Eye tracking, strategies, and sex differences in virtual navigation

https://doi.org/10.1016/j.nlm.2011.09.007Get rights and content

Abstract

Reports of sex differences in wayfinding have typically used paradigms sensitive to the female advantage (navigation by landmarks) or sensitive to the male advantage (navigation by cardinal directions, Euclidian coordinates, environmental geometry, and absolute distances). The current virtual navigation paradigm allowed both men and women an equal advantage. We studied sex differences by systematically varying the number of landmarks. Eye tracking was used to quantify sex differences in landmark utilisation as participants solved an eight-arm radial maze task within different virtual environments. To solve the task, participants were required to remember the locations of target objects within environments containing 0, 2, 4, 6, or 8 landmarks.

We found that, as the number of landmarks available in the environment increases, the proportion of time men and women spend looking at landmarks and the number of landmarks they use to find their way increases. Eye tracking confirmed that women rely more on landmarks to navigate, although landmark fixations were also associated with an increase in task completion time. Sex differences in navigational behaviour occurred only in environments devoid of landmarks and disappeared in environments containing multiple landmarks. Moreover, women showed sustained landmark-oriented gaze, while men’s decreased over time. Finally, we found that men and women use spatial and response strategies to the same extent. Together, these results shed new light on the discrepancy in landmark utilisation between men and women and help explain the differences in navigational behaviour previously reported.

Highlights

► Eye tracking confirmed the view that women navigate by using landmarks more than men. ► Sex differences occurred only in environments devoid of visual landmarks but disappeared with multiple landmarks. ► Sustained landmark-oriented gaze was found in women but not in men. ► Men and women used spatial and response strategies to the same extent. ► The percentage of time looking at landmarks was greater in spatial learners than in response learners.

Introduction

The ability to successfully navigate through known and novel environments is essential in modern life. Finding one’s way to and from locations such as school, work, and home, or orienting oneself in a new city are necessary for daily functioning. However, there is a large variance in the ability to successfully navigate when placed in a novel environment. In particular, numerous studies have found sex differences in navigational ability, some favouring women and others favouring men depending on the method used (Andreano and Cahill, 2009, Astur et al., 1998, Astur et al., 2004, Driscoll et al., 2005, Malinowski, 2001, Sakthivel et al., 1999, Sandstrom et al., 1998, Saucier et al., 2002, Silverman and Eals, 1992, Spiers et al., 2008).

To reach a particular location, one can rely upon two distinct strategies. A spatial strategy involves the construction of a cognitive map of an environment, in which the relative positions of multiple landmarks in space are encoded as the navigator moves. Functional neuroimaging and lesion studies in humans as well as in animals have consistently identified the hippocampus’ role in spatial memory (Bohbot et al., 2004, Hartley et al., 2003, Iaria et al., 2003, Maguire et al., 2003). Conversely, a response strategy involves learning sequences of body movements in response to a stimulus, such as a starting position or a particular environmental feature. Functional neuroimaging studies have associated this response strategy with activation in the striatum, particularly the caudate nucleus (Bohbot et al., 2004, Hartley et al., 2003, Iaria et al., 2003, Packard et al., 1989, White and McDonald, 2002). Women and men were found to be using spatial and response strategies in equal proportions (Bohbot et al., 2004, Iaria et al., 2003, Levy et al., 2005), suggesting that the male advantage in using Euclidian maps or the female advantage in landmark utilisation does not impact the spontaneous use of spatial and response strategies.

Saucier et al. (2002) were able to determine differences in men and women’s preferential cue utilisation in a real-world navigation task. Participants searched for various locations by following two types of instructions, either landmark- or Euclidian-based directions. It was found that women made fewer errors and took less time to complete the task when they were asked to navigate by following landmarks than when they were instructed to use distances and cardinal directions. Men performed equally well using either method. Astur et al. (2004) tested men and women on two virtual spatial memory tests, the Radial Arm Maze and the Morris Water Maze. In both tasks, men took less time to find targets than women, though actual distance traveled did not differ significantly between men and women. The authors hypothesised that these differences in performance reflected different navigational methods between men and women. Similarly, Sandstrom et al. (1998) manipulated the availability of landmarks and room geometry. Men and women were trained in a virtual Morris Water Maze that featured unique room geometry and landmarks. The shape of the room or surrounding landmarks was altered to differentiate which cues were used by participants. While men were able to navigate using either landmarks or room geometry, women’s performance was impaired when landmarks were removed. Levy et al. (2005) did not find sex differences in strategies used in a “T maze” or in performance on a radial arm maze task that contained landmarks later in training. Interestingly, early in training, they found a small but significant bias in women using a spatial strategy consistent with the fact that they used landmarks to orient themselves. Together, these studies suggest that while men and women perform with similar accuracy, they differ in their reliance on landmarks. Men can make use of Euclidian coordinates (Dabbs et al., 1998, Lambrey and Berthoz, 2007, Saucier et al., 2002), environmental geometry (Sandstrom et al., 1998), absolute distances (Dabbs et al., 1998, Postma et al., 2004, Ruggiero et al., 2008), and mental rotation (Malinowski, 2001) in order to orient themselves, and are therefore able to perform equally well when one type of environmental information is missing. Conversely, women seem to rely more on landmarks when navigating (Dabbs et al., 1998, Lambrey and Berthoz, 2007), and their performance decreases when none are available.

In the present study, we tracked eye movements to examine whether variations in gaze behaviour underlie the specific differences in navigational behaviour between men and women. We manipulated the impact of landmark availability on gaze and navigational behaviour by systematically increasing or decreasing the number of landmarks in the environment. It was hypothesised that a sex difference in gaze behaviour would be observed, as previous research has shown differences in eye movement allocations between men and women (Campagne et al., 2005, Miyahira et al., 2000, Miyahira et al., 2000, Mueller et al., 2008). Campagne et al. (2005) followed individuals’ eye movements as they performed a simulated driving task. Following a prolonged navigation period, a sex difference was observed in the gaze pattern. Specifically, whereas men quickly reduced the frequency of glances to an attentional target, the frequency of glances in women did not decrease as quickly, indicating differences in gaze allocation to attentional targets. In accordance with these results, a study by Mueller and colleagues (2008) found that men’s fixation durations decreased faster than women’s in a virtual Morris Water Maze task, where participants had to learn the spatial features of an environment in the learning trials. Moreover, men were found to visually explore more space early in the task than women, which the authors argue is an indication that men were encoding spatial relations between features of the environment more so than women. However, there were no significant sex differences in the amount of time spent looking at environmental features. In a study by Miyahira et al. (2000), the authors investigated sex differences in the distribution of eye movements when viewing fixed scenes. By tracking eye movements using head-mounted video cameras, visual exploration could be accurately measured across four stimuli of increasing complexity, from blank circles to landscape scenes. It was found that eye movements for the scene were significantly different from the simpler stimuli, and mean gazing time of women was higher than that of men. The literature therefore suggests that the distribution of eye movements between men and women differs when a high attentional demand is required or when the complexity of the visual target increases.

If the ability to remember previously visited places depends on the type of information used in the environment, it was hypothesised that this would be reflected in the distribution of gaze to the visual stimuli used as reference points. It was thought that differences in navigational method might result in different visual search patterns, as determined by the frequency and length of landmark-directed eye movements. Specifically, because men typically rely on multiple sources of information to orient themselves, such as room geometry, Euclidian coordinates, or distance from a particular reference point, we hypothesised that this would result in an overall low allocation of visual gaze to landmarks. Conversely, as women usually orient themselves by forming associations between the various landmarks that make up an environment and their position in relation to them, there should be an associated increase in landmark-directed gaze. We therefore expected to see a greater number of landmark-directed eye movements in women than men.

Section snippets

Participants

Seven (four men, three women) members of the student population of McGill University and staff of the Douglas Mental Health University Institute volunteered for the experiment. Participants were between 21 and 37 years old (mean age = 28.17 ± 5.67) and had normal or corrected-to-normal vision. Video game experience was determined for each participant and was correlated with age, sex, strategy, latency, and number of fixations. Informed consent was obtained from all participants in accordance to the

Navigational performance

Men and women differed in their performance on the 4-on-8 virtual maze. Overall, women took longer to complete the task on a given trial (F1,198 = 44.171, p < 0.001) and made more errors (F1,206 = 6.396, p < 0.05) than men. There were no sex differences in the 8-landmark condition in terms of the number of errors or latencies. However, when the number of landmarks was reduced, performance-related sex differences emerged. Women took more time to complete a given trial in the 0-, 2-, 4- and 6-landmark

Discussion

Eye tracking technology was employed to measure eye movements as participants completed a navigation task that could be solved by using a spatial or response strategy. The purpose of this experiment was to examine the differences in visual acquisition of relevant orientation-related information and navigational behaviour between men and women as the number of landmarks systematically varied.

Conclusion

In summary, the present study found evidence of sex differences in the acquisition of visual information and the way it is applied to solve a 4-on-8 virtual maze. Gaze and navigation behaviour are not static processes but are modulated by many factors, in particular the number of landmarks and time. For example, as the number of landmarks available in the environment increases, the proportion of time spent looking at landmarks and the number of landmarks used increases. The sex differences in

Acknowledgments

This project was funded by CIHR grant number: 64381, FRSQ Grant Number: 3234 and the John R. & Clara M. Fraser Memorial Award. We wish to thank the Canadian Foundation for Innovation for an infrastructure grant that funded the laboratory equipment, CFI Project no: 9357. We also wish to thank Jean-Sebastien Provost for his help in constructing the Unreal Maps, Sam McKenzie for his help with the manuscript, and Nicolas Kyle and Wai Keung Kam for coding the data.

References (46)

  • A. Miyahira et al.

    Gender differences of exploratory eye movements: A life span study

    Life Sciences

    (2000)
  • S.C. Mueller et al.

    Sex differences in a virtual water maze: An eye tracking and pupillometry study

    Behavioural Brain Research

    (2008)
  • E.A. Over et al.

    Coarse-to-fine eye movement strategy in visual search

    Vision Research

    (2007)
  • A. Postma et al.

    Sex differences for selective forms of spatial memory

    Brain and Cognition

    (2004)
  • N.J. Sandstrom et al.

    Males and females use different distal cues in a virtual environment navigation task

    Brain Research. Cognitive Brain Research

    (1998)
  • L.S. Tottenham et al.

    Female advantage for spatial location memory in both static and dynamic environments

    Brain and Cognition

    (2003)
  • N.M. White et al.

    Multiple parallel memory systems in the brain of the rat

    Neurobiology of Learning and Memory

    (2002)
  • J.M. Andreano et al.

    Sex influences on the neurobiology of learning and memory

    Learning and Memory

    (2009)
  • C.L. Barkley et al.

    Sex differences in cue perception in a visual scene: Investigation of cue type

    Behavioral Neuroscience

    (2007)
  • A.M. Barnfield

    Development of sex differences in spatial memory

    Perceptual and Motor Skills

    (1999)
  • V.D. Bohbot et al.

    Hippocampal function and spatial memory: Evidence from functional neuroimaging in healthy participants and performance of patients with medial temporal lobe resections

    Neuropsychology

    (2004)
  • V.D. Bohbot et al.

    Gray matter differences correlate with spontaneous strategies in a human virtual navigation task

    Journal of Neuroscience

    (2007)
  • L. Cahill

    Why sex matters for neuroscience

    Nature Reviews Neuroscience

    (2006)
  • Cited by (111)

    • The removal and addition of cues does not impair spatial retrieval and leads to a different metabolic activity of the limbic network in female rats

      2022, Brain Research Bulletin
      Citation Excerpt :

      Males and females are known to behave differently in spatial navigation procedures, with evidence from human (Castillo et al., 2021; Fernandez-Baizan et al., 2019; León et al., 2016; Sneider et al., 2015; Yu et al., 2021) and animal studies (Mifflin et al., 2021; Qi et al., 2016; Safari et al., 2021; Simpson and Kelly, 2012; Yagi et al., 2017). Some authors attribute better male performance to task difficulty (Chamizo et al., 2020), different swimming patterns (Devan et al., 2016), motivation to complete the task (Mifflin et al., 2021), or different spatial strategies, suggesting that males tend to use geometric information whereas females rely more on landmarks during navigation (Aguilar-Latorre et al., 2022; Andersen et al., 2012; Chamizo et al., 2016). Nevertheless, although there are studies including females, most of the research in spatial learning and memory processes using the MWM has been performed only in males, leading to an under-representation of female behavioral and brain-related responses.

    View all citing articles on Scopus
    View full text