The Effects of Direction Type on Wayfinding Strategies, Anxiety Levels, and Confidence of Men and Women
Charlotte Davis and Brigida Palatino
The topic for the current study originated from an email sent by a male campus safety officer describing various parking restrictions on the Connecticut College campus. The content of this email was dense with cardinal directions – north, south, east, and west – and we were confused about where these locations were on campus due to how the locations were described. Our interpretation of this email and our curiosity about how other students might have interpreted the email sparked our interest for the present study. We decided to study wayfinding strategies, anxiety levels, and confidence levels between men and women when giving and receiving directions around campus. The current study elaborates on research in the field of spatial cognition and gender differences related to wayfinding ability, anxiety, and confidence. Previous research has indicated gender differences in wayfinding strategies specifically that men tend to attend to cardinal and distance attributes, whereas women tend to attend to landmarks when navigating or reading maps.
Choi and Silverman (2003) investigated gender differences in the route-learning strategies of children and adolescents using a direction-giving paradigm based on a novel map. In that study, 315 children were recruited from two schools in Canada. Each student was given a folder of pre-assembled cognitive tasks, and they completed the tasks in their classroom. One of the tasks required the participants to learn the shortest route from a point of origin to the destination, another task tested object and location memory, and they also completed a water level task to measure spatial perception. The water level task required participants to indicate how the water line in each of the 12 bottles would appear, if each bottle were half-filled with water and various angles. Finally, the participants completed a spatial relations task that required them to mentally rotate 2-D objects. The results of the study were consistent with the theory that specialized spatial processes that differ between women and men can be attributed to evolutionary pressures related to the division of labor.
In another study, spatial ability and performance on spatial tasks were examined among 160 college students who were given a mental rotations test and a cognitive mapping task (O’Laughlin & Brubaker, 1998). One group viewed photos of a furnished house, and the other group viewed an unfurnished house. After viewing the house, participants were asked to sketch a floor plan of the house including landmarks present in the space. Men performed better than women on the mental rotation portion, but no gender differences were found in the cognitive mapping task. Participants in the landmark condition (furnished house) drew more accurate floor plans than those in the minimal landmark condition (unfurnished house), and women reported less confidence in the accuracy of their floor plans than did men. This research led us to hypothesize that men would report higher levels of confidence than women in wayfinding tasks. Wayfinding strategies are not only demonstrated during written tasks, but new technology has enabled researchers to test this skill using computer simulation programs.
Tlauka, Brolese, Pomeroy and Hobbs (2005) examined gender differences in navigation through computer-simulated spaces. The sample consisted of 32 undergraduate students (16 men and 16 women) who explored two large scale virtual shopping centers with either a hand-held paper map or a digital map displayed on a computer screen. The spatial knowledge of the participants was tested through a wayfinding task, directional and distance estimates, and a map placement task. Results indicated that women required more time than men did to travel from the start location to the finish location when following a route through the simulated shopping center. While following the route, women also made more incorrect navigational decisions and required more time to make directional estimates. Also, the wayfinding performance of men was more accurate than women when they were asked to navigate back to the start location, and placed target objects more accurately on a map of the simulated environment than women. These findings suggest that similar to real-world learning, virtual exploration demonstrates significant gender differences in spatial performance and is a valid tool for evaluating wayfinding strategies. Another study examined the use of technology in navigation.
Technology may aid or hinder the user’s navigation of a given area. Ishikawa, Fujiwara, Imai, and Okabe (2008) examined the effectiveness of a Global Positioning System (GPS)-based mobile navigation system in comparison to paper maps and direct experience of routes. Their sample consisted of 66 college students (11 men and 55 women). At the beginning of the study, participants were given questions regarding their experience using GPS-based navigation system, and they completed the Sense-of-Direction questionnaire. The participants were required to walk six routes finding their way to the goal destinations with a map, GPS, or direct-experience after a guided tour of the route. Results showed that GPS users traveled longer distances and made more stops during the walk than did map users and direct-experience participants. Also, GPS users traveled more slowly, made larger direction errors, drew sketch maps with poorer topological accuracy, and rated wayfinding tasks as more difficult than direct-experience participants did. These findings suggest that technology may actually hinder wayfinding abilities, and the study influenced us to use 2-D maps as the primary navigational tool in the present study.
Wayfinding techniques in relation to gender differences have been demonstrated using computer simulation. Devlin and Bernstein (1995) examined the effectiveness of different kinds of wayfinding information utilized by men and women. The study consisted of 277 visitors (146 men and 151 women) to the Connecticut College campus. The visitors were randomly assigned to one of seven different cue information conditions when viewing a computer simulation of a campus tour. The conditions included text, photos, maps, or combinations of these cues. Participants took a computer test using the touch-screen computer monitor. The researchers found that men were significantly more confident than women that they could find their way. The results also indicated that men preferred visual spatial cues more than women.
In a study of the effects of familiarity on wayfinding, Murakoshi and Kawai (2000) investigated wayfinding behavior in an unfamiliar environment of a complex building. The participants of the study were 24 university freshmen. The main task of the experiment was to navigate back to the start using the shortest possible route after an 8-minute walk within the complex building. The participants also completed a photo memory task, a route memory task, a pointing task, and sketched map of the route. The results indicated that wayfinding performance correlated with performance in the sketch map task, the pointing task, and with route scene memory. However, the researchers found that some participants who either drew incomplete sketch maps or had an inaccurate homing vector could also find their way with a minimum number of errors. Their verbal explanations of their choices revealed that route scenes, schema-like knowledge, environmental constraints, and information-seeking heuristics enabled them to find their way in the artificial environment. These findings suggest that directional sense and memory of landmarks aid in wayfinding.
The ability to maintain orientation in an environment varies between individuals. Lawton (1996) investigated self-reported indoor wayfinding strategies in comparison with previously identified outdoor wayfinding strategies, and she also examined gender differences in the relationships between indoor strategies and levels of spatial anxiety. In this study, participants (104 men and 174 women) were given a landmark learning tour of a floor of a campus building. During the tour participants were asked to learn landmarks located along each corridor, and after the tour participants were given a pointing task in a room without visual access to any of the landmarks of the tour. Using a compass, participants had to indicate the directions of the landmarks on the tour. Results indicated that indoor reliance on directional cues correlated with the outdoor orientation strategy and that indoor reliance on route information correlated with the outdoor route strategy. Consistent with previous research, men reported higher use of the indoor orientation strategy and women reported higher use of the indoor route strategy.
Lawton and Kallai (2002) examined gender and cultural differences in wayfinding strategies and levels of anxiety when performing navigation tasks. The participants in the study were 299 American students (185 women and 114 men) who attended a commuter university and 214 Hungarian students (110 women and 104 men) who attended a commuter university in Hungary. Participants were given a questionnaire to assess wayfinding strategies, wayfinding anxiety, and trait anxiety. Results of the study showed that men reported greater preference for a strategy of orienting using global reference points (i.e., North, South, East, West), whereas women reported greater preference for strategy based on route information (i.e., use of landmarks). As for anxiety, women reported higher levels of wayfinding anxiety than did men. The scales used in Lawton’s study were used in the present study to assess wayfinding strategy and anxiety across gender.
Wayfinding research has demonstrated that familiarity with the environment relates to how efficiently and accurately people can navigate the environment. Devlin (2001) suggested that individuals with extensive familiarity with the environment they are navigating have different wayfinding approaches than those who are unfamiliar with the environment. Her research suggests that maps should be created for the less experienced user and be designed to appeal to larger populations in order to accommodate for individual differences in wayfinding ability (Devlin, 2001). Prestopnik and Roskos-Ewoldsen (2000) also proposed that increased familiarity with an environment provides individuals with more knowledge about an area’s properties and characteristics than they would have in a less familiar environment. Familiarity may increase performance on wayfinding tasks; as familiarity increases, the degree of complexity of the environment becomes less important for accurate navigating. Familiarity was assessed by measured questions concerning how long they had lived in the environment and how familiar they were with the environment. The results showed that both familiarity and sense of direction are directly related to accurate wayfinding ability. Participants who self-reported an accurate sense of direction also navigated the environment more frequently, thus, they also reported being more familiar with the area. The two familiarity measures used were correlated in that people who lived in the area longer also claimed to be more familiar with the area. That relationship between wayfinding ability and familiarity was also tested in the present study to see if wayfinding skills differ between first- and second-year students and third- and fourth-year students at Connecticut College.
Hund and Padgitt (2010) examined how the types of descriptive features contained in wayfinding descriptions influenced sense of direction and strategies when navigating an indoor environment; with specific regard to how men and women approach the task. This sample consisted of three different experimental groups. The purpose of Experiment 1 was to specify how sense of direction, wayfinding strategies, anxiety, and gender relate to the descriptive features provided when giving wayfinding directions. The participants were university students (36 men and 39 women) who were asked to provide directions from various start locations to destinations. As predicted, the results of Experiment 1 showed that people with a self-reported good sense of direction provided correct directions more often than people with a self-reported poor sense of direction. Also, people who said they had a good self-reported sense of direction provided more often detail in their wayfinding descriptions. In addition, women indicated higher route strategy preferences and more spatial anxiety than men did, and men indicated higher orientation strategy preferences than women did. In Experiment 2, participants rated the effectiveness of directions. The sample consisted of 46 men and 33 women. The results showed that the wayfinding directions that received high effectiveness ratings contained more left-right and landmark descriptors than did descriptions that received lower ratings. Lastly, in Experiment 3, participants navigated an indoor environment with directions that had been rated with different levels of effectiveness. The participants were 53 male and 49 female college students, and an unexpected finding was that participants navigated faster when they followed the worst-rated directions than when they followed the best-rated directions.
In the present study, we aimed to test the participants’ ability to give directions and utilize prescribed directions by indicating the route on a 2-D map. Devlin (2001) discussed how maps elicit problems in studies of wayfinding strategy. For example, participants had trouble reading and using maps because of their inability to combine the environment and the map it represents into an effective guidance tool. Thus, maps may be a source of trouble when navigating an unfamiliar, or even a familiar, space. Wayfinding can also provoke anxiety as a result of confusion and disorientation. Devlin (2001) mentioned a study in which students who were instructed to use a map to navigate an area were less confident in their wayfinding skills than those who were not given a map, which demonstrates that maps can inhibit successful wayfinding and be might more problematic than useful.
To summarize, previous research has indicated gender differences in wayfinding strategies, anxiety, and confidence in indoor, outdoor, computer-simulated, familiar, and unfamiliar environments. Studies have also compared GPS navigational aids to maps and direct experience. The purpose of the present study was to investigate gender differences in two wayfinding tasks: indicating a given route on a 2-D map and providing directions from the given start and end location. It was hypothesized that men would prefer cardinal directions, whereas women would prefer landmark directions. It was also hypothesized that women would have higher anxiety levels than men and that men would express higher levels of confidence in wayfinding than women.
Method
Participants
The sample consisted of 50 participants, all of whom were students at Connecticut College. There were 26 women (Mage= 20.7) and 24 men (Mage= 20.7). The cardinal group consisted of 15 women and 10 men. The landmark group consisted of 11 women and 13 men. Participants were recruited through sign-up sheets from the psychology department subject pool and participants were also recruited at the library during its peak hours. Participants were randomly assigned to either the cardinal or landmark group. All participants received either credit hours or candy for their participation.
Materials
Lawton’s Wayfinding Strategy Scale. Lawton’s Wayfinding Strategy Scale (Lawton & Kallai, 2002) was used to assess orientation strategies. The scale consists of 14 strategies for finding one’s way to a location in city or town and 10 strategies for finding one’s way in a building or large complex. The scale includes a 5-point Likert type scale which ranges from Not at all true to Very true, on which participants to rate the degree to which they think that each strategy applies to themselves. A sample item of an orientation strategy is “I kept track of the direction (north, south, east, or west) in which I was going” and a sample item of a route strategy is “I asked for direction telling me how many streets to pass before making each turn.” Scores indicate participants’ wayfinding strategy preferences – orientation or route strategy. The Cronbach’s alpha rating for Lawton’s original test is .79.
Lawton’s Spatial Anxiety Scale. Lawton’s Spatial Anxiety Scale (Lawton & Kallai, 2002) was used to assess anxiety levels in wayfinding tasks. The scale consists of eight wayfinding tasks that might promote anxiety in a participant, such as finding one’s way to an appointment in an unfamiliar area of a town or city. The scale includes a 5-point Likert type scale that ranges from Not at all anxious to Very anxious. Sample items include “Finding my way in an unfamiliar shopping mall, medical center, or large building complex” and “Trying a new route that I think will be a shortcut, without a map.” Higher scores indicate higher levels of anxiety. The Chronbach’s alpha rating for Lawton’s original test .87.
Demographic questionnaire. Participants received a demographics questionnaire that included age, gender, class year, questions regarding their confidence on the wayfinding tasks, and other questions that related to spatial abilities. Participants were also asked to indicate which gender they believed wrote the directions for the first task.
Procedure
The experiment was conducted in Bill Hall and in the Shain Library at Connecticut College. Sign-up sheets were posted in Bill Hall for participants from the Psychology Department subject pool to earn course credit. Participants were also recruited in the library by using candy as an incentive. All participants followed the same procedure and were offered either 30 minutes of credit or candy for their participation.
Upon entering the room, each participant received an informed consent that outlined the purpose of the research. After they had agreed to participate and signed consent, each participant was given two copies of a Connecticut College 2-D campus map labeled “Map 1” and “Map 2." Participants were also given two sets of directions. One group received cardinal-based instructions and the other group received landmark-based instructions. For the first task, the participants were asked to mentally navigate from one place on the campus to another using the directions given and were asked to indicate the route on the provided map. For both cardinal-based and landmark-based instructions, participants received two sets of directions and were asked to indicate the route on the provided maps. Participants were asked to indicate the route of direction set 1 on the map labeled “Map 1” and asked to indicate the route of direction set 2 on the map labeled on “Map 2”. Next, participants were provided with a start destination and an end destination and were asked to write their own directions for this route. For example, participants were asked to provide written directions to a visitor from the basement of the library to the Admissions building. The route was scored using the wayfinding criteria created by Ward et al. (1996). The content of the written directions was coded into six categories: use of cardinality (mention of north, south, east, or west), relational terms (left, right, behind), mention of landmarks (signs, buildings), mileage (mention of distance between two points), omission errors (failure to mention something essential), and commission errors (inclusion of wrong information). After participants had completed this task, they were given the questionnaires about wayfinding strategies and levels of anxiety. Finally, participants received the demographics questionnaire. Participants were also asked to indicate which gender they believed wrote the directions for the first task. Upon completion, all participants were given a debriefing form. The entire study took approximately 30 minutes.
Results
To test the hypothesis that men and women would differ on wayfinding strategy, anxiety, and confidence, a factorial multivariate analysis of variance was employed. There was a significant difference for gender, Wilks’s lambda = .788, F(4,42) = 2.82, p = .037; η2 = .212, and a significant multivariate effect for condition, Wilks’s lambda = .756, F(4,42) = 3.40, p=.017; η2 = .244. The interaction was significant, Wilks’s lambda = .630, F(4,42) = 6.17, p=.001; η2 = .370. A simple effects test indicated that participants reported more anxiety in the cardinal condition than in the landmark condition (p=.0023). Means and standard deviations for orientation, route, anxiety, and confidence scores for the landmark group can be seen in Table 1 and scores for the cardinal group can be seen in Table 2.
To test the hypothesis that women would have higher levels of anxiety than men, a univariate analysis of variance was conducted. The one-way ANOVA revealed that women were reported more anxiety than men, F(1,49) = 7.84, p=.007. Anxiety levels reported by women were significantly higher (M=25.04, SD=7.11) than for men (M=19.5, SD=6.85). To test the hypothesis that men would have higher levels of confidence than women, a univariate analysis of variance was conducted. The one-way ANOVA indicated that there was no significant difference in confidence level between men and women, F(1,49) = .458, p=.502. Refer to Table 1 and 2 for men’s and women’s confidence levels in both conditions.
A MANOVA was conducted to explore anxiety and confidence levels between first- and second-year students to third- and fourth-year students. Results indicated a significant difference in confidence level between first- and second-year students and third- and fourth-year students, Wilks’s lambda = .803, F(4,44) = 2.70, p = .043; η2 = .197. Univariate analysis of variance revealed that first- and second-year students were less confident than third- and fourth-year students in giving and receiving directions, F(1,44) = 9.03, p=.004. See Table 3 for means and standard deviations of first- and second-year students and third- and fourth-year students.
An independent samples t-test was employed to see if participants primed with landmark-based directions would use more landmarks when writing their own directions and whether participants primed with cardinal-based directions would use more cardinal terms when writing their own directions. Participants in the landmark condition, t(48) = -3.54, p=.001, scored significantly differently from those in the cardinal condition, t(48) = 2.27, p=.028. Means and standard deviations for the number of landmark and cardinal references used in written directions can be seen in Table 4.
A chi-square test was conducted to examine the relationship between condition and the assumed gender of the person who wrote the directions. The chi-square indicated that participants in the cardinal condition tended to assume that the directions were written by a man and participants in the landmark condition would indicate that the directions were written by a woman, X2(1, N=46) = 9.23, p=.002. In the cardinal condition, 45% of participants thought the directions were written by woman, and 54% thought the directions were written by a man. In the landmark condition, 87.5% thought the directions were written by a woman, and 12.5% thought the directions were written by a man.
Discussion
The hypothesis that men would prefer cardinal directions compared to landmark directions and women would prefer landmark directions compared to cardinal directions was supported. The factorial MANOVA indicated significant differences between men and women and their preferred wayfinding strategy. This supports results of previous research. When navigating with a 2-D map, our results indicate gendered preferences in written directions; men prefer orientational cues, whereas women prefer route cues.
According to our simple effects test, participants reported higher levels of anxiety in the cardinal condition than in the landmark condition. This finding indicates that, regardless of gender, participants in the cardinal condition were more anxious overall than those in the landmark condition, which suggests that cardinal directions may elicit more anxiety during wayfinding tasks.
The hypothesis that women would report higher anxiety levels than men in wayfinding tasks was supported by the one-way ANOVA. Women reported significantly higher levels of anxiety on wayfinding tasks than did men. Similar to previous research, these results suggest that women report more anxious behavior than men when performing the same navigational task.
The hypothesis that men would report higher levels of confidence in wayfinding tasks than women was not supported by a one-way ANOVA. There were no significant gender differences in self-reported confidence levels. Despite our findings, previous research has indicated that men report more confidence in their wayfinding abilities than women do, possibly due to pressures of masculinity to appear confident. Although that hypothesis was not supported, results indicated that third and fourth-year students reported higher confidence than first- and second-year students in giving and receiving directions. This result indicates that familiarity with the campus could have contributed to higher confidence levels. Therefore, freshmen and sophomores may be less familiar and less confident with the campus than juniors and seniors.
We also found that participants primed with landmark-based directions used more landmarks when writing their own directions, and participants primed with cardinal-based directions used more cardinal terms when writing their own directions. The independent samples t-test demonstrated that the type of directions (i.e. landmark-based versus cardinal-based) influenced how participants wrote their own directions. Participants may change the way they write directions based on what they think the researcher expects; therefore, priming influences how the participant provides directions.
A chi-square analysis indicated that participants in the cardinal condition were more likely to guess that the directions were written by a man and participants in the landmark condition were more likely to guess that the directions were written by a woman. These results add to the wayfinding research by demonstrating that direction content influences perception of gender based on route and orientation terminology. Directions primarily loaded with cardinal directions such as north, south, east, or west were assumed to have been written by a man and directions using landmarks such as a dorm, a building, or a statue were assumed to have been written by a woman.
The current results aid in the understanding of how men and women differ in navigation and wayfinding strategies using a 2-D map. Lawton and Kallai (2002) found that women utilized landmark cues more often than men and also that women reported being more anxious during wayfinding than men, and our results replicate the findings. Women reported higher levels of anxiety on wayfinding tasks compared to men and also preferred to use landmark references when giving and receiving directions.
This study adds to the research on wayfinding and spatial cognition in that familiarity of an environment may influence confidence levels as also suggested by Devlin (2001) and Prestopnik and Roskos-Ewoldse (2000). Our results indicate that third- and fourth-year students reported significantly higher levels of confidence on the wayfinding tasks than did first- and second-year students. Their confidence level may be influenced by the increased familiarity of the campus compared to first- and second-year students; given that juniors and seniors have been at the college longer than their freshmen and sophomore counterparts.
One of the limitations of the present study is a small and homogeneous sample size, which cannot be generalized to the greater population. It is also possible that instructions and directions of the different tasks could have been misinterpreted by the participants. The instructions might have been unclear in terms of if the participants were to answer the wayfinding strategy scale and spatial anxiety scale in reference to the wayfinding activity, the direction-giving task, or to the participants’ daily wayfinding strategies. Assumptions about the researchers’ gender may have persuaded the participant about which gender wrote the directions rather than using the content of the directions. In the task where the participants had to write their own directions from the given start and end location, the participants may have been influenced by the condition they were exposed to – either cardinal-based or landmark-based directions. Whether participants were in the cardinal or landmark group, this first task may have primed their response for the next task. This limitation indicates that the directions may not be representative of the directions they would provide without the prior prompt. Lastly, some of the maps provided to the cardinal group did not contain a compass because we expected participants to recognize the cardinal orientation of campus. Therefore, participants had to be familiar enough with the campus in order to determine the cardinal locations, provided that the top of the map was not north.
One possible threat to the internal validity of the present study could have been the location disparity given that the experiment was conducted in an academic building as well as in a library setting. The environment in which the participants took the survey might have influenced their performance given where they were located on campus. Confounding variables included individual differences regarding spatial cognition and wayfinding expertise when reading and navigating a map. As Devlin (2001) discussed, maps may hinder wayfinding strategies especially when people are unable to connect the environment to what the map represents. As seniors and juniors may be more familiar with the campus environment than freshmen and sophomores, they may have been more likely to associate the map with the campus properly and its features than first- and second-year students who have not had as much exposure. Implications of this study can be applied to the field of cartography in that people respond differently to directional cues and use different wayfinding strategies. Signage should be created with both genders in mind to ease navigation.
Future researchers could follow a similar procedure by adding the Santa Barbara Sense-of-Direction Scale as used in the Ishikawa, Fujiwara, Imai, and Okabe (2008) study. That scale could provide insight as to participants’ self-perception of their wayfinding ability and would be useful to compare their self-reported results to their actual performance on the tasks. Further research should assess perceived ability compared to realistic ability, and analyzing possible discrepancy would be valuable. Also future researchers could ask participants to follow written directions and navigate the route on foot around a given area. Furthermore, another potential topic of research could examine wayfinding abilities in natural environments, such as a wooded environment. Another study could compare natural environments to built environments to see if men and women differ on wayfinding strategies, confidence, and anxiety in different environments.
References
Choi and Silverman (2003) investigated gender differences in the route-learning strategies of children and adolescents using a direction-giving paradigm based on a novel map. In that study, 315 children were recruited from two schools in Canada. Each student was given a folder of pre-assembled cognitive tasks, and they completed the tasks in their classroom. One of the tasks required the participants to learn the shortest route from a point of origin to the destination, another task tested object and location memory, and they also completed a water level task to measure spatial perception. The water level task required participants to indicate how the water line in each of the 12 bottles would appear, if each bottle were half-filled with water and various angles. Finally, the participants completed a spatial relations task that required them to mentally rotate 2-D objects. The results of the study were consistent with the theory that specialized spatial processes that differ between women and men can be attributed to evolutionary pressures related to the division of labor.
In another study, spatial ability and performance on spatial tasks were examined among 160 college students who were given a mental rotations test and a cognitive mapping task (O’Laughlin & Brubaker, 1998). One group viewed photos of a furnished house, and the other group viewed an unfurnished house. After viewing the house, participants were asked to sketch a floor plan of the house including landmarks present in the space. Men performed better than women on the mental rotation portion, but no gender differences were found in the cognitive mapping task. Participants in the landmark condition (furnished house) drew more accurate floor plans than those in the minimal landmark condition (unfurnished house), and women reported less confidence in the accuracy of their floor plans than did men. This research led us to hypothesize that men would report higher levels of confidence than women in wayfinding tasks. Wayfinding strategies are not only demonstrated during written tasks, but new technology has enabled researchers to test this skill using computer simulation programs.
Tlauka, Brolese, Pomeroy and Hobbs (2005) examined gender differences in navigation through computer-simulated spaces. The sample consisted of 32 undergraduate students (16 men and 16 women) who explored two large scale virtual shopping centers with either a hand-held paper map or a digital map displayed on a computer screen. The spatial knowledge of the participants was tested through a wayfinding task, directional and distance estimates, and a map placement task. Results indicated that women required more time than men did to travel from the start location to the finish location when following a route through the simulated shopping center. While following the route, women also made more incorrect navigational decisions and required more time to make directional estimates. Also, the wayfinding performance of men was more accurate than women when they were asked to navigate back to the start location, and placed target objects more accurately on a map of the simulated environment than women. These findings suggest that similar to real-world learning, virtual exploration demonstrates significant gender differences in spatial performance and is a valid tool for evaluating wayfinding strategies. Another study examined the use of technology in navigation.
Technology may aid or hinder the user’s navigation of a given area. Ishikawa, Fujiwara, Imai, and Okabe (2008) examined the effectiveness of a Global Positioning System (GPS)-based mobile navigation system in comparison to paper maps and direct experience of routes. Their sample consisted of 66 college students (11 men and 55 women). At the beginning of the study, participants were given questions regarding their experience using GPS-based navigation system, and they completed the Sense-of-Direction questionnaire. The participants were required to walk six routes finding their way to the goal destinations with a map, GPS, or direct-experience after a guided tour of the route. Results showed that GPS users traveled longer distances and made more stops during the walk than did map users and direct-experience participants. Also, GPS users traveled more slowly, made larger direction errors, drew sketch maps with poorer topological accuracy, and rated wayfinding tasks as more difficult than direct-experience participants did. These findings suggest that technology may actually hinder wayfinding abilities, and the study influenced us to use 2-D maps as the primary navigational tool in the present study.
Wayfinding techniques in relation to gender differences have been demonstrated using computer simulation. Devlin and Bernstein (1995) examined the effectiveness of different kinds of wayfinding information utilized by men and women. The study consisted of 277 visitors (146 men and 151 women) to the Connecticut College campus. The visitors were randomly assigned to one of seven different cue information conditions when viewing a computer simulation of a campus tour. The conditions included text, photos, maps, or combinations of these cues. Participants took a computer test using the touch-screen computer monitor. The researchers found that men were significantly more confident than women that they could find their way. The results also indicated that men preferred visual spatial cues more than women.
In a study of the effects of familiarity on wayfinding, Murakoshi and Kawai (2000) investigated wayfinding behavior in an unfamiliar environment of a complex building. The participants of the study were 24 university freshmen. The main task of the experiment was to navigate back to the start using the shortest possible route after an 8-minute walk within the complex building. The participants also completed a photo memory task, a route memory task, a pointing task, and sketched map of the route. The results indicated that wayfinding performance correlated with performance in the sketch map task, the pointing task, and with route scene memory. However, the researchers found that some participants who either drew incomplete sketch maps or had an inaccurate homing vector could also find their way with a minimum number of errors. Their verbal explanations of their choices revealed that route scenes, schema-like knowledge, environmental constraints, and information-seeking heuristics enabled them to find their way in the artificial environment. These findings suggest that directional sense and memory of landmarks aid in wayfinding.
The ability to maintain orientation in an environment varies between individuals. Lawton (1996) investigated self-reported indoor wayfinding strategies in comparison with previously identified outdoor wayfinding strategies, and she also examined gender differences in the relationships between indoor strategies and levels of spatial anxiety. In this study, participants (104 men and 174 women) were given a landmark learning tour of a floor of a campus building. During the tour participants were asked to learn landmarks located along each corridor, and after the tour participants were given a pointing task in a room without visual access to any of the landmarks of the tour. Using a compass, participants had to indicate the directions of the landmarks on the tour. Results indicated that indoor reliance on directional cues correlated with the outdoor orientation strategy and that indoor reliance on route information correlated with the outdoor route strategy. Consistent with previous research, men reported higher use of the indoor orientation strategy and women reported higher use of the indoor route strategy.
Lawton and Kallai (2002) examined gender and cultural differences in wayfinding strategies and levels of anxiety when performing navigation tasks. The participants in the study were 299 American students (185 women and 114 men) who attended a commuter university and 214 Hungarian students (110 women and 104 men) who attended a commuter university in Hungary. Participants were given a questionnaire to assess wayfinding strategies, wayfinding anxiety, and trait anxiety. Results of the study showed that men reported greater preference for a strategy of orienting using global reference points (i.e., North, South, East, West), whereas women reported greater preference for strategy based on route information (i.e., use of landmarks). As for anxiety, women reported higher levels of wayfinding anxiety than did men. The scales used in Lawton’s study were used in the present study to assess wayfinding strategy and anxiety across gender.
Wayfinding research has demonstrated that familiarity with the environment relates to how efficiently and accurately people can navigate the environment. Devlin (2001) suggested that individuals with extensive familiarity with the environment they are navigating have different wayfinding approaches than those who are unfamiliar with the environment. Her research suggests that maps should be created for the less experienced user and be designed to appeal to larger populations in order to accommodate for individual differences in wayfinding ability (Devlin, 2001). Prestopnik and Roskos-Ewoldsen (2000) also proposed that increased familiarity with an environment provides individuals with more knowledge about an area’s properties and characteristics than they would have in a less familiar environment. Familiarity may increase performance on wayfinding tasks; as familiarity increases, the degree of complexity of the environment becomes less important for accurate navigating. Familiarity was assessed by measured questions concerning how long they had lived in the environment and how familiar they were with the environment. The results showed that both familiarity and sense of direction are directly related to accurate wayfinding ability. Participants who self-reported an accurate sense of direction also navigated the environment more frequently, thus, they also reported being more familiar with the area. The two familiarity measures used were correlated in that people who lived in the area longer also claimed to be more familiar with the area. That relationship between wayfinding ability and familiarity was also tested in the present study to see if wayfinding skills differ between first- and second-year students and third- and fourth-year students at Connecticut College.
Hund and Padgitt (2010) examined how the types of descriptive features contained in wayfinding descriptions influenced sense of direction and strategies when navigating an indoor environment; with specific regard to how men and women approach the task. This sample consisted of three different experimental groups. The purpose of Experiment 1 was to specify how sense of direction, wayfinding strategies, anxiety, and gender relate to the descriptive features provided when giving wayfinding directions. The participants were university students (36 men and 39 women) who were asked to provide directions from various start locations to destinations. As predicted, the results of Experiment 1 showed that people with a self-reported good sense of direction provided correct directions more often than people with a self-reported poor sense of direction. Also, people who said they had a good self-reported sense of direction provided more often detail in their wayfinding descriptions. In addition, women indicated higher route strategy preferences and more spatial anxiety than men did, and men indicated higher orientation strategy preferences than women did. In Experiment 2, participants rated the effectiveness of directions. The sample consisted of 46 men and 33 women. The results showed that the wayfinding directions that received high effectiveness ratings contained more left-right and landmark descriptors than did descriptions that received lower ratings. Lastly, in Experiment 3, participants navigated an indoor environment with directions that had been rated with different levels of effectiveness. The participants were 53 male and 49 female college students, and an unexpected finding was that participants navigated faster when they followed the worst-rated directions than when they followed the best-rated directions.
In the present study, we aimed to test the participants’ ability to give directions and utilize prescribed directions by indicating the route on a 2-D map. Devlin (2001) discussed how maps elicit problems in studies of wayfinding strategy. For example, participants had trouble reading and using maps because of their inability to combine the environment and the map it represents into an effective guidance tool. Thus, maps may be a source of trouble when navigating an unfamiliar, or even a familiar, space. Wayfinding can also provoke anxiety as a result of confusion and disorientation. Devlin (2001) mentioned a study in which students who were instructed to use a map to navigate an area were less confident in their wayfinding skills than those who were not given a map, which demonstrates that maps can inhibit successful wayfinding and be might more problematic than useful.
To summarize, previous research has indicated gender differences in wayfinding strategies, anxiety, and confidence in indoor, outdoor, computer-simulated, familiar, and unfamiliar environments. Studies have also compared GPS navigational aids to maps and direct experience. The purpose of the present study was to investigate gender differences in two wayfinding tasks: indicating a given route on a 2-D map and providing directions from the given start and end location. It was hypothesized that men would prefer cardinal directions, whereas women would prefer landmark directions. It was also hypothesized that women would have higher anxiety levels than men and that men would express higher levels of confidence in wayfinding than women.
Method
Participants
The sample consisted of 50 participants, all of whom were students at Connecticut College. There were 26 women (Mage= 20.7) and 24 men (Mage= 20.7). The cardinal group consisted of 15 women and 10 men. The landmark group consisted of 11 women and 13 men. Participants were recruited through sign-up sheets from the psychology department subject pool and participants were also recruited at the library during its peak hours. Participants were randomly assigned to either the cardinal or landmark group. All participants received either credit hours or candy for their participation.
Materials
Lawton’s Wayfinding Strategy Scale. Lawton’s Wayfinding Strategy Scale (Lawton & Kallai, 2002) was used to assess orientation strategies. The scale consists of 14 strategies for finding one’s way to a location in city or town and 10 strategies for finding one’s way in a building or large complex. The scale includes a 5-point Likert type scale which ranges from Not at all true to Very true, on which participants to rate the degree to which they think that each strategy applies to themselves. A sample item of an orientation strategy is “I kept track of the direction (north, south, east, or west) in which I was going” and a sample item of a route strategy is “I asked for direction telling me how many streets to pass before making each turn.” Scores indicate participants’ wayfinding strategy preferences – orientation or route strategy. The Cronbach’s alpha rating for Lawton’s original test is .79.
Lawton’s Spatial Anxiety Scale. Lawton’s Spatial Anxiety Scale (Lawton & Kallai, 2002) was used to assess anxiety levels in wayfinding tasks. The scale consists of eight wayfinding tasks that might promote anxiety in a participant, such as finding one’s way to an appointment in an unfamiliar area of a town or city. The scale includes a 5-point Likert type scale that ranges from Not at all anxious to Very anxious. Sample items include “Finding my way in an unfamiliar shopping mall, medical center, or large building complex” and “Trying a new route that I think will be a shortcut, without a map.” Higher scores indicate higher levels of anxiety. The Chronbach’s alpha rating for Lawton’s original test .87.
Demographic questionnaire. Participants received a demographics questionnaire that included age, gender, class year, questions regarding their confidence on the wayfinding tasks, and other questions that related to spatial abilities. Participants were also asked to indicate which gender they believed wrote the directions for the first task.
Procedure
The experiment was conducted in Bill Hall and in the Shain Library at Connecticut College. Sign-up sheets were posted in Bill Hall for participants from the Psychology Department subject pool to earn course credit. Participants were also recruited in the library by using candy as an incentive. All participants followed the same procedure and were offered either 30 minutes of credit or candy for their participation.
Upon entering the room, each participant received an informed consent that outlined the purpose of the research. After they had agreed to participate and signed consent, each participant was given two copies of a Connecticut College 2-D campus map labeled “Map 1” and “Map 2." Participants were also given two sets of directions. One group received cardinal-based instructions and the other group received landmark-based instructions. For the first task, the participants were asked to mentally navigate from one place on the campus to another using the directions given and were asked to indicate the route on the provided map. For both cardinal-based and landmark-based instructions, participants received two sets of directions and were asked to indicate the route on the provided maps. Participants were asked to indicate the route of direction set 1 on the map labeled “Map 1” and asked to indicate the route of direction set 2 on the map labeled on “Map 2”. Next, participants were provided with a start destination and an end destination and were asked to write their own directions for this route. For example, participants were asked to provide written directions to a visitor from the basement of the library to the Admissions building. The route was scored using the wayfinding criteria created by Ward et al. (1996). The content of the written directions was coded into six categories: use of cardinality (mention of north, south, east, or west), relational terms (left, right, behind), mention of landmarks (signs, buildings), mileage (mention of distance between two points), omission errors (failure to mention something essential), and commission errors (inclusion of wrong information). After participants had completed this task, they were given the questionnaires about wayfinding strategies and levels of anxiety. Finally, participants received the demographics questionnaire. Participants were also asked to indicate which gender they believed wrote the directions for the first task. Upon completion, all participants were given a debriefing form. The entire study took approximately 30 minutes.
Results
To test the hypothesis that men and women would differ on wayfinding strategy, anxiety, and confidence, a factorial multivariate analysis of variance was employed. There was a significant difference for gender, Wilks’s lambda = .788, F(4,42) = 2.82, p = .037; η2 = .212, and a significant multivariate effect for condition, Wilks’s lambda = .756, F(4,42) = 3.40, p=.017; η2 = .244. The interaction was significant, Wilks’s lambda = .630, F(4,42) = 6.17, p=.001; η2 = .370. A simple effects test indicated that participants reported more anxiety in the cardinal condition than in the landmark condition (p=.0023). Means and standard deviations for orientation, route, anxiety, and confidence scores for the landmark group can be seen in Table 1 and scores for the cardinal group can be seen in Table 2.
To test the hypothesis that women would have higher levels of anxiety than men, a univariate analysis of variance was conducted. The one-way ANOVA revealed that women were reported more anxiety than men, F(1,49) = 7.84, p=.007. Anxiety levels reported by women were significantly higher (M=25.04, SD=7.11) than for men (M=19.5, SD=6.85). To test the hypothesis that men would have higher levels of confidence than women, a univariate analysis of variance was conducted. The one-way ANOVA indicated that there was no significant difference in confidence level between men and women, F(1,49) = .458, p=.502. Refer to Table 1 and 2 for men’s and women’s confidence levels in both conditions.
A MANOVA was conducted to explore anxiety and confidence levels between first- and second-year students to third- and fourth-year students. Results indicated a significant difference in confidence level between first- and second-year students and third- and fourth-year students, Wilks’s lambda = .803, F(4,44) = 2.70, p = .043; η2 = .197. Univariate analysis of variance revealed that first- and second-year students were less confident than third- and fourth-year students in giving and receiving directions, F(1,44) = 9.03, p=.004. See Table 3 for means and standard deviations of first- and second-year students and third- and fourth-year students.
An independent samples t-test was employed to see if participants primed with landmark-based directions would use more landmarks when writing their own directions and whether participants primed with cardinal-based directions would use more cardinal terms when writing their own directions. Participants in the landmark condition, t(48) = -3.54, p=.001, scored significantly differently from those in the cardinal condition, t(48) = 2.27, p=.028. Means and standard deviations for the number of landmark and cardinal references used in written directions can be seen in Table 4.
A chi-square test was conducted to examine the relationship between condition and the assumed gender of the person who wrote the directions. The chi-square indicated that participants in the cardinal condition tended to assume that the directions were written by a man and participants in the landmark condition would indicate that the directions were written by a woman, X2(1, N=46) = 9.23, p=.002. In the cardinal condition, 45% of participants thought the directions were written by woman, and 54% thought the directions were written by a man. In the landmark condition, 87.5% thought the directions were written by a woman, and 12.5% thought the directions were written by a man.
Discussion
The hypothesis that men would prefer cardinal directions compared to landmark directions and women would prefer landmark directions compared to cardinal directions was supported. The factorial MANOVA indicated significant differences between men and women and their preferred wayfinding strategy. This supports results of previous research. When navigating with a 2-D map, our results indicate gendered preferences in written directions; men prefer orientational cues, whereas women prefer route cues.
According to our simple effects test, participants reported higher levels of anxiety in the cardinal condition than in the landmark condition. This finding indicates that, regardless of gender, participants in the cardinal condition were more anxious overall than those in the landmark condition, which suggests that cardinal directions may elicit more anxiety during wayfinding tasks.
The hypothesis that women would report higher anxiety levels than men in wayfinding tasks was supported by the one-way ANOVA. Women reported significantly higher levels of anxiety on wayfinding tasks than did men. Similar to previous research, these results suggest that women report more anxious behavior than men when performing the same navigational task.
The hypothesis that men would report higher levels of confidence in wayfinding tasks than women was not supported by a one-way ANOVA. There were no significant gender differences in self-reported confidence levels. Despite our findings, previous research has indicated that men report more confidence in their wayfinding abilities than women do, possibly due to pressures of masculinity to appear confident. Although that hypothesis was not supported, results indicated that third and fourth-year students reported higher confidence than first- and second-year students in giving and receiving directions. This result indicates that familiarity with the campus could have contributed to higher confidence levels. Therefore, freshmen and sophomores may be less familiar and less confident with the campus than juniors and seniors.
We also found that participants primed with landmark-based directions used more landmarks when writing their own directions, and participants primed with cardinal-based directions used more cardinal terms when writing their own directions. The independent samples t-test demonstrated that the type of directions (i.e. landmark-based versus cardinal-based) influenced how participants wrote their own directions. Participants may change the way they write directions based on what they think the researcher expects; therefore, priming influences how the participant provides directions.
A chi-square analysis indicated that participants in the cardinal condition were more likely to guess that the directions were written by a man and participants in the landmark condition were more likely to guess that the directions were written by a woman. These results add to the wayfinding research by demonstrating that direction content influences perception of gender based on route and orientation terminology. Directions primarily loaded with cardinal directions such as north, south, east, or west were assumed to have been written by a man and directions using landmarks such as a dorm, a building, or a statue were assumed to have been written by a woman.
The current results aid in the understanding of how men and women differ in navigation and wayfinding strategies using a 2-D map. Lawton and Kallai (2002) found that women utilized landmark cues more often than men and also that women reported being more anxious during wayfinding than men, and our results replicate the findings. Women reported higher levels of anxiety on wayfinding tasks compared to men and also preferred to use landmark references when giving and receiving directions.
This study adds to the research on wayfinding and spatial cognition in that familiarity of an environment may influence confidence levels as also suggested by Devlin (2001) and Prestopnik and Roskos-Ewoldse (2000). Our results indicate that third- and fourth-year students reported significantly higher levels of confidence on the wayfinding tasks than did first- and second-year students. Their confidence level may be influenced by the increased familiarity of the campus compared to first- and second-year students; given that juniors and seniors have been at the college longer than their freshmen and sophomore counterparts.
One of the limitations of the present study is a small and homogeneous sample size, which cannot be generalized to the greater population. It is also possible that instructions and directions of the different tasks could have been misinterpreted by the participants. The instructions might have been unclear in terms of if the participants were to answer the wayfinding strategy scale and spatial anxiety scale in reference to the wayfinding activity, the direction-giving task, or to the participants’ daily wayfinding strategies. Assumptions about the researchers’ gender may have persuaded the participant about which gender wrote the directions rather than using the content of the directions. In the task where the participants had to write their own directions from the given start and end location, the participants may have been influenced by the condition they were exposed to – either cardinal-based or landmark-based directions. Whether participants were in the cardinal or landmark group, this first task may have primed their response for the next task. This limitation indicates that the directions may not be representative of the directions they would provide without the prior prompt. Lastly, some of the maps provided to the cardinal group did not contain a compass because we expected participants to recognize the cardinal orientation of campus. Therefore, participants had to be familiar enough with the campus in order to determine the cardinal locations, provided that the top of the map was not north.
One possible threat to the internal validity of the present study could have been the location disparity given that the experiment was conducted in an academic building as well as in a library setting. The environment in which the participants took the survey might have influenced their performance given where they were located on campus. Confounding variables included individual differences regarding spatial cognition and wayfinding expertise when reading and navigating a map. As Devlin (2001) discussed, maps may hinder wayfinding strategies especially when people are unable to connect the environment to what the map represents. As seniors and juniors may be more familiar with the campus environment than freshmen and sophomores, they may have been more likely to associate the map with the campus properly and its features than first- and second-year students who have not had as much exposure. Implications of this study can be applied to the field of cartography in that people respond differently to directional cues and use different wayfinding strategies. Signage should be created with both genders in mind to ease navigation.
Future researchers could follow a similar procedure by adding the Santa Barbara Sense-of-Direction Scale as used in the Ishikawa, Fujiwara, Imai, and Okabe (2008) study. That scale could provide insight as to participants’ self-perception of their wayfinding ability and would be useful to compare their self-reported results to their actual performance on the tasks. Further research should assess perceived ability compared to realistic ability, and analyzing possible discrepancy would be valuable. Also future researchers could ask participants to follow written directions and navigate the route on foot around a given area. Furthermore, another potential topic of research could examine wayfinding abilities in natural environments, such as a wooded environment. Another study could compare natural environments to built environments to see if men and women differ on wayfinding strategies, confidence, and anxiety in different environments.
References
- Choi, J., & Silverman, I. (2003). Process underlying sex differences in route-learning strategies in children and adolescents. Personality and Individual Differences, 34, 1153-1166. doi:10. 0191-8869/02
- Devlin, A. S. (2001). Mind and maze: Spatial cognition and environmental behavior. Westport, CT: Praeger.
- Devlin, A. S., & Bernstein, J. (1995). Interactive wayfinding: Use of cues by men and women. Journal of Environmental Psychology, 15, 23-38. doi:10.0272494&96/010093
- Hund, A. M., & Padgitt, A. J. (2010). Direction giving and following in the service of wayfinding in a complex indoor environment. Journal of Environmental Psychology, 30, 553–56. doi:10.1016/j.jenvp.2010.01.00
- Ishikawa, I., Fujiwara, H., Imai, O., & Okabe, A. (2008). Wayfinding with a GPS-based mobile navigation system: A comparison with maps and direct experience. Journal of Environmental Psychology, 28, 74-82. doi:10.1016/j.jenvp.2007.09.00
- Lawton, C. A. (1996). Strategies for indoor wayfinding: The role of orientation. Journal of Environmental Psychology, 16, 137-145. doi:10.0272-4944/96/020137
- Lawton, C. A., & Kallai, J. (2002). Gender differences in wayfinding strategies and anxiety about wayfinding: A cross-cultural comparison. Sex Roles, 47, 389-401.
- doi:10.0360-0025/02/1100
- Murakoshi S., & Kawai, M. (2000). Use of knowledge and heuristics for wayfinding in an artificial environment. Environment and Behavior, 32, 756-774. doi:10.1177/00139160021972784
- O’Laughlin, E. M., & Brubaker, B. S. (1998). Use of landmarks in cognitive mapping: Gender differences in self report versus performance. Personality and Individual Differences, 24, 595-601. doi:10.1016/S0191-8869(97)00237-7
- Prestopnik, J. L., & Roskos-Ewoldsen, B. (2000). The relations among wayfinding strategy use, sense of direction, sex, familiarity, and wayfinding ability. Journal of Environmental Psychology, 20, 177-191. doi:10.1006/jevp.1999.016.
- Tlauka, M., Brolese, A., Pomeroy, D., & Hobbs, W. (2005). Gender differences in spatial knowledge acquired through simulated exploration of a virtual shopping centre. Journal of Environmental Psychology, 25, 111–118. doi:10.1016/j.jenvp.2004.12.00
- Ward, S. L., Newcombe, N., & Overton, W. F. (1986). Turn left at the church, or three miles north: A study of direction giving and sex differences. Environment and Behavior, 18, 179-213.