Women Who Know Their PlaceSex-Based Differences in Spatial Abilitiesand Their Evolutionary Significance

Ariane Burke & Anne Kandler & David Good

Published online: 31 May 2012# Springer Science+Business Media, LLC 2012

Abstract Differences between men and women in the performance of tests designedto measure spatial abilities are explained by evolutionary psychologists in terms ofadaptive design. The Hunter-Gatherer Theory of Spatial Ability suggests that theadoption of a hunter-gatherer lifestyle (assuming a sexual division of labor) createddifferential selective pressure on the development of spatial skills in men and womenand, therefore, cognitive differences between the sexes. Here, we examine a basicspatial skill—wayfinding (the ability to plan routes and navigate a landscape)—inmen and women in a natural, real-world setting as a means of testing the propositionthat sex-based differences in spatial ability exist outside of the laboratory. Our resultsindicate that when physical differences are accounted for, men and women withequivalent experience perform equally well at complex navigation tasks in a real-world setting. We conclude that experience, gendered patterns of activity, and self-assessment are contributing factors in producing previously reported differences inspatial ability.

Hum Nat (2012) 23:133–148DOI 10.1007/s12110-012-9140-1

A. Burke (*)Département d’Anthropologie, Université de Montréal, C.P. 6128 succursale Centre-Ville, Montréal,QC, Canada H3C 3J7e-mail: a.burke@umontreal.ca

A. KandlerAHRC Centre for the Evolution of Cultural Diversity, Institute of Archaeology, University Collegeof London, 31-34 Gordon Square, London WC1H 0PY, UK

A. KandlerSanta Fe Institute, 1399 Hyde Park Road, Santa Fe, NM 87501, USA

D. GoodDepartment of Social and Developmental Psychology, University of Cambridge, Free School Lane,Cambridge CB2 3RQ, UK

Keywords Evolutionary psychology . Spatial cognition . Gender .Wayfinding

Introduction

Numerous studies report differences between the sexes in the performance of testsdesigned to measure spatial abilities in humans; these differences are thought toreflect fundamental differences in spatial cognition. The observed effect is strongestin tests of spatial perception and mental rotation (Linn and Petersen 1985; Montello etal. 1999; Voyer et al. 1995) in which men reliably outperform women—althoughcontradictory results are also reported (Rilea et al. 2004). Men also tend to performbetter than women in map-reading tests (Beatty 2002; Dabbs et al. 1998), pointingtasks (Lawton 1996), and direction-giving tasks (Dabbs et al. 1998; Lawton 2001). Inturn, women perform better than men in tests designed to assess spatial memory (theability to memorize spatial configurations), in terms of both the content and thelocation of the arrays (Barnfield 1999; Levy et al. 2005; Tottenham et al. 2003). Insome route-learning experiments (Choi et al. 2006), women perform better than men,although no differences are noted in other experiments (Golledge et al. 1993).

As several authors have noted, however, the link between laboratory tests and theperformance of spatial tasks in the real world is not clear (Coluccia and Louse 2004;Jones et al. 2003; Montello et al. 1999; Nadolne and Stringer 2001; Wynn et al.1996). Widely reported differences in visuo-spatial ability in experimental settings,for example, have generally not been addressed in terms of actual wayfinding skills(Bosco et al. 2004). In most mental rotation tasks, geometric space (e.g., rotatingcubes) is the object of study, rather than geographic space—a distinction that maywell be significant (Lobben 2004). Finally, cultural influences on the performance ofspatial tasks (e.g., the opportunity to develop spatial competencies as a function ofgendered roles in a wider social context) are rarely considered when interpreting testresults (Saucier et al. 2002; Shelton and Gabrieli 2004) but are clearly relevant(Lawton 1994; Quaiser-Pohl and Lehmann 2002). Many studies of spatial abilityare performed in the laboratory or in urban settings—in other words, not in culturallyneutral environments. As a result, the implications of differences in spatial abilitybetween men and women observed in the laboratory for the performance of orienta-tion tasks in the real world is unclear (Coluccia and Louse 2004; Gilmartin and Patton1984).

The proposition that men and women differ in their wayfinding abilities in anatural, large-scale environment is central to the “Hunter-Gatherer Theory” of spatialdifferences (HGT), a hypothesis generated by evolutionary psychologists that seeksto explain sex-based differences in spatial abilities in terms of natural and sexualselection, operating on human cognition during the Pleistocene (Eals and Silverman1994; New et al. 2007; Silverman et al. 2000; Silverman and Eals 1992). Accordingto the HGT, sex-specific patterns of spatial behavior emerged with the appearance ofa hunter-gatherer way of life, accompanied by a sexual division of labor (Silvermanand Eals 1992:514). Men in prehistory are assumed to have been predominantlyhunters, ranging widely in unfamiliar surroundings in search of game (Silverman andEals 1992). This heightened mobility in men is thought to have resulted in theadoption of navigation strategies that use global coordinates and rely on distance

134 Hum Nat (2012) 23:133–148

and direction cues (James and Kimura 1997:162). The corresponding argument is thatsmaller-scale mobility, associated with foraging activities, created the context for theselection of spatial memory in women (Eals and Silverman 1994; McBurney et al.1997; New et al. 2007). This hypothesis would explain why men excel in mentalrotation tests, while women excel in object location tests.

In this research we generate new, empirical data in order to test the proposition thatmen and women differ in their wayfinding abilities in a natural, large-scale environ-ment. Various definitions of the term wayfinding exist. In this research, we use thedefinition provided by Golledge (1999:47): “purposeful movement to a specificdestination that is distal and thus cannot be perceived by the traveller.” In wayfinding,the destination may be known or novel and, if it is not connected by an existing pathor track to the point of origin, may involve exploratory travel. Our test population isdrawn from participants in an international orienteering event. The results of thisresearch are used to evaluate the HGT and related hypotheses linking the advent of ahunting and gathering lifestyle to the appearance of sex- or gender-based differencesin spatial ability.

Spatial Cognition and Wayfinding

The ability to plan and carry out navigation tasks is governed by underlying cognitiveprocesses that use two, parallel representations of the world: an egocentric represen-tation, or “transient action-oriented egocentric self-object associations,” and an allo-centric representation, which is a “more enduring . . . object-object or environment-object association” (Burgess 2006:555). Allocentric and egocentric representationsare linked by a process of translation (supra) elsewhere referred to as “spatialupdating”; this process is key to orientating oneself (recognizing one’s currentposition and translating that position onto the enduring representation). Internalrepresentations can be built up either from direct experience of the world or fromexternal aids (maps). Allocentric representations are essential to building “cognitivemaps” or cognitive representations of space (Golledge 1999, 2003; Kitchin 1994).

The process of building cognitive maps from direct experience involves pathintegration and the forming of networks (Golledge 1999; Loomis et al. 1999) as wellas landmark recognition (Foo et al. 2005). Golledge (2003:30) argues that use of acognitive map implies motivated and deliberate encoding of environmental informa-tion. Allocentric representations are updated with new information as individualsexperience their environment. Landmark recognition (the ability to select relevantfeatures of the landscape and consign their position to memory) is one of the skillsthat allows people to orient themselves relative to immediately perceived stimuli(egocentric representation) and then to locate their current position relative to theallocentric representation. Landmark recognition, in other words, is a required ele-ment in the process of translation between spatial perspectives.

In the sport of orienteering, spatial perception, visualization, and spatio-temporalskills are all tested, as well as memory, concentration, and, of course, physical ability.Orienteers make use of mapped terrain detail to generate a referential network (orcognitive map) with “spatial and temporal implications for possible actions” (Johansen1997). The orienteer chooses a route between control points, predicting which

Hum Nat (2012) 23:133–148 135

features of the landscape should appear along the selected route; these will serve ascues for action. The anticipated route is harmonized with the landscape as it isperceived during the run (a.k.a. spatial updating) and cues for action are acted upon.Failure to update their spatial position relative to the map, known in orienteeringcircles as “losing touch with the map,” results in time loss while the competitorrelocates. Relocation strategies include matching the current perspective with per-spectives offered by the map at various hypothesized current locations (searching forrecognizable landmarks) and retracing one’s route to a previously encountered featureon the landscape (Whitaker and Cuqlock-Knopp 1992).

Previous tests of spatial ability in real-world settings are rare. Sandstrom et al.(1998) point out that the study of navigational ability in humans is limited by therestricted availability and inconvenience of using large-scale, novel environments.Two studies conducted on orienteering courses are an exception (Malinowski 2001;Malinowski and Gillespie 2001). In the first study (Malinowski and Gillespie 2001),the participants were military college students; in the second (Malinowski 2001), theywere university students. Both studies report a correlation between mental rotationtasks and wayfinding performance, and gender is offered as the basis of this differ-ence. Although map-reading and orientation skills were provided prior to the events,this does not in itself mean that the effect of prior experience was eliminated.

The sport of orienteering, therefore, provides an excellent opportunity to studywayfinding competence in a large population of men and women of all ages whoregularly use their navigation skills in novel territory and overwhelmingly naturalsettings. Furthermore, men and women frequently run identical courses, increasingthe potential for sex- and gender-based comparisons.

Orienteering

Orienteering is a sport in which participants navigate a course of their own choosingbetween a series of prescribed landscape features, or “control points,” in a predefinedorder as fast as possible. Orienteering competitions take place in a variety of settings,including city parks; in this research we concentrate on events taking place in mostlynatural settings (forests, heathlands).

Orienteering maps are typically produced on a scale of 1:15,000 or 1:10,000 usingbase maps with 5 m topographic contours. Additional information, including fences,paths and roads, cover type (vegetation), and other prominent features (e.g.,distinct trees or boulders), is mapped onto the base map using five standardcolors. Forest cover is categorized by color into three classes according torunability. Participants use compasses to aid navigation and orient the maps,which are drawn using magnetic north (rather than true north). Maps areoverprinted with a course consisting of a series of control points that must bevisited in a given order. Participants compete individually and race at three-minute intervals to discourage following.

Orienteering courses are ranked by degree of technical difficulty (TD) from 1(easiest) to 5* (hardest); in this study we analyzed adult competitive classes runningTD05*. TD 5* is defined as “Hard but fair,” which means that the course shouldoffer significant route choices and should force regular changes in technique (e.g.,

136 Hum Nat (2012) 23:133–148

“long route choice followed by short intricate legs”).1 Course designers are testingcompetitors with respect to the following abilities:

& navigate for long distances using only major contour features& read and interpret complex contours& concentration over long distances& recognition of indistinct features& use all the different skills and adapt speed and technique to changes in the terrain

and orienteering difficulty

Control sites for TD 5* should be distant from obvious relocating features (arelocating feature is an obvious landmark, such as a linear feature, the summit of a largehill, major valley bottoms, or significant slope changes). Finally, two distinct strategiesmay be employed (alone or in combination) to navigate between control points. Acontrol point

will need to be found by either careful map reading all the way in from arelatively distant attack point or running roughly into the vicinity of the control,relocating using the available contour detail, then swinging into the controlitself. In either case, the ability to relate small-scale (5 m high rather than 25 mhigh) relief to contour detail on the map should be being tested.

In most national and international orienteering events, such as the Scottish 6-DayOrienteering Festival, electronic data collection is used to record split times and toverify that a course has been navigated correctly. Competitors carry a SPORTidentdatachip (http://www.sportident.com/en/) and upload timing information from thecontrol points encountered on their course. At the finish, information from each chipis downloaded to a computer and competitors are ranked for each class/course. Theresulting database, combined with course information obtained from the maps, cantherefore be used to assess differences in navigation ability, cross-referenced by ageand gender. Failure to navigate a clean course results in time loss; finishing times,therefore, are a first approximation of wayfinding abilities.

To summarize, in this research we test the proposition that men and women differ intheir ability to plan routes and navigate in a large-scale environment using an expertsample of men and women with commensurable prior experience. The results of thisresearch are used to evaluate hypotheses brought forward by evolutionary psychologistslinking the advent of a hunting and gathering lifestyle to the appearance of sex-baseddifferences in spatial ability.

Methods

Participants

Our target group is drawn from a pool of adult competitors taking part in aninternational orienteering event held every 2 years in the U.K. Participants in the1 The source for definitions, rules, and regulations cited here is the British Orienteering websitehttp://www.britishorienteering.org.uk/images/uploaded/downloads/coaching_support_td5.pdf, accessedMay 5, 2012.

Hum Nat (2012) 23:133–148 137

Scottish 6-Day Orienteering Festival are organized into 17 age categories and segre-gated by sex.2 Courses are set on six novel, mapped areas (averaging 30–35 km2 inarea) used over six successive days of competition.

While it would be wrong to assume that the population under study is represen-tative of the general population, the ratio of women to men in the Scottish 6-Dayevent is the same as reported nationally for people of similar age across all sports,games, and physical activities in the U.K. by the Office for National Statistics (1996and 2002 General Household Surveys; http://www.esds.ac.uk/government/ghs/) andthe Women’s Sports and Fitness Foundation (“Active People Survey” 2008–2009;http://wsff.org.uk/). Gender representation in our study population is, therefore,representative of the gender ratio in the active population of men and women in theU.K. In this research we study adult orienteers, many of whom will have orienteeredfrom a very young age. In the orienteering community, particularly in the U.K. and inScandinavia, where the sport is very popular, men and women are similarly encul-tured with respect to wayfinding. Girls and boys often begin orienteering trainingbefore the age of 10, when they are first allowed to compete. The participants in thisevent, therefore, will have had similar training opportunities, thus lessening the effectof gender-bias on spatial abilities due to previous life experience. A follow-up studywill directly address the issue of previous training (training effect) through theadministration of a questionnaire, which will be used to select a new targetpopulation.

The total number of participants in this study varies from day to day, with samplesizes of between 60 and 80 for women and between 85 and 105 for men (Table 1).Competitors may choose to run fewer than 6 days (for reasons related to work orfamily obligations, fatigue, or injury) without penalty since the final, cumulativescore for the 6-day event is calculated using an individual’s best four results.

Statistical Tests

In this research the number of erroneously identified controls (mis-punches), coursecompletion times, and the actual distance covered (reflecting route choices) are usedto assess the navigation abilities of men and women running the same orienteeringcourses over a six-day period. We chose two courses with technical difficulty (TD)ratings of 5* (the highest technical difficulty rating in orienteering) run by men andwomen aged 40–45 years (course 20) and men and women aged between 45 and50 years (course 18) (Table 1). This choice was dictated by the fact that these were theonly courses where men and women of similar age and experience were runningidentical courses.

Rather than using the number of successfully identified control features, a measureof orienteering ability adopted elsewhere (Malinowski and Gillespie 2001), we usethe number of mis-punches to assess relative success at identifying control features. Amis-punch occurs when a competitor punches an incorrect control point, usually a

2 Participants are divided into 2-year age categories from ages 10 to 35 and 5-year categories from then on.Male and female Elite classes are open to all age groups. There is no maximum age.

138 Hum Nat (2012) 23:133–148

control for another course. The number of mis-punched controls on each course istabulated for men and women and compared as a direct measure of navigation error.

Next, we test the relationship between men’s and women’s finishing times (coursecompletion times) for each day of competition. First, we use a Mann-Whitney test totest for differences in the mean finishing times for men and women over both courses.Nonparametric tests are used in this research because the distribution of men andwomen’s times is not normally distributed but rather skewed toward faster completiontimes. An orienteer may compensate for navigation errors by running faster, and viceversa. That being the case, a competitor’s finishing time in an orienteering eventreflects both their navigation skills and running ability. Luckily, we have a means ofcontrolling for differences in running ability: the run-in. The run-in is the last leg ofeach day and consists of a taped route between the final control and the finishingline—as such, run-in times reflect running ability alone. We use a Mann-Whitney testto test for differences in run-in times, then regression analysis to test the relationshipbetween finishing times and run-in times for both men and women. Since navigationis the variable of interest in this study, we then use the run-in times to control for theeffect of differences in running ability on men’s and women’s finishing times. Theaverage difference in run-in times for men and women on either course is used toadjust the men’s finishing times for each day of competition; the adjusted men’s timesare then compared with the observed women’s finishing times as a means ofcomparing navigation ability whilst controlling for differences in running ability.

Finally, we use the actual distance traveled by competitors as a means of compar-ing men’s and women’s route choices. Actual course lengths (the distance run by acompetitor, rather than the hypothetical course length provided by the planners) areonly available for a restricted sample of competitors. Courses were plotted by thecompetitors on a digitized map using RouteGadget (http://www.routegadget.net/)software hosted by the Spey 2007 website (at http://www.scottish6days.com/), andcourse lengths are generated by the software in numbers of pixels. Digitized routechoices are available for 5 women and 7 men, for course 20 only. We compared the

Table 1 Courses 20 (age 40–45) and 18 (age 45–50). The number of competitors in this table includesmis-punches and disqualified runners

Day 1 Day 2 Day 3 Day 4 Day 5 Day 6

Course 20

Length (km) 5.1 5.5 5.7 5.4 5.6 5.6

Climb (m) 220 230 140 50 85 175

N of controls 17 14 12 17 14 27

Men (40–45) 86 84 85 88 85 84

Women (40–45) 68 66 63 66 68 66

Course 18

Length (km) 4.8 5.2 5.1 4.9 5.2 5.2

Climb (m) 230 210 70 45 90 165

N of controls 16 13 11 18 12 16

Men (45–50) 105 102 104 103 103 99

Women (45–50) 79 76 81 79 76 77

Hum Nat (2012) 23:133–148 139

resulting course lengths (pixel counts) for these competitors as a means of assessingactual distance traveled.

Results

The record of mis-punches shows that women running course 20 were, on average,slightly less prone to mis-punch than men; this trend is reversed in course 18. Overall,men and women orienteers proved equally adept at identifying the correct controlfeatures and mis-punches are rare (Table 2).

Finishing times for women are greater, on average, than finishing times for men(Table 3). Results of an asymmetrical, 2-tailed Mann-Whitney test comparing finish-ing times for men and women (Table 4) confirm that they are significantly different atthe 0.05 level in eight out of twelve cases. Run-in times for men and women alsodiffer significantly. A regression analysis shows that finishing times and run-in timesfor both men and women are unrelated, confirming the hypothesis that coursecompletion times are affected by a more complex set of processes than run-in times(i.e., the combined effect of navigation decisions and running ability).

The run-in times (Table 5) indicate that when running ability alone is measured,women are slower than men by an average of 8.1 % (for course 20) to 8.6 % (forcourse 18). When men’s finishing times are adjusted by the mean difference in run-intimes and compared with the finishing times for women, a 2-tailed Mann–Whitneytest indicates that there is no longer a significant difference between the averagevalues (Table 6). Figure 1 compares the distributions of women’s finishing times andmen’s adjusted finishing times for a representative sample. Similar results areobtained for course 18 and 20 for all 6 days. We conclude that observed differences

Table 2 Percentage of competitors identifying an incorrect control feature (mp)

Day1 Day 2 Day 3 Day 4 Day 5 Day 6 Average

Course 18

mp (M) 0 6 3 2 5 4

mp (W) 1 9 0 0 4 9

N (M) 105 102 104 103 103 99

N (W) 79 76 81 79 75 77

% mp (M) 0.00 5.88 2.88 1.94 4.85 4.04 3.26

% mp (W) 0.95 11.84 0.00 0.00 5.33 11.68 4.96

Course 20

mp (M) 0 6 0 6 9 3

mp (W) 2 4 0 2 3 5

N (M) 86 84 85 88 85 84

N (W) 68 66 63 66 68 66

% mp (M) 0.00 7.14 0.00 6.81 10.58 3.57 4.68

% mp (W) 2.98 6.06 0.00 3.03 4.41 7.57 4.00

140 Hum Nat (2012) 23:133–148

in finishing times between women and men are rooted in differences in runningability, rather than wayfinding ability.

Finally, a comparison of the actual course lengths (the real distance covered by acompetitor as a function of individual route choice) available for men (N07) andwomen (N05) on course 20 shows that men’s and women’s route choices areessentially identical in terms of length.

Discussion

Spatial Tasks in a Real-World Setting

Many of the arguments linking performance in tests of spatial abilities (e.g., mentalrotation tests, water level tests, tests of spatial memory) and real-world skills such aswayfinding are untestable or unsupported (Coluccia and Louse 2004; Jones et al.2003; Nadolne and Stringer 2001; Wynn et al. 1996). The spatial abilities required tosolve problems at one spatial scale may be disassociated from those necessary for

Table 3 Finishing times for women and men (courses 18 and 20): summary statistics

Day 1 Day 2 Day 3 Day 4 Day 5 Day 6

Course 20 Men

Mean 61.33 83.49 63.85 56.57 78.84 71.22

SD 15.04 27.44 14.16 11.79 17.33 15.29

Women

Mean 66.15 86.9 69.94 62.81 84.04 76.43

SD 10.94 21.46 16.38 13.29 18.85 17.69

Course 18 Men

Mean 61.24 76.53 53.59 55.25 72.17 65.40

SD 14.49 21.56 10.35 12.29 16.87 15.10

Women

Mean 69.23 86.38 59.47 61.38 78.39 75.59

SD 14.97 20.61 11.83 11.36 15.11 14.19

Table 4 Comparison of the dis-tribution of men’s and women’sfinishing times for courses 18 and20 (p-values)

Mann-Whitneytest (course 20)

Mann-Whitneytest (course 18)

Day 1 0.169 0.000

Day 2 0.095 0.001

Day 3 0.018 0.001

Day 4 0.008 0.000

Day 5 0.062 0.006

Day 6 0.084 0.000

Hum Nat (2012) 23:133–148 141

solving problems at another scale—for example, mental rotation tests may not betesting the skills required to solve real-world navigation problems (Hegarty et al.2006). In this study, large-scale spatial abilities are measured at the appropriate scale,in terms of the speed, accuracy (number of correctly identified controls), and effi-ciency with which navigation tasks are carried out in a real-world setting.

Men and women in this study do not differ significantly in their ability to find thecorrect control points—which is a direct measure of wayfinding ability. Differencesin course completion times for men and women of comparable age running the samecourse disappear once a sex-based difference in running ability is controlled for.Finally, men’s and women’s route choices do not differ in terms of the actual distancerun by a competitor. Although the shortest distance between two points is theoreti-cally the fastest route choice (and therefore the most desirable), distance is not theonly factor determining the “best” route choice when orienteering. Terrain roughness(amount of climb, vegetation type) and individual fitness levels affect route selection,

Table 5 Mean values of run-in times for men and women, in seconds, and % difference between the meansfor all 6 days

Men Women % difference

Mean SD Mean SD

Course 20

Day 1 28.17 4.96 29.77 5.82 5.6798

Day 2 58.12 11.67 63.71 11.39 9.6180

Day 3 25.86 5.41 27.66 4.30 6.9606

Day 4 28.98 4.12 32.31 4.13 11.4907

Day 5 18.16 3.16 19.56 2.92 7.7093

Day 6 25.51 4.91 27.42 3.87 7.4873

Course 18

Day 1 27.05 4.45 29.45 4.55 2.21

Day 2 57.55 10.59 62.72 10.41 8.9

Day 3 23.87 3.87 27.14 5.64 13.7

Day 4 30.10 6.81 32.90 4.93 9.3

Day 5 18.51 7.51 20.14 3.27 8.81

Day 6 25.84 6.02 28.06 4.59 8.59

Table 6 Comparison betweenadjusted men’s finishing times andobserved finishing times for womenfor courses 18 and 20 (p-values)

Mann-Whitney testsignificance (course 20)

Mann-Whitney testsignificance (course 18)

Day 1 0.707 0.808

Day 2 0.953 0.639

Day 3 0.564 0.750

Day 4 0.320 0.817

Day 5 0.957 0.535

Day 6 0.872 0.199

142 Hum Nat (2012) 23:133–148

making it almost impossible to determine a single “best” route for all competitors(Myrvold 1996). A key element in successful course completion for many, if notmost, orienteers is the ability not only to plot a route but also to calibrate one’s ownphysical capacities and the likelihood of being able to move swiftly along it. If we usedistance as a measure of the efficiency of route choice, our data indicate an overallsimilarity between men and women, although the population available for study is toosmall to draw any real conclusions. Our results, therefore, indicate that wayfindingabilities in men and women with previous orienteering experience are essentiallyequal.

Globally, our results suggest that the differential performances of men and womenin experimental tests of spatial abilities are not linked to an underlying sex-baseddifference in spatial cognition. Instead, gender-based differences in training opportu-nity linked to habitual activity patterns are proposed as the more likely cause of theobserved differences. The importance of similar activity patterns promoting equalspatial performance in men and women in a real-world setting is demonstratedethnographically (Istomin and Dwyer 2009). The existence of a training effect onspatial skills is demonstrated elsewhere (Gibson and Bergman 1954). Even theapparently robust, sex-based differences in the performance of mental rotation tasksare reversible when a target population of men and women is provided with task-appropriate training and experience (Baenninger and Newcombe 1989; Feng et al.2007). Training in orienteering skills provided prior to studies by Malinowski (2001);Malinowski and Gillespie 2001) did not have a similar leveling effect, but the training

Fig. 1 Comparison of the distributions of women’s finishing times and men’s adjusted finishing times forcourse 20, Day 1

Hum Nat (2012) 23:133–148 143

was of short duration and participants had no prior experience in orienteering. Thereare many reasons why this lack of a training effect might exist. The set of skillsinvolved in successfully completing an orienteering course is both extensive andcomplex, and experience using those skills in an authentic environment can be a keyfactor in their consolidation and subsequent successful deployment. Furthermore, thefailure of a training regime by itself may indicate nothing about the ability of thosebeing trained to learn.

The extent to which a more general sex- or gender-linked factor could underlie theobserved differences in performance in both laboratory and mundane conditionsrequires further exploration. One would expect differential experience below somecritical threshold to have a differential effect on performance, and that thresholdwould be relatively high for activities such as orienteering, which integrate a set ofcognitive and perceptual skills that are deployed in a physically demanding motortask. The impact of that experience could be compounded by a generic gender-linkedeffect mediated through variation in levels of self-efficacy (Bandura 1997).

Many studies report a systematic difference between men and women in theirlevels of perceived self-efficacy in different domains, with consequent negativeeffects on performance and personal development. This has long been observed instudies of mathematics for example (Mau and Bikos 2000; O’Brien et al. 1999;Pajares and Miller 1994; Randhawa 1994) and in studies of spatial abilities (Quaiser-Pohl and Lehmann 2002). This gender-linked effect is, however, overturned in casesof substantial experience and effective performance (Reis and Park 2001). It is quiteconceivable that in the case of a developed sense of one’s wayfinding skills, a similarprocess could be at work. The authentic experience of successfully deploying one’sknowledge and skills could lead to increased levels of perceived self-efficacy inwomen, overturning the typical effect of gender. Thus, documented differences inspatial performance would be related to gender, which amplifies culturally basedpreferences and differential experience. An analysis of prior training and experience,perceived self-efficacy levels, and gender in an ecologically valid environment wouldallow the interpretation of differences in wayfinding performance as a function ofgender, self-efficacy, or experience. We will return to this conjecture in a future study.

Conclusion

Our results indicate that when sex-based differences in running ability (speed andstamina) are accounted for, men and women perform equally well at complexnavigation tasks in a real-world setting. We measured this by considering the numberof erroneously identified control points, course completion times (controlling forrelative differences in running ability), and the actual distances covered by compet-itors during a multi-day orienteering event. Our research does not support the claimthat men and women have different spatial abilities relating to movement in the realworld as a result of differential selection acting upon spatial cognition. We suggestinstead that the differences in spatial abilities consistently reported in the literature arethe result of gender bias in training opportunities, rather than underlying sex-baseddifferences in spatial cognition. It seems reasonable to propose that underlyingcognitive skills are present in all adults but that they are developed (expressed as

144 Hum Nat (2012) 23:133–148

“spatial skills”) in adults as needed, according to their habitual activity patterns—aproposal consistent with current neurophysiological research (Maguire et al. 2003).These developments may not be uniform across a population (Janzen et al. 2007) orbetween populations. Alternatively, the experimental tests used to measure spatialskills in the laboratory may not reflect actual, real-world wayfinding skills.

In conclusion, we see no reason to support an evolutionary hypothesis for theemergence of sex-based differences in spatial cognition, such as the hunter-gatherertheory of spatial ability. Another evolutionary hypothesis, closely linked to the HGT,argues in favor of female superiority in object location memory as a result of thepredominance of gathering in women’s foraging activities (McBurney et al. 1997;Neave et al. 2005; New et al. 2007; Voyer et al. 2007; but see Dabbs et al. 1998;James and Kimura 1997; Postma et al. 2004). If women have biologically inheritedstrengths in object-location, which assesses topological memory, this could hypothet-ically make them better orienteers than men since landscape features would be readilyretained in memory and accessed during the run without consulting the map. Ourresults do not support this hypothesis either.

Similar outcomes for men and women in this study do not rule out the possibilitythat men and women are using different cognitive strategies toward the same end,however, since the present study does not address the issue of wayfinding strategies.This question forms the basis of a follow-up study that the authors will conduct inconjunction with the 2009 6-Day Scottish Orienteering Festival.

Acknowledgments This research would not have been possible without the support of the AHRC Centrefor the Evolution of Cultural Diversity, Institute of Archaeology, UCL, with special thanks to Dr. JamesSteele (Director, CECD), and the Leverhulme Centre for Human Evolutionary Studies, Cambridge Uni-versity. The help and advice of organizers of the 2009 Scottish 6-Day Orienteering Festival is gratefullyacknowledged, with special thanks to Mr. Gareth Bryan-Jones. A.K. was supported by a Leverhulme TrustEarly Career Fellowship. The authors would also like to acknowledge Prof. F. Grine, Dr. Jay Stock, and Dr.Tom Smulders for their comments during the preparation of this manuscript.

References

Baenninger, M., & Newcombe, N. (1989). The role of experience in spatial test performance: a meta-analysis. Sex Roles, 20, 327–344.

Bandura, A. (1997). Self-efficacy: the exercise of control. New York: W.H. Freeman.Barnfield, A. M. C. (1999). Development of sex differences in spatial memory. Perceptual and Motor

Skills, 89(1), 339–350.Beatty, W. W. (2002). Sex difference in geographical knowledge: driving experience is not essential.

Journal of the International Neuropsychological Society, 8(6), 804–810.Bosco, A., Longoni, A. M., & Vecchi, T. (2004). Gender effects in spatial orientation: cognitive profiles and

mental strategies. Applied Cognitive Psychology, 18(5), 519–532.Burgess, N. (2006). Spatial memory: how egocentric and allocentric combine. Trends in Cognitive

Sciences, 10(12), 551–557.Choi, J., McKillop, E., Ward, M., & L’Hirondelle, N. (2006). Sex-specific relationships between route-

learning strategies and abilities in a large-scale environment. Environment and Behavior, 38(6), 791–801.

Coluccia, E., & Louse, G. (2004). Gender differences in spatial orientation: a review. Journal of Environ-mental Psychology, 24(3), 329–340.

Dabbs, J. M., Chang, E. L., Strong, R. A., & Milun, R. (1998). Spatial ability, navigation strategy, andgeographic knowledge among men and women. Evolution and Human Behavior, 19(2), 89–98.

Hum Nat (2012) 23:133–148 145

Eals, M., & Silverman, I. (1994). The hunter-gatherer theory of spatial sex differences: proximatefactors mediating the female advantage in recall of object arrays. Ethology and Sociobiology, 15(2), 95–105.

Feng, J., Spence, I., & Pratt, J. (2007). Playing an action video game reduces gender differences in spatialcognition. Psychological Science, 18, 850–855.

Foo, P., Warren, W. H., Duchon, A., & Tarr, M. J. (2005). Do humans integrate routes into a cognitive map?Map- versus landmark-based navigation of novel shortcuts. Journal of Experimental Psychology.Learning, Memory, and Cognition, 31(2), 195–215.

Gibson, E. J., & Bergman, R. (1954). The effect of training on absolute estimation of distance over theground. Journal of Experimental Psychology, 48(6), 473–482.

Gilmartin, P. P., & Patton, J. C. (1984). Comparing the sexes on spatial abilities: map-use skills. Annals ofthe Association of American Geographers, 74(4), 605–619. doi:10.1111/j.1467-8306.1984.tb01477.x.

Golledge, R. G. (1999). Human wayfinding and cognitive maps. In R. G. Golledge (Ed.), Wayfindingbehaviour (pp. 5–45). Baltimore: Johns Hopkins University Press.

Golledge, R. G. (2003). Human wayfinding and cognitive maps. In M. Rockman & J. Steele (Eds.),Colonization of unfamiliar landscapes: the archaeology of adaptation (pp. 25–43). London and NewYork: Routledge.

Golledge, R. G., Ruggles, A. J., Pellegrino, J. W., & Gale, N. D. (1993). Integrating route knowledge in anunfamiliar neighborhood: along and across route experiments. Journal of Environmental Psychology,13(4), 293–307.

Hegarty, M., Montello, D. R., Richardson, A. E., Ishikawa, T., & Lovelace, K. (2006). Spatial abilities atdifferent scales: individual differences in aptitude-test performance and spatial-layout learning. Intel-ligence, 34(2), 151–176.

Istomin, K. V., & Dwyer, M. J. (2009). Finding the way: a critical discussion of anthropological theories ofhuman spatial orientation with reference to reindeer herders of Northeastern Europe and WesternSiberia. Current Anthropology, 50(1), 29–49.

James, T. W., & Kimura, D. (1997). Sex differences in remembering the locations of objects in an array:location-shifts versus location-exchanges. Evolution and Human Behavior, 18(3), 155–163.

Janzen, G., Jansen, C., & van Turennout, M. (2007). Memory consolidation of landmarks in goodnavigators. Hippocampus, 18(1), 40–47.

Johansen, B. T. (1997). Thinking in orienteering. Scientific Journal of Orienteering, 13, 38–46.Jones, C. M., Braithwaite, V. A., & Healy, S. D. (2003). The evolution of sex differences in spatial ability.

Behavioral Neuroscience, 117(3), 403–411.Kitchin, R. M. (1994). Cognitive maps: what are they and why study them? Journal of Environmental

Psychology, 14(1), 1–19.Lawton, C. A. (1994). Gender differences in way-finding strategies: relationship to spatial ability and

spatial anxiety. Sex Roles, 30(11–12), 765–779.Lawton, C. A. (1996). Strategies for indoor wayfinding: the role of orientation. Journal of Environmental

Psychology, 16(2), 137–145.Lawton, C. A. (2001). Gender and regional differences in spatial referents used in direction giving. Sex

Roles, 44(5–6), 321–337.Levy, L. J., Astur, R. S., & Frick, K. M. (2005). Men and women differ in object memory but not

performance of a virtual radial maze. Behavioral Neuroscience, 119(4), 853–862.Linn, M. C., & Petersen, A. C. (1985). Emergence and characterization of sex differences in spatial ability:

a meta-analysis. Child Development, 56(6), 1479–1498.Lobben, A. K. (2004). Tasks, strategies, and cognitive processes associated with navigational map reading:

a review perspective. The Professional Geographer, 56(2), 270–281.Loomis, J. M., Klatzky, R. L., Golledge, R. G., & Philbeck, J. W. (1999). Human navigation by path

integration. In R. G. Golledge (Ed.), Wayfinding behavior (pp. 125–151). Baltimore: Johns HopkinsUniversity Press.

Maguire, A. M., Spiers, H. J., Good, C. D., Hartley, T., Frackowiak, R. S. J., & Burgess, N. (2003).Navigation expertise and the human hippocampus: a structural brain imaging analysis. Hippocampus,13(2), 250–259.

Malinowski, J. C. (2001). Mental rotation and real-world wayfinding. Perceptual and Motor Skills, 92(1),19–30.

Malinowski, J. C., & Gillespie, W. T. (2001). Individual differences in performance on a large-scale, real-world wayfinding task. Journal of Environmental Psychology, 21(1), 73–82.

Mau, W. C., & Bikos, L. H. (2000). Educational and vocational aspirations of minority and female students:a longitudinal study. Journal of Counseling and Development, 78(2), 186–194.

146 Hum Nat (2012) 23:133–148

McBurney, D. H., Gaulin, S. J. C., Devineni, T., & Adams, C. (1997). Superior spatial memory ofwomen: stronger evidence for the gathering hypothesis. Evolution and Human Behavior, 18(3),165–174.

Montello, D. R., Lovelace, K. L., Golledge, R. G., & Self, C. M. (1999). Sex-related differences andsimilarities in geographic and environmental spatial abilities. Annals of the Association of AmericanGeographers, 89(3), 515–534.

Myrvold, B. (1996). Is it possible to find a “best” route? A look at accuracy and significance in route choicecomparison. Scientific Journal of Orienteering, 12, 19–36.

Nadolne, M. J., & Stringer, A. Y. (2001). Ecologic validity in neuropsychological assessment: prediction ofwayfinding. Journal of the International Neuropsychological Society, 7(6), 675–682.

Neave, N., Hamilton, C., Hutton, L., Tildesley, N., & Pickering, A. T. (2005). Some evidence of a femaleadvantage in object location memory using ecologically valid stimuli. Human Nature, 16(2), 146–163.

New, J., Krasnow, M. M., Truxaw, D., & Gaulin, S. J. C. (2007). Spatial adaptations for plant foraging:women excel and calories count. Proceedings of the Royal Society B-Biological Sciences, 274(1626),2679–2684.

O’Brien, V., Martinez-Pons, M., & Kopala, M. (1999). Mathematics self-efficacy, ethnic identity, gender,and career interests related to mathematics and science. The Journal of Educational Research, 92(4),231–235.

Pajares, F., & Miller, M. D. (1994). Role of self-efficacy and self-concept beliefs in mathematical problem-solving: a path-analysis. Journal of Educational Psychology, 86(2), 193–203.

Postma, A., Jager, G., Kessels, R. P. C., Koppeschaar, H. P. F., & van Honk, J. (2004). Sex differences forselective forms of spatial memory. Brain and Cognition, 54(1), 24–34.

Quaiser-Pohl, C., & Lehmann, W. (2002). Girls’ spatial abilities: charting the contributions of experiencesand attitudes in different academic groups. British Journal of Educational Psychology, 72, 245–260.

Randhawa, B. S. (1994). Self-efficacy in mathematics, attitudes, and achievements of boys and girls fromrestricted samples in 2 countries. Perceptual and Motor Skills, 79(2), 1011–1018.

Reis, S. M., & Park, S. (2001). Gender differences in high-achieving students in math and science. Journalfor the Education of the Gifted, 25(1), 52–73.

Rilea, S. L., Roskos-Ewoldsen, B., & Boles, D. (2004). Sex differences in spatial ability: a lateralization offunction approach. Brain and Cognition, 56(3), 332–343.

Sandstrom, N. J., Kaufman, J., & Huettel, S. A. (1998). Males and females use different distal cues in avirtual environment navigation task. Cognitive Brain Research, 6(4), 351–360.

Saucier, D. M., McCreary, D. R., & Saxberg, J. K. J. (2002). Does gender role socialization mediate sexdifferences in mental rotations? Personality and Individual Differences, 32(6), 1101–1111.

Shelton, A. L., & Gabrieli, J. D. E. (2004). Neural correlates of individual differences in spatial learningstrategies. Neuropsychology, 18(3), 442–449.

Silverman, I., Choi, J., Mackewn, A., Fisher, M., Moro, J., & Olshansky, E. (2000). Evolved mechanismsunderlying wayfinding: further studies on the hunter-gatherer theory of spatial sex differences.Evolution and Human Behavior, 21(3), 201–213.

Silverman, I., and Eals, M. (1992). Sex differences in spatial abilities: evolutionnary theory and data. In J.Barkow, L. Cosmides and L. Tooby (Eds.), The adapted mind: evolutionary psychology and thegeneration of culture (pp. 487–503). Oxford University Press.

Tottenham, L. S., Saucier, D., Elias, L., & Gutwin, C. (2003). Female advantage for spatial locationmemory in both static and dynamic environments. Brain and Cognition, 53(2), 381–383.

Voyer, D., Postma, A., Brake, B., & Imperato-McGinley, J. (2007). Gender differences in object locationmemory: a meta-analysis. Psychonomic Bulletin and Review, 14(1), 23–38.

Voyer, D., Voyer, S., & Bryden, M. P. (1995). Magnitude of sex differences in spatial abilities: a meta-analysis and consideration of critical variables. Psychological Bulletin, 117(2), 250–270.

Whitaker, L. A., & Cuqlock-Knopp, G. (1992). Navigation in off-road environments: orienteering inter-views. Scientific Journal of Orienteering, 8, 55–71.

Wynn, T. G., Tierson, F. D., & Palmer, C. T. (1996). Evolution of sex differences in spatial cognition.Yearbook of Physical Anthropology, 39, 11–42.

Ariane Burke obtained a PhD in anthropology from New York University in 1992 and is currently aprofessor in the Department of Anthropology at the Université de Montréal (Canada). Her research interestsinclude studying the causes of consequences of hominin dispersals and landscape archaeology.

Hum Nat (2012) 23:133–148 147

Anne Kandler obtained a PhD in mathematics from the Chemnitz University of Technology (Germany) in2006. She currently holds the Omidyar Fellowship at the Santa Fe Institute, where she is pursuing herinterest in modeling and analyzing cultural shift processes.

David Good is a lecturer in the School of Social and Developmental Psychology, University of Cambridge.His research interests include examining how our understanding of human communication can contribute tothe design and use of new informational and communication technologies and studying the role of socialfactors in the evolution of language and intelligence.

148 Hum Nat (2012) 23:133–148