.58 63.75 72.92 78.33 83.68 57.50

. consider a multilevel virtual three-dimensional building with two levels – the first level, where participants will perform three wayfinding tasks considering an everyday situation, and the second level, in which an emergency event will occur and participants have to escape from a fire (i.e. emergency situation); . use the same floor plan for first and second levels; . use the 12 corridors (Table 1) selected from previous research conducted by Vilar et al. (2013); . distribute the 12 corridors in three consecutive sections of the building’s plan with a room between each section (Figure 2); . consider three wayfinding tasks (i.e. 1 – find the Europe Ballroom to pay the registration fee; 2 – find the information desk at the America Ballroom; 3 – find the elevators to go to the second floor and find the Oceania Auditorium where the lecture will occur), numbered on Figure 2;

Figure 2. Top view of the floor plan with the three sections and with the location of the 12 selected corridor intersections. Numbers 1, 2 and 3 show where the wayfinding instructions were delivered to the participants.

516

E. Vilar et al.

Downloaded by [UOV University of Oviedo] at 06:53 22 October 2014

Figure 3.

Examples of fire with smoke in the second floor of the VE during the emergency situation.

. consider a controlled navigation approach – create strategies to avoid backtracking (e.g. closing doors after a routechoice); . maintain the same travelled distance for all sections from the starting point (when the task is given) to the destination point; . use virtual characters (embodied autonomous agents [Allbeck and Badler 2002]) to provide the instructions for wayfinding tasks; . insert fire and smoke to create the emergency situation in the second floor (Figure 3); . insert decorative elements (e.g. furniture, paintings) and sounds (e.g. ambient music, emergency alarm, fire) to create a more realistic environment. In experimental conditions where posted signs were available, directional signs were inserted in the ground floor (i.e. everyday condition) and exit signs were inserted in the second floor (i.e. emergency condition) of the virtual building. To create the critical situations, the signs were always positioned to point to the directions opposite to those that were considered the most probable choice (see Table 1) according to the results of the study conducted by Vilar et al. (2013). The directional signs used for this study had the same features of the vertical directional system tested by Vilar, Rebelo, and Noriega (2012) that verified the effectiveness of this signage configuration in facilitating wayfinding tasks. Exit signs are symbol-based and consistent with the International Organization for Standardization’s 3864-1 (ISO 2002) standard. ISO standard exit signs are required by law to illustrate an arrow and running figure in a doorway. Figure 4 shows examples of signs (i.e. directional and emergency egress) placed in the VE.

Figure 4. Top image shows an example of the directional sign in the first floor and the bottom image shows an example of the exit signs in the second floor.

Ergonomics

517

Downloaded by [UOV University of Oviedo] at 06:53 22 October 2014

2.4.3. Scenario and wayfinding task To increase participant involvement, a scenario was created. At the beginning of the experimental session, the following cover story was given to the participants:You were invited to give a lecture in an important conference at a hotel and conference centre. The conference staff told you that you must talk with the receptionists to complete your registration at the conference and to know the location where the lecture will take place. As the city has a lot of traffic, you are late for your presentation. Please, complete your registration and find your lecture room as soon as possible.Participants were also told that they should behave as they would in a real-life situation. The receptionist (virtual character) would help participants to complete his/her conference registration and to know the location where the presentation will be made. When the participant reaches the second level where the presentation is to occur, he/she is informed that a fire has been detected on the premises. It was considered a controlled navigation approach because the corridors already passed by the participant were closed by doors during everyday wayfinding and by fire and smoke in the emergency situation. Thus, for each choice point, when participants chose one of the two alternative corridors, the corridor of the path that was not chosen was closed by a door (or fire), forcing them to continue along their initial selected path. At the beginning of each section, there was a room that was used to deliver the wayfinding task via virtual characters and to calculate the participants’ partial performance. 2.5.

Procedure

Before starting the VR-based components, all participants were asked to sign a consent form and advised that they could stop participation at any time. The average duration of each experimental session was approximately 30 minutes, divided into a training session and an experimental session. Participants were told that the experimenter’s objective was to evaluate new software for VR simulation, so they ought to fulfil the tasks as accurately and as quickly as possible. Participants were unaware of the real objective of the experiment. The VR-based wayfinding portion of the study began with the training session which had the following main objectives: (1) to familiarise participants with the simulation set-up; (2) to allow them to practise the use of navigation and visualisation devices, to bring their virtual movements closer to their realistic/natural actions; (3) to homogenise differences in the participant’s performance using joystick; and (4) to make a preliminary check for symptoms of simulator sickness (participants were asked to report whether they felt any discomfort). Participants were encouraged to explore freely and navigate into the VE, as quickly and efficiently as they could, without time restrictions. Participants were further instructed to inform the experimenter when they felt relaxed and comfortable with the equipment. The researcher also monitored participants’ control of the navigation device by verifying their accuracy in executing some tasks, such as circumnavigating a pillar placed in the middle of a room without bumping into this element and walking through a zigzag corridor without touching the walls. Only after verifying some of these equipment-related skills did the researcher permit the participant to start the experimental session. No dialogue between the participant and the researcher was allowed after the simulation started. The interaction started in the ground floor of a hotel and convention centre where participants received three wayfinding tasks (i.e. find three different locations in the building) from virtual characters present in the VE. The last task sent the participants to the second floor of the building via an elevator. Once they exited the elevator, an auditory alarm sounded and they were prevented from further elevator use. Thus, participants were faced with finding an emergency egress point by navigating through the second floor to escape from the fire. Smoke and flames always appeared behind the participants in the VE, thereby closing the corridor that they had already passed and preventing retreat in that direction. If the participants reached a time limit of 20 minutes inside the simulation, the experimental session was stopped to prevent eye fatigue, or simulation sickness, or both. Simulator sickness was mainly evaluated through participants’ verbalisations. However, the researcher also monitored them during the interaction for symptoms such as redness of face, nausea, dizziness and sweating (Kennedy, Hettinger, and Lilienthal 1990; Keshavarz and Hecht 2011). At the end of the experimental test, a post-task questionnaire was used to collect demographic information such as age, gender, occupation and dominant hand. Participants were also asked to answer, in seven-point scale format, questions related to their perceived hazard and overall involvement during the interaction with the simulation. 2.6.

Results and discussion

Criteria for presenting results are related to the choices favouring the environmental variables (i.e. corridor width and brightness) direction in the everyday and emergency situation in no-signs versus signs conditions. Even in the signs condition, in which it was expected that participants would comply with signs, for simplicity of results presentation, the same presentation criteria was maintained. Participants’ route performance considers the directional choices recorded for the entire route (12 corridor intersections). Table 2 summarises the results obtained for all conditions. The corridors are

a

72.05 75.83 87.87 89.58 81.67 91.25 89.58 63.75 72.92 78.33 83.68 57.50

Right Left Left Right Right Left Front Right Left

% of choices towards the environmental variablea

Right Front Left

Variable directiona

Predicted results were attained from Vilar et al. (2013) study.

Width Brightness Brightness and width C4 (F) Brightness and width C5 (T) Brightness C6 (F) Brightness and width C7 (T) Brightness and width C8 (F) Width C9 (T) Width C10 (F) Brightness C11 (T) Brightness C12 (F) Width Participants’ route performance (%) SD

C1 (T) C2 (F) C3 (T)

Corridor (inter- Variable (presection type) dicted attractor)a

53.10 68.80 40.60 71.90 40.60 64.84 11.54

68.80

81.30 65.20

84.40

65.60 75.00 65.60

No-signs % choice towards environmental variable

3.10 0 0 3.10 12.50 3.12 5.07

0

0 9.40

3.10

0 0 6.20

Signs % choice towards environmental variable

Everyday situation

56.30 37.50 68.80 71.90 28.10 61.98 14.03

78.10

78.10 43.80

31.30

78.10 81.30 90.60

No-signs % choice towards environmental variable

15.60 6.20 6.20 6.20 6.20 18.49 21.66

21.90

12.50 15.60

6.20

65.60 31.20 28.10

Signs % choice towards environmental variable

Emergency situation

Table 2. Results considering the predicted directions from previous study (Vilar et al. 2013), participants’ route performance and percentages of choices favouring the corridors with the environmental variables for everyday and emergency situations in no-signs versus signs conditions. Corridors were arranged according to their disposition on the building’s plan.

Downloaded by [UOV University of Oviedo] at 06:53 22 October 2014

518 E. Vilar et al.

Ergonomics

519

presented according to their disposition on the building’s plan. All statistical analyses were conducted using IBM SPSS v.20. The statistical significance level was set at 5%.

Downloaded by [UOV University of Oviedo] at 06:53 22 October 2014

2.6.1.

Everyday versus emergency situation

Considering the participants’ behaviour regarding the route-choice in the studied situations, the McNemar test was used to evaluate if the situational context (everyday vs. emergency) significantly affected the choices towards the environmental variable (i.e. corridors width and brightness in three situations, wider corridors, wider and brighter corridors, and narrower and brighter corridors). Data from the no-signs condition were analysed and results from the McNemar test are presented in Table 3, separated by intersection type. Overall results suggest that, when decisions were made between left and right corridors (T-type), participants did not change their behaviour (i.e. if in the everyday situation, they decided to follow the wider corridor in an intersection, in the emergency situation they tend to follow the same pattern of use). Changes were verified only for C3 and C9. For the C3 corridor intersection, the percentage of choices towards the direction of the environmental variable increased in the emergency situation. However, for intersection C9, when in the emergency situation participants changed their behaviour and preferred to follow the left path, instead of the wider path on the right side (preferred in the everyday situation). It could be explained by the ‘direction strategy’ (Ho¨lscher et al. 2006), since before facing this decision, participants crossed diagonally a room from the right to the left side, so the next decision to the left minimised their global deviation. When front corridors were available (‘F-type’ intersections), changes were also observed for intersections C4 and C10. The presence of a front corridor in the emergency situation noticeably influenced route selection behaviour. The predicted direction for C4 was to the right. The predicted behavioural trend was observed in the everyday situation; however, for the emergency situation, most of the participants preferred to follow the front corridor. For C10, while in the everyday situation there may have been task-related influence that acted to direct people towards the side corridor, when exposed to the emergency situation (without this task influence) participants preferred to follow the front corridor as predicted. To verify if the signage has more influence on participants’ route-choice in the everyday than in the emergency situation, as predicted by the second hypothesis, the McNemar test was performed considering the situational context in the signs conditions. Data from the signs condition were analysed and results from the McNemar test are presented in Table 4, separated by intersection type. Overall, results suggest that for the signs conditions, the effect of the environmental variables was stronger for the emergency situation, resulting in lower influence of signage influence on participants’ route-decision. However, statistical analysis revealed that only in four intersections (i.e. C1, C2, C3 and C7), the participants changed their behaviour influenced by the situational context. For these four intersections, the lower influence of signage was verified. 2.6.2. Everyday situation (no-signs versus signs) No-Signs condition. Data related to the everyday situation when no signs were available shows that, in this condition, participants choose mostly the corridors in which the variables width and brightness were manipulated, suggesting a heavy reliance on the environmental affordances. Table 3. Percentages of directional choices attained in everyday and emergency situations, considering no-signs conditions with the results from the McNemar change test, separated by intersection type.

Intersection type ‘T-type’

‘F-type’

Corridor intersection C1 C3 C5 C7 C9 C11 C2 C4 C6 C8 C10 C12

Everyday

Emergency

% choices towards environmental variable

% choices towards environmental variable

65.60 65.60 81.30 68.80 68.80 71.90 75.00 84.40 65.20 53.10 40.60 40.60

78.10 90.60 78.10 78.10 37.50 71.90 81.30 31.30 43.80 56.30 68.80 28.10

McNemar change test

x (1) x 2(1) x 2(1) x 2(1) x 2(1) x 2(1) x 2(1) x 2(1) x 2(1) x 2(1) x 2(1) x 2(1) 2

¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼

0.643; p ¼ 0.424; N 6.125; p ¼ 0.008; N 0.000; p ¼ 1.000; N 0.364; p ¼ 0.549; N 5.062; p ¼ 0.021; N 0.000; p ¼ 1.000; N 0.083; p ¼ 0.774; N 13.474; p , 0.001; N 2.083; p ¼ 0.146; N 0.000; p ¼ 1.000; N 4.267; p ¼ 0.035; N 0.643; p ¼ 0.424; N

¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼

32 32 32 32 32 32 32 32 32 32 32 32

520

E. Vilar et al.

Table 4. Percentages of directional choices attained in everyday and emergency situations, considering signs conditions with the results from the McNemar change test, separated by intersection type.

Intersection type ‘T-type’

Downloaded by [UOV University of Oviedo] at 06:53 22 October 2014

‘F-type’

Corridor intersection C1 C3 C5 C7 C9 C11 C2 C4 C6 C8 C10 C12

Everyday

Emergency

% choices towards environmental variable

% choices towards environmental variable

0 6.20 0 0 0 3.10 0 3.10 9.40 3.10 0 12.50

65.60 28.10 12.50 21.90 6.20 6.20 31.20 6.20 15.60 15.60 6.20 6.20

McNemar change test

x (1) x 2(1) x 2(1) x 2(1) x 2(1) x 2(1) x 2(1) x 2(1) x 2(1) x 2(1) x 2(1) x 2(1) 2

¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼

19.048; 4.000; 2.250; 5.143; 0.500; 0.000; 8.100; 0.000; 0.167; 1.500; 0.500; 0.167;

p p p p p p p p p p p p

, ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼

0.001; N 0.039; N 0.125; N 0.016; N 0.500; N 1.000; N 0.002; N 1.000; N 0.688; N 0.219; N 0.500; N 0.688; N

¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼

32 32 32 32 32 32 32 32 32 32 32 32

Binomial tests were performed considering the percentages of choices for the ‘T-type’ intersections for the three manipulated situations with the variables corridor width and brightness (i.e. wider corridors, brighter and wider corridor, brighter and narrower corridors). Results from the binomial test revealed that participants made their choices favouring: (1) wider corridors (observed proportion ¼ 0.672, p ¼ 0.004, one-tailed, N ¼ 64), (2) brighter and wider corridors (observed proportion ¼ 0.632, p ¼ 0.004, one-tailed, N ¼ 64) and (3) brighter and narrower corridors (observed proportion ¼ 0.766, p , 0.001, one-tailed, N ¼ 64). These results suggest the environmental variable corridor brightness is a stronger factor of attraction than the environmental variable corridor width. When ‘F-type’ intersections were analysed, results from the binomial test showed that participants only made their choices favouring the environmental variable when it was the situation with brighter and wider corridor option (observed proportion ¼ 0.734, p , 0.001, one-tailed, N ¼ 64). However, for this intersection type, the continuous navigation approach used for the current experiment, in which this corridor was linked to others and participants had to accomplish a task, could have reinforced the ‘direction strategy’ (Ho¨lscher et al. 2006). According to these authors, people tend to choose a direction and minimise their global deviation, trying to maintain a straight heading whenever possible such that they avoid local directional deviations at junctions. Such a ‘direction strategy’ could explain the unexpected results found for the intersection C12, where participants mostly followed the front corridor instead of a side corridor, which was wider. For intersection C10, while it was expected people would follow the front corridor, the majority (59.4%) preferred to follow a side corridor. However, for this specific situation, an influence of the task may serve to explain the observed behaviour. Because participants were tasked with finding a room and that, for this particular intersection, they saw a brighter corridor in front of them without any door; they might have anticipated that their decision options were limited. Signs condition. In contrast to the no-signs condition, results attained from the everyday situation with available signs show that participants mostly complied with signage, suggesting a greater reliance on the signs rather than on the environmental cues. 2.6.3.

Emergency situation (no-signs versus signs)

No-signs condition. In this condition, when participants were in an emergency situation without emergency signs available, reliance on environmental affordances, which was found strong in 8 of the 12 intersections, seems to be affected by the intersection type. Binomial tests were performed considering the percentages of choices for the ‘T-type’ and ‘F-type’ intersections for the three manipulated situations with the variables corridor width and brightness (i.e. wider corridors, brighter and wider corridor, brighter and narrower corridors). Results for the ‘T-type’ intersections revealed that participants made their choices favouring: brighter and wider corridors (observed proportion ¼ 0.84, p , 0.001, one-tailed, N ¼ 64), and brighter and narrower corridors (observed proportion ¼ 0.75, p , 0.001, one-tailed, N ¼ 64). However, when only corridor width was considered, participants seemed to make their directional choices randomly (observed proportion ¼ 0.58, p ¼ 0.130, one-tailed, N ¼ 64). As found for the everyday situation, the emergency condition results for ‘T-type’ intersection suggest corridor brightness as a stronger environmental variable than the corridor width.

Downloaded by [UOV University of Oviedo] at 06:53 22 October 2014

Ergonomics

521

When ‘F-type’ intersections were analysed, results from the binomial test showed that, in emergency situations, the front corridor had a strong influence on participants’ decision. In this type of intersection, results from binomial tests indicated that participants made their choices favouring the front corridor when the side corridor was wider and brighter (observed proportion ¼ 0.63, p , 0.030, one-tailed, N ¼ 64) and when it was wider and darker (observed proportion ¼ 0.75, p , 0.001, N ¼ 64). When only the corridor width was considered, participants seemed to make their choices randomly (observed proportion ¼ 0.58, p ¼ 0.130, one-tailed, N ¼ 64). In the emergency situation, the influence of the front corridor was more noticeable, suggesting that participants might have used wayfinding strategies such as the ‘Least-angle strategy’ (e.g. Conroy-Dalton 2001, 2003; Hochmair and Frank 2000; Hochmair and Karlsson 2005), in which people tend to proceed in the direction of the target suggesting a preference for most direct direction, and/or the ‘direction strategy’ (Ho¨lscher et al. 2006). This influence can be observed for corridor intersections C4, C6 and C12, where participants preferred to follow the front corridor, which was narrower and/or darker than the side corridor. An unexpected result was also found for intersection C9. In this particular ‘T-type’ situation, participants found the choice point after crossing a room diagonally such that turning right could represent an effect of direction strategy. The decision in this choice point might be influenced only by a differentiation between corridor widths (right was narrower than left) and it is possible that the variable in this situation was not strong enough to superimpose a wayfinding strategy. Also, participants might be aware of proximal temporal and spatial cues (e.g. the fire direction) that could influence their decisions. Signs condition. In the sign condition, the reliance on environmental variables decreases along the egress route, with a highest reliance found in the first three intersections. For the first intersection (C1), 65.6% of the participants preferred to follow the wider corridor, which was in the opposite direction to that posted on the emergency egress sign, following the trend found in the no-signs condition. Second and third corridor intersections (i.e. C2 and C3) also presented a considerable percentage of choices favouring the direction of the environmental variable (opposite to the way posted on the sign). Data from this study do not allow us to determine the causality of the observed route-choice, but it suggests that the ISO-type exit sign posted in this situation was not enough to produce the desired safety behaviour in the observed situation (emergency egress). Potential explanations for these findings are task differences or sign salience. Regarding the first one, the lack of emergency validation in the pre-movement phase can explain the observed behaviour. According to Kobes, Oberije´, and Duyvis (2010), people perform first a cue validation where they become aware of the danger from external stimuli. In this study, even though an auditory sign (fire siren) could be heard and smoke and flames could be seen from the beginning of the emergency situation, participants who were engaged in an everyday wayfinding situation may have not validated the danger in the initial moments. The other potential explanation for the findings is related to sign salience. The fact that participants were facing an emergency situation may have affected their attentional capacity and, consequently, more salient signs would be required to attract attention. However, the findings may need further validation with the collection of other data (e.g. eye-gaze, participants self-reports) before general application. 2.6.4.

Influence of situational context and signage on participants’ route performance

A 2 £ 2 mixed ANOVA was performed considering signage as the between-subjects factor, and situational context as the within-subjects factor. The mixed ANOVA results yielded a significant interaction effect (F(1, 62) ¼ 13.903, p , 0.001), indicating that the effect of the situational context was not the same in the no-signs and signs and conditions (everyday no-signs: M ¼ 64.84, SD ¼ 11.54; everyday signs: M ¼ 3.13, SD ¼ 5.08; emergency no-signs: M ¼ 61.98, SD ¼ 14.03; emergency signs: M ¼ 18.49, SD ¼ 21.66). Participants’ route performance was calculated considering the choices favouring the predicted direction made by the participants along the entire route. In this way, higher percentages indicate that participants were more likely to follow the corridors with the environmental variables along the route. Results indicate that when posted signs were inserted the percentage of participants’ route performance was significantly higher for the emergency than for the everyday situation, suggesting that participants followed the direction of the corridors with the environmental variables. However, when signs were absent, percentages of participants’ route performance did not differ in the two situational contexts (i.e. every day and emergency situations). There were significant main effects for the situational context (F(1, 62) ¼ 6.537, p ¼ 0.013) and for the signage condition (F(1, 62) ¼ 400.102, p , 0.001). Figure 5 shows a graph with results. The percentages of choices favouring environmental variables direction are higher in the emergency sign condition than in the everyday signs condition. This result could be explained by the Yerkes –Dodson law (Wickens 1992), which states that the relationship between the level of arousal and performance is defined by an inverted U-shaped function. In this way, performance increases until an optimum level of arousal is reached and high arousal leads to the performance degradation. Such an effect could be explained by a degradation in performance that occurs as a consequence of the selectivity of

522

E. Vilar et al.

Downloaded by [UOV University of Oviedo] at 06:53 22 October 2014

Figure 5.

Graph with the two factors, situational context and signage, considering the variable participants’ route performance (%).

attention or ‘tunnelling’ to different environmental or internal cues, narrowing the spotlight of attention (Easterbrook 1959), which is a more specific effect of the emergency situation. Results reported by Loftus, Loftus, and Messo (1987) lend credence to this ‘tunnelling’ explanation because their efforts to examine eye movements as an indicator of the location of overt attention, verified through eye tracking that the presence of a weapon (vs. a non-weapon condition) elicited eye movements consistent with attentional narrowing. Tang, Wu, and Lin (2009) verified lower percentages of compliance with an exit sign when it was at the first intersection where participants were confronted with this directional information. However, in the second and third intersections, the percentage of persons attending to the signs increased. This pattern of behaviour was also observed in this study, when analysing the higher percentages of choices favouring the corridors with the environmental variables (which were the opposite of the direction posted on the exit signs).

2.6.5.

Post-task questionnaire

The post-task questionnaire included questions related to the perceived level of hazard and overall involvement during the interaction with the simulation. Two specific questions of interest were asked: (1) ‘How would you rate the hazard level on the ground floor (from reception desk, where you received directions to the elevator)?’ (2) ‘How would you rate the hazard level on the second floor (when you exited the elevator until you reached the exit door)?’ Participants responded using a seven-point rating scale (1 – very low; 7 – very high). When asked to classify the perceived hazard level on the ground floor (i.e. everyday situation), 11 participants classified it as medium (N ¼ 64, Median ¼ 4, IQR ¼ 2). However, when asked about the second floor (i.e. emergency egress), 11 classified it as very high (N ¼ 32, Median ¼ 6, IQR ¼ 2). The Wilcoxon matched-pairs signed-ranks test was performed and results showed statistically significant differences between the declared perceived hazard level in both floors (T ¼ 15.25, z ¼ 2 5.131, p , 0.001, one-tailed, N ¼ 64). This finding allowed us to perform a manipulation check because it was confirmed that participants perceived differences between the two simulated situations.

3.

Conclusion and limitation

The current study examined the relative influence of environmental variables (i.e. corridor width and brightness) and signage (i.e. directional and exit signs), when presented in competition, on participants’ route-choices in two situational variables (i.e. everyday vs. emergency situation), during an indoor wayfinding task in VE. Three hypotheses were formulated. The first hypothesis (H1) considered that if no signage is available, regardless of the situations (i.e. everyday and emergency), people tend to rely on the environmental affordances (i.e. corridor width and brightness) to find the way. It was partially confirmed due to the fact that this hypothesis is true only for the ‘T-type’ intersection in the everyday situation. For the emergency situation, considering ‘T-type’ intersections, the corridor width variable did not seem to influence the participants’ route-choice. For intersections with a front corridor as an available alternative (i.e. ‘F-type’) some wayfinding strategies such as the ‘Least-angle strategy’ (e.g. Conroy-Dalton 2001, 2003; Hochmair and Frank 2000; Hochmair and Karlsson 2005) and the ‘direction strategy’ (Ho¨lscher et al. 2006) seem to be used by the participants and overlap the studied environmental variables. For everyday situations, only when the side corridor was the brightest and widest, participants tend to follow its direction. For emergency situations, the influence of the front corridor was more noticeable, with participants preferring to follow narrower and dark front corridors instead of a wider and brighter side corridors.

Downloaded by [UOV University of Oviedo] at 06:53 22 October 2014

Ergonomics

523

The second hypothesis (H2) considered that when signage and environmental variables are presented in competition, the signage influence on participants’ route-decision will be lower in the emergency situation. Overall, descriptive results suggested that the effect of the environmental variables was stronger for the emergency situation, resulting in lower influence of signage influence on participants’ route-decision. Nonetheless, only in four intersections was the lower influence of signage in the emergency conditions statistically verified. Given the observation that high percentages of participants in an emergency and stressful condition chose not to follow the signs (Table 2), these choices are disturbing because lives may be at risk. The existence of an interaction effect, such that situational context (i.e. everyday vs. emergency situation) and the presence or absence of contradictory information (i.e. with or without signs) also influence the percentage of choices towards a particular direction, served to confirm the third hypothesis (H3). When posted signs are available in both situational contexts (i.e. everyday and emergency) creating a critical situation, behavioural differences were observed when participants had to decide whether they should follow the corridors with the environmental variables in 12 corridors intersections along the route. Higher percentages of disagreement between participants’ route-choices and direction posted by exit signs were found in the emergency situation. Attained results were consistent with those reported by Tang, Wu, and Lin (2009), with the percentages of route-choices where participants followed the corridors with the environmental variables decreased along the escape route suggesting that, with the repeated exposure to an exit sign people increased their compliance with it. It is important to highlight that during the entire emergency egress route (12 choice-points), a high percentage of participants’ route-choices (almost 20%) were different from that posted on the exit sign, and more noticeable disagreement was found for the first three decisions. This finding is of paramount importance because following the direction opposite of the path to the emergency egress could foreseeably make people walk greater distances and spend more time than necessary to escape from a hazardous situation and could potentially increase the likelihood of injury or death. One limitation of the present research is the use of behaviour as the main dependent measure. Other complementary measures such as eye-gaze and/or self-reports would provide at least some insights into what participants looked at and thought about while navigating, as well as indications about the existence of a possible ‘tunnelling’ effect. The degradation in performance, which occurs as a consequence of the selectivity of attention or ‘tunnelling’ to different environmental or internal cues, narrowing the spotlight of attention (Easterbrook 1959) could explain the percentages of choices favouring environmental variables direction considering situational contexts (i.e. everyday and emergency), with higher percentages of choice in the emergency sign condition than in the everyday signs condition. Thus, although it would be very informative to understand the reasons underlying the participants’ navigational behaviour, it should be noted that this study was not designed to investigate the decision-making but instead the decision-taking processes. That is, in the current experiment, we know that signs are in the field of view of participants, but we do not know if participants visually searched or gazed at them and, if they did so, yet continued to follow the environmental variables. According to the communications –human information (C-HIP) model (Wogalter 2006), once the message (i.e. sign) reaches the receiver, to be successful (i.e. to promote an intended behaviour) it must successfully pass five processing stages until behavioural compliance occurs. In this case, results suggest that in the emergency situation, there is a low compliance with the signs, but we do not know in which stage of the process a blockage occurred, if any existed, for example if there is an attentional failure or a comprehension one. Further research employing other methods and tools, such as eye-trackers, or a different methodological approach, can help verify and determine which influence, on the navigational behaviours, is attributable to the signs and environmental variables. Nevertheless, with the necessary caution, the current results have implications for both the design of buildings and the design of signage systems. Critical questions for practitioners are whether static wall-mounted signs are able to successfully direct people when in competition with environmental variables, as well as if environmental variables can be used to influence route-choice within complex buildings. Given the main findings from this work, static wall-mounted signs with destination names and arrows pointing the way can be sufficient for everyday situations, and for buildings with low simultaneous flow of people, like the one studied in this research. However, the ISO-type exit signs (static and usually found ‘running man’ sign) for the studied conditions presented themselves as not so efficient in directing people to a safe place. Our findings regarding the environmental variables suggest that such variables can be used to improve wayfinding performance.

Funding This research was supported by the Portuguese Science and Technology Foundation (FCT) [grant numbers PTDC/PSI-PCO/100148/2008, and SFRH/BD/38927/2007].

524

E. Vilar et al.

Downloaded by [UOV University of Oviedo] at 06:53 22 October 2014

References Allbeck, J. M., and N. I. Badler. 2002. “Embodied Autonomous Agent.” In Handbook of Virtual Environments: Design, Implementation, and Applications, edited by K. M. Stanney, 313– 332. Mahwah, NJ: Lawrence Erlbaum Associates. Bourne, L. E., and R. A. Yaroush. 2003. Stress and Cognition: A Cognitive Psychological Perspective, NASA Technical Report, NASA, p. 123. Conroy-Dalton, R. 2001. “Spatial Navigation in Immersive Virtual Environments.” Doctoral diss., University of London. http://dis covery.ucl.ac.uk/1111/ Conroy-Dalton, R. 2003. “The Secret Is to Follow Your Nose: Route Path Selection and Angularity.” Environment and Behavior 35 (1): 107– 131. doi:10.1177/0013916502238867. Duarte, E., F. Rebelo, J. Teles, and M. S. Wogalter. 2013. “Behavioral Compliance for Dynamic versus Static Signs in an Immersive Virtual Environment.” Applied Ergonomics. doi:http://dx.doi.org/10.1016/j.apergo.2013.10.004. Easterbrook, J. A. 1959. “The Effect of Emotion on Cue Utilization and the Organization of Behavior.” Psychological Review 66 (3): 183– 201. doi:10.1037/h0047707. Gamberini, L., P. Cottone, A. Spagnolli, D. Varotto, and G. Mantovani. 2003. “Responding to a Fire Emergency in a Virtual Environment: Different Patterns of Action for Different Situations.” Ergonomics 46 (8): 842 – 858. doi:10.1080/ 0014013031000111266. Hochmair, H., and A. U. Frank. 2000. “Influence of Estimation Errors on Wayfinding-Decisions in Unknown Street Networks – Analyzing the Least-Angle Strategy.” Spatial Cognition and Computation 2 (4): 283– 313. doi:10.1023/A:1015566423907. Hochmair, H., and V. Karlsson. 2005. “Investigation of Preference between the Least-Angle Strategy and the Initial Segment Strategy for Route Selection in Unknown Environments.” In Spatial Cognition IV. Reasoning, Action, Interaction, edited by C. Freksa, M. Knauff, B. Krieg-Bru¨ckner, B. Nebel, and T. Barkowsky. Vol. 3343, 79 –97. Berlin: Springer. Ho¨lscher, C., T. Meilinger, G. Vrachliotis, M. Bro¨samle, and M. Knauff. 2006. “Up the Down Staircase: Wayfinding Strategies in Multilevel Buildings.” Journal of Environmental Psychology 26 (4): 284– 299. doi:http://dx.doi.org/10.1016/j.jenvp.2006.09.002. Ishihara, S. 1988. Tests for Colour-Blindness: 38 Plates Edition. Tokyo: Kanehara. ISO (International Organization for Standardization). 2002. Graphical Symbols - Safety Colors and Safety Signs. Part 1: Design Principles for Safety Signs in Workplaces and Public Areas. ISO 3864-1. Geneva: International Organization for Standardization. Kennedy, R. S., L. J. Hettinger, and M. G. Lilienthal. 1990. “Simulator Sickness.” In Motion and Space Sickness, edited by G. H. Crampton, 317– 342. Boca Raton, FL: CRC Press. Keshavarz, B., and H. Hecht. 2011. “Validating an Efficient Method to Quantify Motion Sickness.” Human Factors: The Journal of the Human Factors and Ergonomics Society 53 (4): 415– 426. doi:10.1177/0018720811403736. Kobes, M., N. Oberije´, and M. Duyvis. 2010. “Case Studies on Evacuation Behaviour in a Hotel Building in BART and Real Life.” In Pedestrian and Evacuation Dynamics – PED 2008, edited by W. W. F. Klingsch, C. Rogsch, A. Schadschneider, and M. Schreckenberg, 183– 201. Berlin: Springer. Loftus, E. F., G. R. Loftus, and J. Messo. 1987. “Some Facts about ‘Weapon Focus’.” Law and Human Behavior 11 (1): 55– 62. doi:10.1007/bf01044839. Mantovani, G., L. Gamberini, M. Martinelli, and D. Varotto. 2001. “Exploring the Suitability of Virtual Environments for Safety Training: Signals, Norms and Ambiguity in a Simulated Emergency Escape.” Cognition, Technology & Work 3 (1): 33 – 41. doi:10. 1007/pl00011519. O’Neill, M. J. 1991. “Effects of Signage and Floor Plan Configuration on Wayfinding Accuracy.” Environment and Behavior 23 (5): 553– 574. doi:10.1177/0013916591235002. Purser, D. A., and M. Bensilum. 2001. “Quantification of Behaviour for Engineering Design Standards and Escape Time Calculations.” Safety Science 38 (2): 157– 182. doi:10.1016/s0925-7535(00)00066-7. Robinson, E. S. 1933. “The Psychology of Public Educations.” American Journal of Public Health 23 (2): 123– 128. doi:10.2105/AJPH. 23.2.123. Scharine, A. A., and M. K. McBeath. 2002. “Right-Handers and Americans Favor Turning to the Right.” Human Factors: The Journal of the Human Factors and Ergonomics Society 44 (2): 248– 256. doi:10.1518/0018720024497916. Shih, N.-J., C.-Y. Lin, and C.-H. Yang. 2000. “A Virtual-Reality-Based Feasibility Study of Evacuation Time Compared to the Traditional Calculation Method.” Fire Safety Journal 34 (4): 377– 391. doi:10.1016/s0379-7112(00)00009-6. Tang, C.-H., W.-T. Wu, and C.-Y. Lin. 2009. “Using Virtual Reality to Determine How Emergency Signs Facilitate Way-finding.” Applied Ergonomics 40 (4): 722–730. doi:10.1016/j.apergo.2008.06.009. Taylor, L. H., and E. W. Socov. 1974. “The Movement of People toward Lights.” Journal of Illuminating Engineering Society 3: 237– 241. Vilar, E., F. Rebelo, and P. Noriega. 2012. “Indoor Human Wayfinding Performance Using Vertical and Horizontal Signage in Virtual Reality.” Human Factors and Ergonomics in Manufacturing & Service Industries. doi:10.1002/hfm.20503. Vilar, E., F. Rebelo, P. Noriega, J. Teles, and C. Mayhorn. 2013. “The Influence of Environmental Features on Route Selection in an Emergency Situation.” Applied Ergonomics 44 (4): 618– 627. doi:http://dx.doi.org/10.1016/j.apergo.2012.12.002. Vilar, E., L. Teixeira, F. Rebelo, P. Noriega, and J. Teles. 2012. “Using Environmental Affordances to Direct People Natural Movement Indoors.” Work: A Journal of Prevention, Assessment and Rehabilitation 41: 1149– 1156. doi:10.3233/WOR-2012-0295-1149. Wickens, C. D. 1992. Engineering Psychology and Human Performance. 2nd ed. New York: Harper Collins. Wogalter, M. S. 2006. “Communication-Human Information Processing (C-HIP) Model.” In Handbook of Warnings, edited by M. S. Wogalter, 51 – 62. Mahwah, NJ: Lawrence Erlbaum Associates. Xie, H., L. Filippidis, E. R. Galea, D. Blackshields, and P. J. Lawrence. 2011. “Experimental Analysis of the Effectiveness of Emergency Signage and Its Implementation in Evacuation Simulation.” Fire and Materials 36 (5– 6): 367– 382. doi:10.1002/fam.1095.