Spatial mental representations derived from spatial descriptions: The predicting and mediating roles of spatial preferences, strategies, and abilities

Learning and Individual Differences 21 (2011) 150–157

Contents lists available at ScienceDirect

Learning and Individual Differences

j ourna l homepage: www.e lsev ie r.com/ locate / l ind i f

Spatial mental representations derived from survey and route descriptions: Whenindividuals prefer extrinsic frame of reference

Chiara Meneghetti ⁎, Francesca Pazzaglia, Rossana De BeniDepartment of General Psychology, University of Padua, Italy

⁎ Corresponding author. Department of General PsycPadova, Italy.

E-mail address: [email protected] (C. Men

1041-6080/$ – see front matter © 2010 Elsevier Inc. Aldoi:10.1016/j.lindif.2010.12.003

a b s t r a c t

Article history:Received 22 September 2009Received in revised form 9 December 2010Accepted 20 December 2010

Keywords:Spatial individual differencesExtrinsic frame of referenceSpatial descriptionsMental representation

The present research investigates the role of individual differences in preference for adopting extrinsic frameof reference (EFR) in ability to represent mentally spatial information learned through survey and routedescriptions. A sample of 191 participants (100 females and 91 males) was categorized as four groups withhigh (H-EFR), medium-high (MH-EFR), medium-low (ML-EFR) and low (L-EFR) preference. The groupslistened to two spatial descriptions in survey and route perspectives, subsequently performing true/falseverification and map drawing tasks. They also performed a number of visuo-spatial, verbal and geographicalpointing tasks. Results showed a general better performance of the H-EFR group and of males – in comparisonwith lower-ability counterparts and females respectively – in spatial text recall tasks and in other spatialmeasures. Moreover, spatial ability interacts with text perspective: in verification test the H-EFR groupoutperformed the lower groups in survey and route inferential sentences and in map drawing the superiorityof H-EFR group was shown with survey (but not with route) text. Overall, the results suggest that individualswith high preference for extrinsic frame of reference are able to manage environment information in bothspatial perspectives even if they prefer to express it in survey format. That preference is specifically sustainedby spatial abilities.

hology, Via Venezia, 8, 35100

eghetti).

l rights reserved.

© 2010 Elsevier Inc. All rights reserved.

1. Introduction

1.1. Spatial descriptions: survey and route perspective

Oneway of acquiring spatial information is to read or listen to a textdescribing an environment. Spatial descriptions convey environmentalinformation by typically adopting two perspectives, i.e. route or survey,or a combination (Taylor & Tversky, 1992). Route descriptions representthe space from an egocentric perspective (i.e., path view) and use anintrinsic frame of reference (e.g., “to your left”, “behind you”, etc.);survey descriptions represent the space from an allocentric perspective(i.e., a bird's-eye view) and use an extrinsic frame of reference such ascompass directions (i.e., north, south, east, and west).

A question still discussed in the literature is the extent to which theperspective is maintained in spatial mental representation; in somecases, spatialmental representationhas been shown to be dependent onthe perspective learned (Perrig & Kintsch, 1985; Bosco, Filomena,Sardone, Scalisi, & Longoni, 1996; Pazzaglia, Cornoldi, & Longoni, 1994),but in other cases independent of it (Taylor & Tversky, 1992). Perrig andKintsch's (1985) and Taylor and Tversky's (1992) studies are tradition-ally compared for their divergent results. Perrig and Kintsch (1985)found that people's performance of spatial inferential sentences after

reading a route or a surveydescriptions ismodulated byperspective andsex of reader: females were better with the sentences expressed in theperspective as learned than with sentences in a different perspective(not directly learned); males had a better performance with surveysentences than route. This supports the assumption that spatialperspective modulates mental representation. By contrast, Taylor andTversky (1992) found that participants were equally accurate inverifying inferential sentences expressed in both spatial perspectives,independently of those learned. These results support the notion thatspatial information is incorporated into abstractmental representationsthat can be viewed or visualized from several different perspectives.

At the moment, however, results supporting perspective indepen-dence have not been fully corroborated. Whether or not spatialmental representations are independent has been shown to vary as afunction of various factors such as time of text experience (Bosco et al.,1996; Brunyé & Taylor, 2008a), perspective switching withindescriptions (Lee & Tversky, 2005), goals (Taylor, Naylor, & Chechile,1999), instructions (Noordzij, Van der Lubbe, & Postma, 2005, 2006),and test type (Chabanne, Péruch, Denis, & Thinus-Blanc, 2004;Noordzij & Postma, 2005; Shelton & McNamara, 2004; Noordzij,Zuidhoek & Postma, 2006). Recent studies have shown that theprocessing of two types of spatial texts is underpinned by visuo-spatial working memory (WM) but route descriptions involve morespatial sequential WM processes (Brunyé & Taylor, 2008b; Deyzac,Logie, & Denis, 2006; Pazzaglia, Meneghetti, De Beni, & Gyselinck,2010). Overall these studies showed that oneway to disambiguate the

151C. Meneghetti et al. / Learning and Individual Differences 21 (2011) 150–157

results in favor or otherwise of perspective independence is to takeinto account factors that have a role in the construction of mentalrepresentation.

1.2. Spatial descriptions and individual differences

One aspect thatmay have a central role in themodulation of spatialmental representation in the function of spatial perspective is that ofindividual differences. For example, as mentioned above, Perrig andKintsch (1985) found that gender influences the type of mentalrepresentation built (females being more perspective-dependent).Individual differences in spatial ability are also relevant in thisfield andinfluence the construction of mental representation derived fromspatial descriptions. Some studies comparing individualswithhighandlow spatial ability found differences in thememorization of survey text(Bosco et al., 1996; Pazzaglia, 2008) where individuals with highspatial competence appeared to have higher performance than theirlower competence counterparts; in route text, studies found bothsimilar performance (e.g., Meneghetti, Gyselinck, Pazzaglia & De Beni,2009) and differences between groups (e.g., Pazzaglia & Cornoldi,1999). Other studies focusing on individual differences in visuo-spatialWM found that individuals with high spatial span had better spatialtext recall than their lower-ability counterpart (Gyselinck, De Beni,Pazzaglia, Meneghetti, &Mondoloni, 2007; Pazzaglia & Cornoldi, 1999;texts in route perspective).

In this study we opted to take into account individual differencesin the preference for frame of reference. Spatial cognition literatureproposes a distinction between intrinsic (or egocentric) and extrinsic(or allocentric) frame of references that are usually associated withthe use of route and survey strategies, respectively (e.g., Darken &Peterson, 2002; Siegel & White, 1975). Route strategy is predomi-nantly sensorimotor and ego-oriented (i.e., involves the person'spoint of view): a person represents the environment throughsequences of actions and landmark locations related to his/herposition. Conversely, survey strategy (also called orientation strategyby Lawton, 1994) represents the whole environment, integratinginformation on landmark positions and their relations independentlyof the personal point of view. Survey strategy is prototypicallyrepresented by the use of global reference points, such as the cardinalpoints (north, south, east and west). The use of compass directionrepresents the modality whereby people keep track of their positionin the whole environment using extrinsic frame of reference.Information about the preferential use of route and survey strategiesis usually obtained through self-measures (e.g., Lawton, 1994;Pazzaglia, Cornoldi, & De Beni, 2000): respondents are asked toimagine themselves in common real-life situations (e.g., being in anunfamiliar city or a natural outdoor environment) and to state howthey represent the environment and orient themselves to find theright direction. Preference for using allocentric strategy arises whenpeople say they represent the environment by creating mental mapsand use global reference points (e.g., cardinal points and sunposition); preference for route strategy arises when people say theyrepresent the environment by remembering routes connecting oneplace to another and use egocentric frame of reference (e.g., turns,streets, and landmarks in relation to their own position). Studies showthat the use of orientation strategies is related to gender differences:males prefer survey strategies, and in general report using cardinaldirections, environmental geometry and metric distances; femalesprefer to use route strategy, and in general they report navigating onthe basis of local landmarks and familiar routes (e.g., Coluccia & Louse,2004; Chai & Jacobs, 2009; Halpern, 2000; Lawton, 1994, 2010; Nori,Mercuri, Giusberti, Bensi, & Gambetti, 2009; Wolbers & Hegarty,2010).

The preferential use of these strategies is related to laboratorytasks testing visuo-spatial ability such as the Mental Rotation Test(MRT, Vandenberg & Kuse, 1978). For example, Pazzaglia and De Beni

(2001) found that high-scoring survey participants had betterperformance in MRT than their lower counterparts. In addition,individuals with high MRT ability had higher scores in surveystrategies and performed better in perspective-taking tasks aftermap learning than the lower group (De Beni, Pazzaglia, & Gardini,2006; Pazzaglia & De Beni, 2006). Furthermore the preference forsurvey strategies also holds environment tasks, such as pointing toplaces (i.e., pointing task); indeed, studies found that individuals witha high preference for survey strategy showed greater accuracy inpointing to unseen locations, whereas the use of a route strategy wasunrelated to pointing performance (Lawton, 1996). Recently it hasbeen shown that survey strategies modulate environment learning inrelation to spatial perspective. Pazzaglia and Taylor (2007) identifiedindividuals with high and low preference for survey strategy, wholearned a city path through virtual navigation (i.e., route perspective)or map inspection (i.e., survey perspective). They found thatindividuals with low survey strategies had poorer navigationperformance after learning the environment in survey perspectivethan in route, whereas the high survey group made few navigationerrors in both spatial perspectives learned, proving to be lesssusceptible to the effect of the spatial perspective.

Overall, these studies showed that the predilection for extrinsicframe of reference (preferred by males) is associated with visuo-spatial ability such as the performance of visuo-spatial tasks (such asMRT) and environment tasks such as pointing to places andnavigation. The main aim of our present research is to verify whetherthe preference for extrinsic frame of reference influences environ-mental learning even when spatial information is conveyed throughlanguage. To date there has been no investigation of the relationbetween preference for the use of extrinsic frame of reference andspatial mental representation derived from survey and routedescriptions.

1.3. Aim of the present research

This paper examines whether preference for extrinsic frame ofreference modulates the environment learning using descriptionsexpressed in survey and route perspectives. The use of cardinal pointsfor orientation was assessed through a self-rating scale, the Senseof Direction and Spatial Representation Scale (SDSR Scale, Pazzagliaet al., 2000), which was applied in order to split the sample(composed of males and females) into four groups: high (H-EFR),medium-high (MH-EFR), medium-low (ML-EFR) and low (L-EFR)preference for extrinsic frame of reference. Gender and group wereconsidered as independent variables. All participants performed aseries of visuo-spatial tasks: geographical pointing tasks – placeswithin and outside the city (i.e., Padua) – , MRT (Vandenberg & Kuse,1978), Spatial Indication Task (Nori & Giusberti, 2003), Visual task(Nori & Giusberti, 2003) and a visuo-spatial WM task, the Corsi Blocktask (Corsi, 1972) in forward and backward versions. They alsoperformed two verbal tasks: reading comprehension task (Cornoldi,Rizzo, & Pra Baldi, 1991) and the Digit span (Wechsler, 1981). Nextthey listened to route and survey descriptions, and then performed:(i) a true/false sentences task (categorized as filler, paraphrased, androute and survey inferential) and (ii) a map drawing task.

Hypotheses.

1. Encoding of survey and route descriptionsa. General expectations. It will be assessed whether all partici-

pants construct a spatial mental representation dependent (assuggested by Perrig & Kintsch, 1985) or independent (assuggested by Taylor & Tversky, 1992) of the perspective learnt.Given that previous studies showed that themaintenance of theperspective in spatial mental representation is related toparticular learning conditions (e.g., Brunyé & Taylor, 2008a;

152 C. Meneghetti et al. / Learning and Individual Differences 21 (2011) 150–157

Lee & Tversky, 2005; Noordijz & Postma, 2005), we hypothesizethat mental representation is modulated, at least in part, byspatial perspective. Thus, in the verification task completedafter text learning we expect: (i) better performance ininferential survey questions than in route questions aftersurvey text learning; (ii) better performance in inferentialroute questions than in survey questions after route textlearning. The effect of spatial perspective on map drawingperformance was also explored.

b. Individual differences expectations. We suppose that thepreferences for extrinsic frame of reference can modulate theconstruction of spatial mental representation in the function ofperspectives learnt as suggested by Pazzaglia and Taylor(2007). Thus we expect that when environment is learnt insurvey and route perspectives using descriptions individualswith a high preference for that type of strategy will have betterperformance in spatial recall tasks (verification test and mapdrawing) in comparison with their lower-ability counterparts,appearingmore flexible and perspective-independent. It will beexamined whether the differences are observable betweenextremes of groups (H-EFR vs. L-EFR) or are extendable also tomedium groups (MH-, ML-EFR).Gender differences in favor of males are also expected, aspreviously shown in the verification test by Perrig and Kintsch(1985). This might be extended to map drawing, a taskrequiring participants to reproduce the environment in globallayout and in which males can have better performancecompared with females (e.g., Coluccia & Louse, 2004).

2. Visuo-spatial abilitiesThe performance of visuo-spatial tasks can be subject to spatialindividual and gender differences in favor of individuals with highpreference for extrinsic frame of reference and of males respec-tively. In particular, we suppose that individuals with goodpreference for extrinsic frame of reference will show (relative totheir lower counterparts) a better performance in some spatialtasks, such as MRT (as shown by Pazzaglia & De Beni, 2001, 2006)and in pointing tasks (as shown by De Beni et al., 2006; Lawton,1996). In these spatial tasks gender differences (where malesoutperform females) are also expected, as previously found withMRT (e.g., Linn & Petersen, 1985; Jansen & Heil, 2010) and pointingtasks (e.g., Coluccia & Louse, 2004; Lawton, 1996; 2010; Montello,Lovelance, Golledge, & Self, 1999).Spatial individual differences (in favor of the H-EFR group) can alsobe found in other related factors of SDSR, such as to form a map ofthe environment (as indicated by Pazzaglia et al., 2000; Pazzaglia &De Beni, 2001, 2006). It will be assessed whether spatial andgender differences can be found in the performance of the otherspatial tasks (SIT, Corsi Block task).

Table 1Percentile criteria, number of participants, range of scores, means and standard errors of“knowledge and use of compass directions in orientation tasks” of four spatial groups:low (L-EFR), medium-low (ML-EFR), medium-high (MH-EFR) and high (H-EFR)preference for extrinsic frame of reference groups as a function of females (N=100)and males (N=91).

Percentile Number ofparticipants

Rangeof scores

Means and standarderrors

Females L-EFR ≤25th 20 3 M=3.00 SE=.00ML-EFR 26th–50th 26 4–5 M=4.61 SE=.10MH-EFR 51th–75th 28 6–7 M=6.46 SE=.10H-EFR ≥76th 26 8–13 M=9.85 SE=.29

Males L-EFR ≤25th 26 3–4–5 M=4.11 SE=.14ML-EFR 26th–50th 26 6–7 M=6.38 SE=.14MH-EFR 51th–75th 18 8–9 M=8.44 SE=.17H-EFR ≥76th 21 10–13 M=10.71 SE=.15

2. Method

2.1. Participants

A total of 191 undergraduates (100 females and 91 males) fromthe University of Padua voluntarily participated. Mean age was20.65 years (SD=1.99). The factor “knowledge and use of compassdirections in orientation tasks” of the SDSR Scale (Pazzaglia et al.,2000) was used to select four groups of individuals: high (H-EFR),medium-high (MH-EFR), medium-low (ML-EFR) and low (L-EFR)preference for extrinsic frame of reference. This factor sums the scoresof three items requiring participants to express whether in a naturaloutdoor environment they could spontaneously identify the cardinalpoints (see the three items in Appendix A; Cronbach alpha: .75). Theother four factors of the scale are: general sense of direction, map,route and visual representations (internal reliability: .75: other

psychometric information is reported in Pazzaglia et al., 2000 and inDe Beni et al., 2006). Given that analysis showed gender differences inthe factor used to select groups F (1, 189)=6.65 p≤ .01 η2=.04where males (M=7.18, SD=2.57) had higher ratings than females(M=6.17, SD=2.63), as previously found by Pazzaglia and De Beni(2001, 2006), the four spatial groups were individuated according topercentile criteria calculated separately for males and females in thesample. The percentile ranges and mean scores of males and femalesare reported in Table 1. It should be noted that the number ofparticipants in groups is not completely equal and this is attributed todifferences in distribution (Asymmetry index: Males: .14 ES=.50;Females: .65 ES=.24). The “knowledge and use of compass directionsin orientation tasks” scores differed between four groups in bothfemales F (3, 96)=288.51 1 p≤ .001 η2=.90 and males F (3, 87)=367.42 p≤ .001 η2=.93 (see Table 1).

2.2. Materials

2.2.1. Spatial descriptionsTwopairs of spatial descriptionswere used, adapted fromPazzaglia

et al. (2010), describing a Tourist Center and a Holiday Farm in surveyand route perspective. All four descriptions depicted fourteen land-marks in their respective locations andwere similar in length (numberof words: Tourist Center: survey 309, route 305; Holiday Farm: survey302, route 303). The survey texts initially provide the general structureof the environment and then define the relation between landmarkswithin the environment using canonical terms such as “north,” “south-east,” etc. In the route texts a person imagines walking along a route;the positions of landmarks are defined as the person walks along byegocentric terms such as “left,” “right,” etc.

2.2.2. Verification testThirty-two sentences (half true and half false) were prepared for

each text: 8 filler (describing non-spatial characteristics of land-marks), 8 paraphrased (giving spatial information directly expressedby the text maintaining the same perspective), 8 survey inferentialand 8 route inferential (requiring participants to judge spatialrelations between landmarks not explicitly described in the textexpressed in survey and route perspectives respectively).

2.2.3. Visuo-spatial and verbal measuresMental Rotation Test (MRT, Vandenberg & Kuse, 1978). This task

measures mental rotation ability. It comprises 20 items, eachpresenting one 3D target figure and four possible matches (assembledcubes), the task being to find the two figures identical to the target butrotated in space (time limit for the task 8 min). One point wasassigned when both figures were correctly identified for each item.

Table 2Mean accuracy and standard errors of filler, paraphrased, route and survey inferentialsentences as a function of route and survey text.

Sentence Route text Survey text Total

Filler M 6.79 6.57 6.67SE .07 .08 .05

Paraphrased M 6.50 6.58 6.56SE .09 .10 .08

Route inferential M 5.52 4.61 5.09SE .11 .10 .07

Survey inferential M 4.57 5.20 4.90SE .13 .11 .08

153C. Meneghetti et al. / Learning and Individual Differences 21 (2011) 150–157

Spatial Indication Task (SIT, Nori & Giusberti, 2003). This uses a 2Dmap of a fictitious city on which a route is traced. Participants have tosay out loud the turns taken from start to end points without rotatingthe map (total turns 17). One point was assigned for each spatialdirection given in the correct order.

Visual Task (VT, adapted from Nori & Giusberti, 2003). The taskcontains seven items, each comprising five pictures of houses: onetarget (presented for 3 s) plus four alternatives (one identical to thetarget and three different in some detail). One point was assigned foreach correct alternative selected.

Reading Comprehension Task (RCT, Cornoldi et al., 1991). Thisconsisted of reading an abstract text and answering 10 multiple-choice inferential questions, during which participants were allowedto consult the text; there was no time limitation. One point wasassigned for each correct answer.

Working Memory (WM) measures. The Corsi Block task (Corsi,1972) consists of tapping sequences of blocks arranged irregularly on aboard. The Digit Span task (Wechsler, 1981) consists of sayingsequences of digits. Participants have to reproduce sequences ofblocks/numbers of increasing length, in forward or reverse order. Inbothmeasures the sequence length varied from2 to9 blocks/digits (twosequenceswereused for each length). Thefinal score corresponds to themaximum length of sequences correctly reproduced.

Geographical Pointing Task (GPT). Participants had to point to thedirection of three separate locations within the city of Padua (trainstation, “Prato della valle” square, and the “Caffè Pedrocchi”) and alsotowards three Italian cities (Milan, Venice and Bologna). The directionof the target location was indicated on a circle, with participantsdrawing a line from the center to the circumference. The absoluteangular degree error was calculated for each item given by differencebetween correct angle and angle indicated by participants.

2.2.4. ProcedureThe tasks and texts were administered in two sessions. In the first,

participants were tested in groups (20 to 25 in each one) and thefollowing spatial and verbal tasks were administered: SDSR scale,MRT, VT, RCT and GPT (in that order; this guaranteed that self-evaluations of the SDSR scale were not influenced by the performanceof objective tasks). The second session took place a few days later andparticipants were individually tested; participants first performed theSIT and WM measures. Then, assigned to one of the four possiblecombinations of perspective and environment presented, theylistened twice to one survey and one route description and thenperformed: (i) verification test using the E-prime 1 program(Schneider, Eschman, & Zuccolotto, 2002); sentences were presentedrandomly and remained on the screen until participants pressed key 1on the keyboard for “true” or key 2 for “false”; the next sentence waspresented after an interval of 1.5 s. Correct/incorrect responses andresponse times were recorded; (ii) map drawing, making a sketch ofthe environment on a sheet of paper.

3. Results

3.1. Encoding of survey and route descriptions

3.1.1. Verification test

Accuracy. The accuracy was scored by averaging the number of correcttrue/false answers. A 4×2×2×4 analysis of variance (ANOVA) wasperformed, with group (L-EFR vs. ML-EFR vs. MH-EFR vs. H-EFR) andgender (males vs. females) as between-participants factors, perspec-tive (survey vs. route) and sentence (filler vs. paraphrased vs. surveyinferential vs. route inferential) as within-participant factors. Thedifference between means was considered significant when p was≤.01.

a. General results. The significantmain effect of sentence F (3, 549)=209.99 p≤ .001 η2=.53 was found. Post hoc comparisons (withBonferroni correction) showed in general better responses in filler andparaphrased sentences (which did not differ from each other) than forsurvey and route inferential sentences (which did not differ from eachother) with psb .001 (see Table 2). The interaction perspective×sentenceF (3, 549)=23.54 p≤ .001 η2=.12 was significant and it is imputed todifferences in inferential sentences between the two texts (psb .001): inroute text the inferential route sentences (M=5.52 SE=.11) weremore accurate than survey ones (M=4.57 SE=.13); in survey text theresults were inverted: the inferential survey sentences (M=5.20SE=.11) were better in comparison with route ones (M=4.61SE=.10; see Table 2). These results confirmed the hypothesis that forthewhole samplemental representation is dependent on the perspectivelearnt.

b. Individual differences results. The main effects of gender F (1,183)=4.85 p=.03 η2=.03 and of group F (3, 183)=7.68 p≤ .001η2=.11 were significant; the means showed that males (M=5.90SE=.07) outperformed females (M=5.61 SE=.06); post hoccomparisons showed that the H-EFR group had a better performancein comparison with all three groups (MH-EFR p=.02- ML-EFR and L-EFR ps≤ .001; see Table 3) which did not differ from each other.

The interaction group×sentence F (9, 549)=4.53 p≤ .001 η2=.07was significant. Post hoc comparisons showed that the groups differedin inferential route (p≤ .01) and survey (p≤ .001) sentences – but notin filler and paraphrased sentences: in route sentences the H-EFRgroup had better performance relative to L and ML-EFR – which didnot differ from each other (ps≤ .01); the MH-EFR group did not differfrom the other three groups. In survey sentences the H-EFR group hadbetter performance in comparison with L-, ML- and MH-EFR groups,which did not differ from each other (ps≤ .001; see Table 3). Theseresults revealed that individuals with H-EFR had a superior ability totheir counterparts in answering route and survey inferential sen-tences, showing that they build a mental representation that is moreperspective-independent.

The interaction gender×text×sentencewasmarginally significantF (3, 549)=2.55 p=.05 η2=.02. The post hoc comparisons showedgender differences only in survey inferential sentences of route text(p≤ .01) where males (M=4.99 SE=.18) outperformed females(M=4.19 SE=.17); for the other sentences no significant differenceswere found.

Response times. The average number of syllables/response times ofcorrect sentences for each category (filler, paraphrased, inferentialsurvey and inferential route) was calculated (in ms).

a. General results. The results of the 4×2×2×4 ANOVA showedsignificantmain effect of perspective F (1, 183)=26.63 p≤ .001 η2=.13and sentence, F (3, 549)=710.10 p≤ .001 η2=.80; the means showedthat the answers for route text were rather faster than for survey text(see Table 4); post hoc comparisons showed that all sentences differedfrom each other (with ps≤ .001): the filler sentences were the shortest,followed by the paraphrased, then by the route inferential and thelongest were the survey inferential sentences (see Table 4). The

Table 3Mean accuracy and standard errors of filler, paraphrased, route and survey inferentialsentences as a function of low (L-EFR), medium-low (ML-EFR), medium-high (MH-EFR) and high (H-EFR) preference for extrinsic frame of reference groups.

Sentence type L-EFR ML-EFR MH-EFR H-EFR

Filler M 6.61 6.72 6.64 6.64SE .11 .10 .11 .11

Paraphrased M 6.38 6.37 6.67 6.76SE .15 .14 .15 .15

Route inferential M 4.82 4.76 5.14 5.47SE .15 .14 .15 .15

Survey inferential M 4.57 4.59 4.55 5.72SE .17 .16 .17 .17

Total M 5.65 5.64 5.78 6.16SE .09 .08 .09 .09

Table 5Mean and standard errors of map drawing accuracy of low (L-EFR), medium-low (ML-EFR), medium-high (MH-EFR) and high (H-EFR) preference for extrinsic frame ofreference groups.

Text L-EFR ML-EFR MH-EFR H-EFR

Route M 9.02 9.39 9.85 9.52SE .53 .50 .56 .52

Survey M 7.85 8.43 9.86 11.32SE .46 .43 .49 .45

Total M 8.43 8.91 9.85 10.42SE .39 .37 .41 .38

154 C. Meneghetti et al. / Learning and Individual Differences 21 (2011) 150–157

interaction perspective×sentence F (3, 549)=36.57 p≤ .001 η2=.17was significant. Post hoc comparisons showed that the differencesbetween the two spatial texts in this case were found in routeparaphrased and route inferential sentences: the answers to thosequestions were faster in route than in survey text (with ps≤ .001; seeTable 4); the performance of filler and survey inferential sentences didnot differ between route and survey text. b. Individual differencesresults: the main effect of group, the main effect of gender andinteractions involving these variables were not significant).

3.1.2. Map drawingOne point was assigned when participants correctly wrote/drew a

landmark position on the map. The protocols were scored by twoindependent judges and proved to be strongly correlated (r=.96p≤ .001). Analyses were carried out on the scores assigned by the firstjudge (the experimenter).Three scores of males (one of ML and two ofMH-EFR groups) were not included in the analysis because anylandmarks were correctly identified (score=0). a. The general result(i.e., main effect of perspective) was not found to be significant.

b. Individual differences results. The 4 (group: L-EFR vs. ML-EFR vs.MH-EFR vs. H-EFR)×2 (gender: males vs. females)×2 (perspective:survey vs. route) ANOVA showed significant main effects of genderF (1, 180)=4.77 p=.03 η2=.05 and of group F (3, 180)=5.39 p≤ .01η2=.09. The means showed that males (M=9.83 SE=.29) out-performed females (M=8.98 SE=.27); post hoc comparisonsshowed that the H- EFR group had better performance relative to L-(p≤ .01) and ML- (p=.03) EFR groups; the MH-EFR group did notdiffer from the other three groups (see Table 5).

The interaction group×perspective F (3, 180)=5.08 p≤ .01η2=.08 was found to be significant. The post hoc comparisonsshowed that spatial groups differed in survey (but not in route) text:the H-EFR performed better than the MH-EFR (p≤ .01) and these twogroups had better performance in comparison with ML- and L-EFRgroups (ps≤ .01, see Table 5). These results indicated that in mapdrawing the differences between spatial groups (in favor of H-EFR)were detected only in survey text, showing that individuals with high

Table 4Mean and standard errors of response times (in ms) of filler, paraphrased, route andsurvey inferential sentences as a function of route and survey text.

Sentence type Route Text Survey Text Total

Filler M 172.96 175.52 174.24SE 2.90 3.01 2.56

Paraphrased M 235.76 286.76 261.26SE 3.92 5.43 3.91

Route inferential M 291.91 318.79 305.35SE 4.57 5.14 4.16

Survey inferential M 375.84 361.04 368.44SE 7.85 6.28 6.26

Total M 269.12 285.53SE 3.95 3.82

preference for extrinsic frame of reference are more able to expressspatial information in survey format.

3.2. Visuo-spatial abilities

A multivariate ANOVA was carried out for SDSR dimensions(except “knowledge and use of compass directions in orientationtasks,” the factor used to select the four groups), visuo-spatial (MRT,SIT, VT, and Corsi Block task), verbal tasks (RCT and Digit Span) andpointing tasks (GPT—places within and outside the city) as dependentfactors and gender and group as independent factors. The resultsshowed the significant multivariate effects of group F (42, 516)=2.08p≤ .001 η2=.15 and gender F (14, 170)=7.38 p≤ .001 η2=.38. Inparticular, the univariate ANOVAs showed group differences in thefollowing variables: MRT, GPT – places within and outside the city – ,general sense of direction and map representation. On the MRT(accuracy), GPT – places outside the city – (degree errors) and senseof direction, post hoc comparisons showed that the H- and MH-EFRgroups (which did not differ from each other) had better performancein comparison with ML- and L-EFR groups (which did not differ fromeach other in GPT and MRT, but did differ in sense of direction, whereL- had lower scores thanML-EFR). On GPT – placeswithin the city –H-and MH-EFR groups (which did not differ from each other) had betterperformance in comparisonwith the L- EFR group and only H-EFR hadbetter performance in comparison with ML-EFR. On map represen-tation the H-EFR group had higher scores than the L- and ML-EFRgroups (which did not differ from each other) and MH-EFRwas betterthan L-EFR (Means, standard errors of H-, MH-,ML- andH-EFR groups,F, p and η2 values are reported in Table 6).

The univariate ANOVAs showed gender differences (where maleshad higher scores than females) in the following variables: MRT, SIT,GPT – places within and outside the city – , general sense of direction,map representation, and Corsi Block task in backward version (means,standard errors ofmales and females, F, p and η2 values are reported inTable 7).

4. Discussion of results and conclusions

The aim of the present study was to investigate whether spatialmental representation derived from environment descriptions inroute and survey perspective was modulated by individual genderand spatial differences. The theoretical premises were: (i) studiesshowed that mental representation derived from survey and routetexts could be independent of (Taylor & Tversky, 1992) or dependenton (Perrig & Kintsch, 1995) the perspective learnt; (ii) to disambig-uate these results recent studies analyzed the role of other factors thatpotentially modulated the effect of perspective (e.g., Brunyé & Taylor,2008a; Lee & Tversky, 2005; Noordijz & Postma, 2005); (iii) one of thecentral factors was spatial ability. The latter was taken into account inthe present research and we particularly analyzed individualdifferences in the preference for extrinsic frame of reference thatproved to be related to spatial task performance (e.g., De Beni et al.,2006; Lawton, 1996; Pazzaglia & De Beni, 2001). Four spatial groups ofmales and females were individuated according to their scores for

Table 6Mean, standard errors and F, p, η2 values of self-evaluations in SDSR scale (general sense of direction and map representation), MRT, GPT (places within and outside the city) of low(L-EFR), medium-low (ML-EFR), medium-high (MH-EFR) and high (H-EFR) preference for extrinsic frame of reference groups.

L-EFR ML-EFR MH-EFR H-EFR F, p, η2

General sense of direction M 10.77 12.46 13.33 14.47 F (3, 183)=10.82 p≤ .001 η2=.15(H –EFR=MH-EFR)N(L-EFRbML-EFR) ps from ≤.01 to ≤.001

SE .48 .46 .49 .47

Map representation M 4.99 5.40 6.04 6.65 F (3, 183)=8.90 p≤ .001 η2=.13H –EFRN(L-EFR=ML-EFR) ps≤ .001MH –EFRNL-EFR

SE .25 .23 .25 .24

GPT (places within the city) M 45.23 36.78 34.25 26.39 F (3, 183)=4.90 p≤ .01 η2=.07(H –EFR=MH-EFR)N(L-EFR) with p≤ .001 and p=.03H –EFRNML-EFR p=.03

SE 3.53 3.29 3.59 3.48

GPT (places outside the city) M 43.84 37.07 24.14 23.65 F (3, 183)=9.28 p≤ .001 η2=.13(H –EFR=MH-EFR)N(L-EFR=ML-EFR) ps≤ .001

SE 3.29 3.07 3.34 3.24

MRT M 7.10 7.23 8.77 8.72 F (3, 183)=2.73 p≤ .05 η2=.04(H –EFR=MH-EFR)N(L-EFR=ML-EFR) ps≤ .05

SE .57 .53 .57 .56

155C. Meneghetti et al. / Learning and Individual Differences 21 (2011) 150–157

“knowledge and use of compass directions in orientation tasks” on theSDSR Scale (Pazzaglia et al., 2000). First, we investigated the role ofpreference for extrinsic frame of reference in environment learningusing spatial descriptions in survey and route perspectives. Second,we analyzed whether that preference was sustained by a spatialprofile through analysis of visuo-spatial and verbal task performance.

4.1. Encoding of survey and route descriptions

We analyzed the effect of spatial perspective in the whole sample(a. general results) and in relation to individual differences inextrinsic frame of reference (b. individual differences results). Giventhat previous studies showed that mental representation wasinfluenced, at least in part, by spatial perspective (e.g., Brunyé &Taylor, 2008a; Lee & Tversky, 2005; Noordijz & Postma, 2005), for thewhole sample of the present research we also expected to find aninfluence of spatial perspective. We supposed, however, that strongpreference for an extrinsic frame of reference allowed the construc-tion of a mental representation more flexible and less influenced byspatial perspective (as previously suggested by Pazzaglia & Taylor,2007).

Table 7Mean, standard errors and F, p, η2 values of self-evaluations in SDSR scale (generalsense of direction and map representation), MRT, GPT (places within and outside thecity) of male and female groups.

Males Females F, p, η2

Generalsense of direction

M 13.66 11.86 F (1, 183)=14.78 p≤ .001 η2=.08

SE .34 .32Map representation M 6.10 5.44 F (1, 183)=7.38 p≤ .01 η2=.04

SE .18 .17GPT (places withinthe city)

M 27.36 43.97 F (1, 183)=22.86 p≤ .001 η2=.11

SE 2.57 2.39GPT (places outsidethe city)

M 25.95 38.40 F (1, 183)=14.79 p≤ .001 η2=.08

SE 2.35 2.23MRT M 9.65 6.27 F (1, 183)=36.92 p≤ .001 η2=.17

SE .40 .38SIT M 13.96 10.07 F (1, 183)=43.25 p≤ .001 η2=.19

SE .43 .41Corsi Block task(backward version)

M 5.93 5.49 F (1, 183)=6.68 p=.01 η2=.04

SE .12 .12

(a) The general findings sustained the hypothesis that mentalrepresentation derived from survey and route descriptions wasperspective-dependent, supporting Perrig and Kintsch's (1985) study.Indeed the results of the verification test showed that after route textlearning participants were more accurate when answering routeinferential questions than survey questions; conversely, after surveytext learning they were more accurate in answering survey inferentialquestions than route questions. The response times showed, however,that route sentences – both paraphrased and inferential – wereanswered more rapidly in route text than in survey text. Furthermore,gender modulated the effect of spatial perspective. In fact, resultsshowed that males performed better than females in survey sentenceswhen text was encoded in route perspective. This finding corroboratedthe results found by Perrig and Kintsch (1985)wherebymales created amental representationwithmap-like characteristics,whereas in femalesmental representation was more dependent on the perspective learnt.

(b) The analysis of individual differences showed general effects(for both recall measures) in spatial and gender factors in favor of H-EFR individuals and males respectively. These results revealed thatspatial knowledge of environment acquired through verbal input wassubject to individual gender and spatial differences; this findingextends the results found for spatial tasks (e.g., Pazzaglia & De Beni,2001; De Beni et al., 2006) or derived from environment learningacquired through visual input (e.g., De Beni et al., 2006; Pazzaglia & DeBeni, 2006; Lawton, 1996).

As expected, however, analysis of the preference for extrinsic frameof reference introduced changes in the general results. Some differencesbetween recall tasks were found. The findings for map drawing showedthat the group with a strong preference for extrinsic frame of referencefound it easier than did its less able counterparts to express spatialinformation learnt in survey format. Indeed, the results showed that theH-EFR individuals drew maps more accurately than the lower-abilitygroups (MH-, ML- and L-EFR) in survey text (but not in route text).Nevertheless, the individuals with high spatial competence appeared tobe able to handle information expressed in both spatial perspectives, asshown by verification tests; more specifically, in the latter measure theH-EFR group was better than its less able counterparts at answeringspatial inferential questions both in survey and in route perspectives.Overall these results suggested thatwhen environmentwas learnt usingspatial language – and not only when it was learnt using visual input(Pazzaglia & Taylor, 2007) – individuals with a high preference forextrinsic frame of reference constructed a perspective-independentmental representation (as shown by verification tests). When they hadto reproduce spatial information previously learnt, however, theyshowed the ability to organize it from a survey perspective (as shownby map drawing).

156 C. Meneghetti et al. / Learning and Individual Differences 21 (2011) 150–157

4.2. Visuo-spatial abilities

The profile emerging above was completed by analysis ofperformance in visuo-spatial tasks. We expected to find individualgender and spatial differences in typical visuo-spatial tasks (such asMRT) and environment tasks (such as pointing tasks) as shown in theliterature (e.g., De Beni et al., 2006; Pazzaglia & De Beni, 2001, 2006;Linn & Petersen, 1985; Coluccia & Louse, 2004).

As expected, the individuals with a high preference for extrinsicframe of reference (H- and MH-EFR groups) gave a superiorperformance (in comparison with L- and ML-EFR groups) in MRT andin the geographical pointing tasks of unseen landmarks within andoutside the city. These results confirmed that individuals who preferredto adopt extrinsic frame of reference had better perspective-takingability and performed better in the MRT (e.g., De Beni et al., 2006;Pazzaglia & De Beni, 2001, 2006). In these tasks gender differences infavor of males were shown too, confirming previous results found bothwith mental rotation (e.g., Linn & Petersen, 1985; Jansen & Heil, 2010)and with pointing tasks (e.g., Coluccia & Louse, 2004; Lawton, 1996,2010; Montello et al., 1999). Furthermore, spatial (in favor of the H-EFRgroup) and gender (in favor of males) differences in SDSR scale factors(e.g., Pazzaglia et al., 2000) related to knowledge and use of cardinalpoints such as to represent the environment usingmaps and also in thesense of direction (e.g., Pazzaglia & De Beni, 2001, 2006) were found.

Gender differences (but not spatial differences) in favor of maleswere found in the Corsi Block task (backward version) and SIT. TheCorsi Block task is a task measuring visuo-spatial WM and thebackward version is a task more high-demanding in comparison toforward version: according Cornoldi and Vecchi (2003) model thebackward version involves more active spatial resources than theforward version and it is subject to gender differences (e.g., Vecchi &Girelli, 1998). Our results, consistently confirmed a superior perfor-mance of males in the backward version of the Corsi Block task. Forthe SIT the gender differences could be explained by the fact that, asfor MRT, rotation abilities are required. Indeed, in this task of givingcorrect right/left directions participants need to rotate their ownperspective to that of their street map. Furthermore, gender andgroup differences did not affect performance in visual and verbal tasksas previously found (e.g., Blakemore, Berenbaum, & Liben, 2009;Cornoldi & Vecchi, 2003; Hyde & Linn, 1988; Kimura, 1999).

In summary, the present study showed that the mental represen-tation derived from route and survey descriptions is perspective-dependent. Individual differences in gender and in spatial compe-tence, however, influenced the ability to process spatial informationin route and survey perspectives. In fact, when individuals with a highpreference for extrinsic frame of reference learnt environmentinformation using verbal input and were forced to use survey androute information (as in the verification test) they appeared to bemore able than the other groups with a less strong preference (andthis wasmore pronounced inmales). Moreover, when the same groupof individuals reproduced in a map the spatial information learnt (asin map drawing) the performance was facilitated when the text waslearnt in the survey perspective. This modality for processing spatialinformation is supported by a profile that is specifically spatial.

Acknowledgments

The authors wish to thank Marco Agrioli, Francesca Del Maschio,Angela Mason and Elisabetta Polloniato for collecting the data.

Appendix A

Items testing “knowledge and use of compass directions” from theSDSR scale (Pazzaglia et al., 2000), used to split the sample into low

(L-EFR), medium-low (ML-EFR), medium-high (MH-EFR) and high(H-EFR) preference for extrinsic frame of reference groups.

5. When you are in a natural, open environment (mountains, seaside,and country), can you naturally identify the cardinal points, i.e.where north, south, east and west are?

1

6. When you are in a town, can you naturally identify the cardinalpoints, i.e. do you find it easy to find where north, south, east andwest are?

1

10. When you are outdoors and have to point out a compass direction(north-south-east-west), do you (circle one)?

a. point immediatelyb. need to think before pointingc. have difficulty

References

Blakemore, J. E. O., Berenbaum, S. A., & Liben, L. S. (2009). Gender development. NewYork, NY: Psychology Press.

Bosco, A., Filomena, S., Sardone, L., Scalisi, T. G., & Longoni, A. M. (1996). Spatial modelsderived from verbal descriptions of fictitious environments: The influence of studytime and the individual differences in visuo-spatial ability. Psychology Beitrage, 38,451−464.

Brunyé, T. T., & Taylor, H. A. (2008a). Extended experience benefits spatial mentalmodel development with route but not survey descriptions. Acta Psychologica, 127,340−354.

Brunyé, T. T., & Taylor, H. A. (2008b). Working memory in developing and applyingmental models from spatial descriptions. Journal of Memory and Language, 58,701−729.

Chabanne, V., Péruch, P., Denis, M., & Thinus-Blanc, C. (2004). Mental scanning ofimages constructed from visual experience or verbal descriptions: The impact ofsurvey and route perspective. Imagination, Cognition and Personality, 23, 163−171.

Chai, X. J., & Jacobs, L. F. (2009). Sex differences in directional cue use in a virtuallandscape. Behavioral Neuroscience, 123, 276−283.

Coluccia, E., & Louse, G. (2004). Gender differences in spatial orientation: A review.Journal of Environmental Psychology, 24, 329−340.

Cornoldi, C., Rizzo, A., & Pra Baldi, A. (1991). Prove avanzate MT di comprensione dellalettura [Advanced MT reading comprehension task]. Florence: OrganizzazioniSpeciali.

Cornoldi, C., & Vecchi, T. (2003). Visuo-spatial working memory and individualdifferences. Philadelphia, PA: Psychology Press.

Corsi, P. M. (1972). Human memory and the medial temporal region of the brain [Doctoraldissertation]. McGill University, Montreal.

Darken, R., & Peterson, B. (2002). Spatial orientation, wayfinding and representation. InK. Stanney (Ed.), Handbook of virtual environments: Design, implementation andapplications (pp. 493−518). Mahwah, NJ: Lawrence Erlbaum.

De Beni, R., Pazzaglia, F., & Gardini, S. (2006). The role of mental rotation and age inspatial perspective-taking tasks: When age does not impair perspective-takingperformance. Applied Cognitive Psychology, 20, 807−821.

Deyzac, E., Logie, R. H., & Denis, M. (2006). Visuospatial working memory and theprocessing of spatial descriptions. British Journal of Psychology, 97, 217−243.

Gyselinck, V., De Beni, R., Pazzaglia, F., Meneghetti, C., & Mondoloni, A. (2007). Workingmemory components and imagery instructions in the elaboration of a spatialmental model. Psychological Research, 71, 373−382.

Hyde, J. S., & Linn, M. C. (1988). Gender differences in verbal ability: A meta-analysis.Psychological Bulletin, 104, 53−69.

Halpern, D. F. (2000). Sex differences in cognitive abilities. Hillsdale, NJ: Erlbaum.Jansen, P., & Heil, M. (2010). Gender differences in mental rotation across adulthood.

Experimental Aging Research, 36, 94−104.Kimura, D. (1999). Sex and cognition. Cambridge, MA: MIT Press.Lawton, C. A. (1994). Gender differences in way-finding strategies: relationship to

spatial ability and spatial anxiety. Sex Roles, 30, 765−779.Lawton, C. A. (1996). Strategies for indoor wayfinding: the role of orientation. Journal of

Environmental Psychology, 16, 137−145.Lawton, C. A. (2010). Gender, spatial abilities, and wayfinding. In J. C. Chrisler, & D.

McCreary (Eds.), Handbook of gender research in psychology, Vol 1: Gender researchin general and experimental psychology (pp. 317−341). US: Springer.

Lee, P. U., & Tversky, B. (2005). Interplay between visual and spatial: The effect oflandmark descriptions on comprehension of route/survey spatial descriptions.Spatial Cognition and Computation, 5, 163−185.

157C. Meneghetti et al. / Learning and Individual Differences 21 (2011) 150–157

Linn, M. C., & Petersen, A. C. (1985). Emergence and characterization of genderdifferences in spatial abilities: A meta-analysis. Child Development, 56, 1479−1498.

Meneghetti, C., Gyselinck, V., Pazzaglia, F., & De Beni (2009). Individual differences inspatial text processing: High spatial ability can compensate for spatial workingmemory interference. Learning and Individual Differences, 19, 577−589.

Montello, D. R., Lovelance, K. L., Golledge, R. G., & Self, C. M. (1999). Sex-relateddifferences and similarities in geographic and environmental spatial abilities.Annals of the Association of American Geographers, 89, 515−534.

Noordzij, M. L., & Postma, A. (2005). Categorical and metric distance information inmental representations derived from route and survey descriptions. PsychologicalResearch, 69, 221−232.

Noordzij, M. L., Van der Lubbe, R. H. J., & Postma, A. (2005). Strategic and automaticcomponents in the processing of linguistic spatial relations. Acta Psychologica, 119,1−20.

Noordzij, M. L., Van der Lubbe, R. H. J., & Postma, A. (2006). Electrophysiological supportfor strategic processing of spatial sentences. Psychophysiology, 43, 277−286.

Noordzij, M. L., Zuidhoek, S., & Postma, A. (2006). The influence of visual experience onthe ability to form spatial mental models based on route and survey descriptions.Cognition, 100, 321−342.

Nori, R., & Giusberti, F. (2003). Cognitive styles: Errors in directional judgments.Perception, 32, 307−320.

Nori, R., Mercuri, N., Giusberti, F., Bensi, L., & Gambetti, E. (2009). Influences of genderrole socialization and anxiety on spatial cognitive style. American Journal ofPsychology, 4, 497−505.

Pazzaglia, F. (2008). Text and picture integration in comprehending and memorizingspatial descriptions. In J. F. Rouet, & R. K. Lowe (Eds.), Understanding multimediadocuments (pp. 43−59). New York, NY: Springer-Verlag.

Pazzaglia, F., & Cornoldi, C. (1999). The role of distinct components of visuo-spatialworking memory in the processing of texts. Memory, 7, 19−41.

Pazzaglia, F., Cornoldi, C., & De Beni, R. (2000). Differenze individuali nellarappresentazione dello spazio: presentazione di un questionario autovalutativo[Individual differences in representation of space: presentation of questionnaire].Giornale Italiano di Psicologia, 3, 627−650.

Pazzaglia, F., Cornoldi, C., & Longoni, A. M. (1994). Limiti di memoria e specificità dirappresentazione nel ricordo di descrizioni spaziali “dall'alto” ed “entro il percorso”

[Limits of memory and the specificity of representations in the memory of surveyand route descriptions]. Giornale Italiano di Psicologia, 21, 267−286.

Pazzaglia, F., &DeBeni, R. (2001). Strategies of processing spatial information in survey andlandmark-centred individuals. European Journal of Cognitive Psychology, 13, 493−508.

Pazzaglia, F., & De Beni, R. (2006). Are people with high and low mental rotationabilities differently susceptible to the alignment effect? Perception, 35, 369−383.

Pazzaglia, F., Meneghetti, C., De Beni, R., & Gyselinck, V. (2010). Working memorycomponents in survey and route spatial text processing. Cognitive Processing, 11,359−369.

Pazzaglia, F., & Taylor, H. A. (2007). Perspective, instruction, and cognitive style inspatial representation of a virtual environment. Spatial Cognition and Computation:An Interdisciplinary Journal, 7, 349−364.

Perrig, W., & Kintsch, W. (1985). Propositional and situational representations of text.Journal of Memory and Language, 24, 503−518.

Schneider, W., Eschman, A., & Zuccolotto, A. (2002). E-Prime User's Guide. Pittsburgh:Psychology Software Tools Inc..

Siegel, A.W., &White, S. H. (1975). The development of spatial representations of large-scale environments. In H. W. Reese (Ed.), Advances in child development andbehavior, Vol. 10. (pp. 9−55)New York, NY: Academic Press.

Shelton, A. L., & McNamara, T. P. (2004). Orientation and perspective dependence inroute and survey learning. Journal of Experimental Psychology: Learning, Memory,and Cognition, 30, 158−170.

Taylor, H. A., Naylor, S. J., & Chechile, N. A. (1999). Goal-specific influences on therepresentation of spatial perspective. Memory and Cognition, 27, 309−319.

Taylor, H. A., & Tversky, B. (1992). Spatial mental models derived from survey and routedescriptions. Journal of Memory and Language, 31, 261−292.

Vandenberg, S. G., & Kuse, A. R. (1978). Mental rotation, a group test of three-dimensional spatial visualization. Perceptual and Motor Skills, 47, 599−604.

Vecchi, T., & Girelli, L. (1998). Gender differences in visuo-spatial processing: Theimportance of distinguishing between passive storage and active manipulation.Acta Psychologica, 99, 1−16.

Wechsler, D. (1981). Wechsler adult intelligence scale–revised (WAIS-R). New York, NY:The Psychological Corporation.

Wolbers, T., & Hegarty, M. (2010).What determines our navigational abilities? Trends inCognitive Sciences, 14, 138−146.