0% found this document useful (0 votes)

76 views49 pages

Individual Differences in Spatial Abilities

Uploaded by

Jorge Garcia

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

76 views49 pages

Individual Differences in Spatial Abilities

Uploaded by

Jorge Garcia

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 49

4

Individual Differences in Spatial Abilities

Mary Hegarty and David A. Waller

abstract
This chapter reviews research on spatial abilities, which is concerned with individ-
ual differences in how people mentally represent and manipulate spatial informa-
tion to perform cognitive tasks. We first review factor analytic studies of spatial
abilities. This research tradition provided strong evidence that spatial ability is
differentiated from general intelligence and that it is not a single, undifferentiated
construct, but instead is composed of several somewhat separate abilities. We next
review analyses of performance on spatial abilities tests by cognitive psychologists,
which has shown that different spatial abilities may depend more or less on speed
of processing, strategies, quality of spatial images, active maintenance of spatial
information, and central executive processes. Third, we examine individual dif-
ferences in large-scale or environmental spatial abilities such as wayfinding and
navigation. Research on this topic has begun to characterize the factor structure of
large-scale spatial abilities and these abilities’ relation to more traditional measures
of spatial abilities. Finally, we consider some of the functions of spatial ability in
occupational and academic performance, including surgery, mechanical reasoning,
and mathematical problem solving.

introduction
The ability to represent and process spatial information is important for
many common activities, such as finding our way to and from places in
the environment, moving furniture, packing a suitcase, and catching a ball.
It is also related to skilled performance in many occupations, such as car
mechanic, airline pilot, and surgeon. It has been linked to academic suc-
cess in mathematics and science. We are aware that people differ in spatial
abilities. We know that some people have a better sense of direction than

121

others, that not everyone is cut out to be a pilot, and that only some people
are able to visualize the complex spatial relations between atoms in an or-
ganic compound. Do all of these activities rely on a single spatial ability
or do they depend on different types of spatial ability? What are the es-
sential differences in spatial representations and processes between people
with more and less spatial ability? In this chapter we explore the different
dimensions of spatial abilities, the cognitive processes that underlie these
different abilities, and the functions of spatial abilities in everyday life and
professional activities.
This chapter is different from most other chapters in this volume in
that it explores the natural variation in performance among individuals,
rather than variation due to experimental treatments. As Cronbach (1957)
pointed out, experimental psychology and differential psychology (or cor-
relational psychology) have been separate areas of research in the history
of psychology. Because most research in cognitive psychology is exper-
imental, it is worth considering why research on individual differences
is important. First, a complete account of spatial cognition must account
for the variance among people, as well as the central tendency, especially
because individual differences in spatial abilities are so pervasive. Second,
from a practical or applied perspective, understanding cognitive differ-
ences among people can inform decisions about whom to select or recruit
for a job or training program. Third, by capitalizing on the natural varia-
tion among people, the study of individual differences can be an important
strategy in basic research on spatial cognition. For example, if we are in-
terested in the extent to which spatial representations and processes are
important for a particular task, such as solving a mathematics problem,
we can use the natural variation among individuals to identify those with
more or less spatial ability and observe how they solve the problem. Unlike
experimental methods, which tend to create discrete differences between
experimental conditions, individual differences in ability vary continu-
ously across the population and therefore allow us more precision in ex-
amining the relation between some spatial process and a criterion. We can
also find natural variation in aspects of spatial cognition that are difficult
to control experimentally, such as the speed of mental rotation. Finally,
differences among individuals on some tasks might be large enough to
make the effects of other variables (perhaps those manipulated by an ex-
perimentalist) difficult to detect. Research on individual differences can
thus inform investigators about how to examine the effects that interest
them by factoring out effects due to individual differences. A particu-
larly powerful design is to combine the use of experimental methods and

differential methods in the same study, in order to examine aptitude-

treatment interactions (Cronbach, 1957).
In this chapter, we will review research on individual differences in
spatial cognition from a somewhat historical perspective. We first review
the factor analysis literature, which dominated early research in spatial
abilities. The goals of this research movement were to develop measures
of spatial ability, to establish its separability from general intelligence, and
to examine the extent to which it is made up of different abilities. Next we
consider research on the analysis of spatial abilities from the perspective of
cognitive psychology. This research movement attempted to discover the
basic perceptual and cognitive processes that differentiate people with high
and low ability. Third, we examine individual differences in large-scale or
environmental spatial abilities such as wayfinding and navigation. Until
recently, this type of spatial ability has received relatively little research
attention. Indeed, most of the individual differences literature has been
dominated by analyses of paper-and-pencil tests, which measure spatial
abilities at a smaller scale. Finally, we consider some of the functions of
spatial ability in occupational and academic performance.
In this chapter, we will be concerned primarily with the measurement,
classification, and characterization of spatial abilities. We will not discuss
the underlying causes of differences in spatial ability, for example, the ex-
tent to which they can be trained or are based on innate factors. These issues
are dealt with in some detail in the chapter on sex differences (Halpern &
Collaer, Chapter 5).

the factor-analytic approach to visuospatial abilities

The scientific study of individual differences in visuospatial abilities began
in the early 20th century as a result of widespread interest in predicting
academic and vocational success through the use of standardized tests.
Spatial testing emerged from early attempts to measure mechanical abil-
ity, typically in order to predict success in technical occupations. Because
these abilities were generally assumed to require physical manipulation,
many of these tests were based on performance on tasks that required as-
sembly or manipulation of actual objects. Smith (1964) describes some of
these early tests as consisting of “peg-boards, eye-boards, tests of tapping
and aiming, tests of tweezer-dexterity, of wire-bending, tests involving the
manipulation of nuts and bolts, and many others” (p. 41). As instruments
such as Binet and Simon’s Scales of Intelligence (1905) grew in popular-
ity, ability tests began to shift their emphasis from physical manipulation

to quantitatively scored multiple-choice questions that could be answered

by means of paper-and-pencil tests. Paper-and-pencil assessments enabled
testing in large groups, and as a result large data sets emerged that slowly
began to clarify the structure of intelligence and the nature of visuospatial
abilities (for a summary, see Smith, 1964). It took several decades to de-
velop a consistent and powerful means of analysis as well as to overcome
the prevailing bias to associate all aspects of intellectual ability with a sin-
gle, general (typically verbal) intelligence. However, by 1939, Kelly (1928),
Koussy (1935), Eysenck (1939), and Thurstone (1938) had all shown that
spatial ability was distinct from verbal ability. These researchers made a
primary distinction between verbal and spatial intelligence and noted that
the latter accounted for a significant amount of variance in scores on large
batteries of intellectual functioning. This conclusion is no longer disputed.
One of the most common methods used to describe the underlying struc-
ture of the intellect is called factor analysis. Factor analysis is a statistical
technique that examines the patterns of correlations among a large num-
ber of variables (e.g., scores on many different paper-and-pencil tests). The
technique attempts to reduce the information contained in many variables
to a smaller number of constructs called factors. Mathematically, a factor
is simply a weighted sum of each of the variables. Yet factors are typically
conceptualized as representing an underlying ability. Tests that correlate
highly with a given factor may be called markers for a factor and are said
to load on that factor. Although generally helpful and very widely used,
the factor-analytic approach has several limitations. For example, evidence
for underlying mental factors can only be derived from testing batteries
that include markers for that factor. This and other problems, discussed
in length next, limit the effectiveness of factor analysis as a means of un-
derstanding the structure of the intellect. Yet much of the progress made
toward understanding visuospatial ability arose from the factor-analytic
tradition.
During the middle of the 20th century, research on individual differ-
ences in spatial abilities focused on determining the factor structure of spa-
tial ability. Investigators such as Guilford and Lacey (1947), Zimmerman
(1954), Thurstone (1950), and French (1951) all found that large batteries
of spatial tests yielded evidence for several separable subcomponents of
spatial ability. By the 1960s, many investigators no longer treated spatial
ability as a unitary factor – rather, as an amalgam of several correlated
factors. However, theorists differed on how best to characterize these fac-
tors. Some of these different characterizations, which we now discuss, are
summarized in Table 4.1. Michael, Gulford, Fruchter, and Zimmerman
(1957) published a review of then-existing literature, concluding that there

table 4.1. Summary of the spatial abilities factors identified in some of the primary
studies using the factor-analytic approach, including representative spatial abilities
tests cited by the authors as markers for each factor.

Tests cited as typical

Study Factors identified markers for each factor
Michael, Guilford, 1. Spatial Visualization - Paper Folding, Form Board
Fruchter, & (see Figure 4.1a)
Zimmerman, 2. Spatial Relations and - Cube Comparisons Test
1957 Orientation (see Figure 4.1b), Guilford–
Zimmerman Spatial
Orientation (see Figure 4.2),
Card Rotations (see Figure 4.3a)
3. Kinesthetic Imagery - Hands test∗
McGee, 1979b 1. Spatial Visualization - Paper Folding
2. Spatial Orientation - Cube Comparisons,
Guilford–Zimmerman
Spatial Orientation
Lohman, 1988 1. Spatial Visualization - Paper Folding, Form Board,
Cube Comparisons
2. Spatial Relations - Card Rotations

3. Spatial Orientation - Guilford–Zimmerman

Spatial Orientation
Carroll, 1993 1. Spatial Visualization - Paper Folding, Form Board,
Cube Comparisons,
Guilford–Zimmerman
Spatial Orientation
2. Spatial Relations - Card Rotations
3. Closure Speed - Snowy Pictures (see Figure 4.3b)
4. Flexibility of Closure - Hidden Figures (see Figure 4.3c)
5. Perceptual Speed - Identical Pictures (see Figure 4.3d)
6. Visual Memory - Silverman–Eals visual
memory task (see Figure 4.4)

was consistent evidence for three factors of spatial ability: (a) Spatial
relations and orientation, which was theorized to involve the ability to
understand the arrangement of elements within a visual stimulus, pri-
marily with respect to one’s body frame of reference; (b) Visualization,
which was thought to require the mental manipulation of objects; and

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 25 Jul 2018 at 02:51:40, subject to the Cambridge Core
Cambridge Books Online © https://doi.org/10.1017/CBO9780511610448.005
terms of use, available at https://www.cambridge.org/core/terms. Cambridge University Press, 2010
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511610448.005
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 25 Jul 2018 at 02:51:40, subject to the Cambridge Core

126

figure 4.1. Sample items from three different tests that load on the spatial visualization, as this factor
is described by Lohman (1988) and Carroll (1993). Because many different types of tests load on this
factor, three representative examples are given: (a) Items on the Paper Folding test from Ekstrom et al.
(1976, top) require examinees to choose which of the five alternatives (right) represents the appearance
of the paper if it has been folded and punched as indicated on the left; (b) The middle panel illustrates
three sample items from the Cube Comparisons test (Ekstrom et al., 1976). Examinees determine
whether the two depicted cubes represent the same or different cubes; (c) The lower panel illustrates
a sample item from the Form-Board test. Examinees determine which of the shapes on the bottom
could be rearranged to complete the rectangle above them.
Cambridge Books Online © Cambridge University Press, 2010
Individual Differences in Spatial Abilities 127

(c) Kinesthetic imagery, which was believed to be associated with left-right

discrimination. Michael et al. admitted that none of these descriptions
yielded perfectly clear-cut distinctions empirically, and despite many ef-
forts to distinguish these factors, correlations among tests that served as
markers for these factors were often very high (Borich & Bauman, 1972;
Vincent & Allmandinger, 1971; Smith, 1964) – occasionally approaching
the tests’ own reliability (Roff, 1952).
Despite the shortcomings of the Michael et al. description of spatial
ability, a strikingly similar model prevailed through the 1980s, primarily as
a result of extensive summaries and reviews conducted by McGee (1979a)
and Lohman (1979). McGee pointed out that much of the confusion and
inconsistencies concerning the factor structure of spatial ability is due
to inconsistent naming conventions among investigators. In his review,
McGee concluded that most researchers have distinguished what he
regarded as the same two fundamental dimensions of spatial ability (see
Table 4.1). The strongest factor, spatial visualization (or VZ), was first
defined by Guilford and Lacey (1947) and is typically described as an
ability to manipulate, rotate, twist, or invert objects without reference
to one’s self. This ability is required for successful performance on
tasks such as the Paper Folding test from the Ekstrom, French, Harman,
and Dermen (1976) test battery (Figure 4.1a). For McGee, the second
prominent dimension of spatial ability was spatial orientation (SO), which
he characterized as “the comprehension of the arrangement of elements
within a visual stimulus pattern and the aptitude to remain unconfused
by the changing orientation in which a spatial configuration may be
presented” (McGee, 1979b, p. 893). Spatial orientation is often conceived
as an ability to imagine the appearance of objects from different orienta-
tions (perspectives) of the observer and has been described by Thurstone
as the “ability to think about those spatial relations in which the body
orientation of the observer is an essential part of the problem” (Thurstone,
1950, as reported in McGee, 1979a). A typical marker for this ability is the
Guilford–Zimmerman (1948) Spatial Orientation test (Figure 4.2), which
requires users to determine the change in position implied by two views
from the prow of a boat. However, McGee’s concept of spatial orientation
was somewhat broader than that of Thurstone. For McGee, tests of mental
rotation – which do not typically involve imagining a change in the
body orientation of the observer – were considered as markers for SO.
In another influential review, Lohman (1979) supported the existence
of three major spatial abilities factors (Table 4.1). The first was spatial
relations, which is defined by tests that require speeded performance of

128

figure 4.2. Three sample items from the Guilford–Zimmerman test of Spatial Orientation. Exami-
nees determine which of five choices represents the change in orientation from the top to the bottom
view of a boat’s prow. Consulting Psychologists Press. Reprinted by permission.

Cambridge Books Online © Cambridge University Press, 2010

Individual Differences in Spatial Abilities 129

simple (two-dimensional) mental rotation items (such as the Card Rota-

tions test shown in Figure 4.3a). Lohman later referred to this factor as
speeded rotation (Lohman, 1988). Lohman’s second factor was spatial ori-
entation, which he defined as the ability to imagine how a stimulus array
will appear from another perspective – a more restricted definition than
that of McGee, and similar to that of Thurstone (1950). Lohman agreed with
McGee on the definition of his third factor, spatial visualization, but em-
phasized that it is defined by difficult spatial tasks that require a sequence
of transformations of a spatial representation, and more complex stimuli.
Examples of tests loading on the spatial visualization factor are the paper-
folding test, form-board test, and mental rotation of three-dimensional
figures, shown in Figure 4.1.
In the most extensive study to date, Carroll (1993) surveyed and reana-
lyzed more than 90 data sets that bear on the factor structure of visuospatial
ability. Carroll examined the support for five visuospatial factors in the cat-
egory that he referred to as “abilities in the domain of visual perception.”
These were visualization (VZ), spatial relations (SR), closure speed (CS),
flexibility of closure (CF), and perceptual speed (P). A fifth factor, visual
memory (MV) is described in a chapter of Carroll’s book on abilities in the
domain of learning and memory. It should be noted that Carroll’s defini-
tion of visual-spatial abilities was somewhat broader than those of McGee
(1979b) and Lohman (1979). For example, Lohman acknowledged the exis-
tence of closure speed, perceptual speed, and visual memory, but referred
to them as minor factors that are not central to what is meant by “spatial
ability.”
Carroll found strong and consistent support for a spatial visualization
factor that emphasizes “power in solving increasingly difficult problems
involving spatial forms” (Carroll, 1993, p. 315). He also found support for
closure speed (CS) and flexibility of closure (CF), both of which involve
the ability to identify a stimulus (or part of a stimulus) that is either em-
bedded in or obscured by visual noise. In general, tests of closure speed
do not give the examinee information about the stimulus to be discov-
ered. The target stimulus typically depicts a common object, and items are
scored by the time to recognize these targets (see the sample item from
the Snowy Pictures test in Figure 4.3b). Closure speed thus involves the
ability to access representations quickly from long-term memory. On the
other hand, tests of flexibility of closure tend to include the target stimu-
lus (typically a geometric design), so that examinees know what to search
for (see the sample item from the Hidden Figures tests in Figure 4.3c).

130

figure 4.3. Sample items from tests of different spatial abilities factors identified by Carroll (1993), all
taken from Ekstrom et al. (1976): (a) The Card Rotations test is a measure of spatial relations (speeded
rotation). Examinees must determine which of the items on the right are rotations of the item on the
left of the line (as opposed to its mirror image); (b) The Snowy Pictures test is a test of closure speed.
Examinees must identify the object in the picture (in this case, an anchor); (c) The Hidden Figures test
is a test of flexibility of closure. Examinees must determine which of the shapes, labeled A to E, is
contained in each of the complex figures below; (d) The Identical Pictures test is a test of Perceptual
Speed. Examinees must identify which of the pictures on the right is identical to the one on the left.
Cambridge Books Online © Cambridge University Press, 2010
Individual Differences in Spatial Abilities 131

Flexibility of closure is thus more closely associated with the ability to

hold a stimulus in working memory while attempting to identify it from a
complex pattern.
Carroll also found evidence for a factor called perceptual speed (P),
which was described by French (1951) as “speed in comparing figures or
symbols, scanning to find figures or symbols, or carrying out other very
simple tasks involving visual perception.” Tasks that measure this fac-
tor often involve either matching two stimuli or searching a field for a
target stimulus (see the sample item from the Identical Pictures test in
Figure 4.3d). These tasks require virtually no mental transformations of
stimuli and typically rely more on visual than spatial processing. Finally,
Carroll recognized support for the existence of a visual memory factor,
described by Ekstrom et al. (1976) as “the ability to remember the con-
figuration, location, and orientation of figural material.” Tests of visual
memory generally require examinees to study a stimulus configuration
for several minutes. These configurations can be either a maplike depic-
tion of an environment, an array of objects, or a configuration of meaning-
less shapes. Examinees must then either recognize or recall the identity or
the location of objects within the configuration (Figure 4.4). Visual-spatial
memory has been of interest recently because Silverman and Eals (1992)
have found significant gender differences favoring women on tests of spa-
tial memory (see also Montello, Lovelace, Golledge, & Self, 1999). More
recently, Robert and Ecuyer-Dab (2000) have shown that women’s advan-
tage in this task extends to memory of both stationary and repositioned
objects.
A somewhat surprising result of Carroll’s study is that he failed to find
strong and consistent evidence for the separability of spatial relations from
the spatial visualization factor; only 7 of the 94 data sets that he examined
showed such a distinction. Spatial relations, associated with the ability
to perform relatively simple mental transformations (typically rotations)
quickly, was identified in 21 of the data sets reviewed by Carroll. Both
Carroll and Lohman agree that a distinction between these two factors
emerges only if several tests of each type appear in the testing battery. De-
spite this observation, the distinction between spatial relations and spatial
visualization has remained important in the cognitive analysis of spatial
test performance.
Carroll also failed to find consistent support for the separability of spatial
orientation (SO) and spatial visualization (SV). Many investigators have
supported the existence of spatial orientation (Guilford & Zimmerman,

figure 4.4. A test of visual memory developed by Silverman & Eals (1992).
After studying the picture on the left, examinees are shown one of the pic-
tures on the right. In one condition (upper panel) examinees must put a cross
through all the items that were not in the picture that they studied. In the
other condition (lower panel) they must circle the items that were in the same
place in the picture that they studied and put a cross through the items that
have moved from their position in the studied picture. From J. H. Barkow,
L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolutionary psychology and the
generation of culture (pp. 357–358). Oxford: Oxford University Press. Reprinted by
permission.

1948; McGee, 1979b; Thurstone, 1950) and it continues to appear promi-

nently in descriptions of spatial ability. However, several studies have
found that spatial orientation and spatial visualization are highly re-
lated (Borich & Bauman, 1972; Goldberg & Meredith, 1975; Vincent &
Allmandinger, 1971), with markers correlating as highly as .75 (Roff, 1952).
Although he generally supported the distinction of spatial orientation from
spatial visualization, Lohman (1979) also concluded that “It is clear that the
Guilford–Zimmerman Spatial Orientation and Spatial Visualization tests
do not measure different factors.” Finally, Carroll failed to find evidence
for a factor known as spatial scanning, which was described by Ekstrom
et al. (1976) as “speed in exploring visually a wide or complicated spatial
field.”

Evaluation of the Factor-Analytic Literature

The factor-analytic tradition has delineated several separable dimensions
of visuospatial ability, and these abilities enable investigators to conceptu-
alize the domain of spatial intelligence. Yet the factor analytic approach is
not without its problems and limitations. Perhaps one weakness is the fact
that most factor analytic studies of spatial ability have, until recently, gen-
erally employed exploratory factor analyses, the technique used to derive a
relatively small set of factors from a large set of variables. Exploratory factor
analysis can be used without any prior understanding of the interrelations
among a set of variables. There are several variations of exploratory fac-
tor analysis techniques (and different methods of rotating solutions), and
the use of different techniques can result in different conclusions being
drawn. The choice of which technique to use is often based on subjective
grounds. As a result, many different exploratory factor-analytic models
have proliferated, each unable to disconfirm alternative models of spa-
tial ability (Gustafsson, 1984; Sternberg, 1981). An increasingly popular
approach that attempts to remedy some of these problems is called confir-
matory factor analysis. This technique relies critically on prior specification
of the underlying factor structure. In confirmatory factor analysis, the re-
searcher begins with a hypothesized factor structure and then tests how
plausible it is given the observed correlations among measures. Gustafsson
(1984) has argued that this class of model allows a sharper analytic tool,
capable of disconfirming theories in a way that exploratory analyses can-
not. Later in the chapter, we will report some examples of recent research
that has used this type of model.
Another shortcoming of the factor-analytic approach concerns the as-
sumptions it makes in terms of strategies. Correlational analyses such as
factor analysis not only assume that all examinees use the same strategy,
but that a given examinee uses the same strategy on all items. Both of these
assumptions are in general false (Barratt, 1953; Just & Carpenter, 1985;
Lohman, 1988). That different strategies can be used to solve the same test
item greatly confounds the correlational analyses of underlying mental
processes. As we will see, research aimed at understanding strategies used
on spatial ability tests has been enormously helpful in understanding the
role of individual differences in spatial ability.
We have seen that, despite the widespread belief in the independent
existence of SO, Carroll was unable to establish conclusive support for this
factor. The confusion over the existence of spatial orientation highlights
another problem with the factor analytic approach: Factor analyses are

only able to describe the underlying structure of the variables that go into
the analysis. If markers for a relevant or important factor are left out of
a testing battery, the factor obviously cannot emerge from the analysis.
Similarly, if a factor is poorly represented by too few or weak markers,
then the recovered factor structure may not be very reliable.
Although Carroll’s extensive reanalysis found little support for spatial
orientation, this does not mean that such an ability does not exist – rather,
it is possible that this ability has simply been poorly assessed historically.
Theoretically, the critical distinction between spatial orientation and spa-
tial visualization is that spatial orientation is related to the ability to imag-
ine changing one’s viewpoint (perspective), whereas spatial visualization
involves imagining the movement of objects. Barratt (1953) and others
have noted that test items putatively measuring spatial orientation can
often be solved by imagining the stimulus moving rather than the imag-
ining a change in the viewpoint of the observer. For example, although
the Guilford–Zimmerman test of Spatial Orientation (Figure 4.2) is often
thought to be a strong marker for SO, this test does not require adoption of
alternate perspectives and most examinees do not do so (Barratt, 1953). It is
thus possible that spatial orientation ability represents a separable spatial
factor, but there have not been enough pure tests of this ability for it to
be adequately measured and adopted into psychometric models of spatial
ability.
Kozhevnikov and Hegarty (2001) have recently shown evidence of a
perspective-taking ability that is distinct from spatial visualization. In their
study, participants were shown a two-dimensional array of objects, and
were asked to imagine themselves facing a particular direction within the
array. They then indicated the direction to a target object from that per-
spective. Verbal reports from the participants indicated that the dominant
strategy used to solve the test items was to imagine oneself reoriented with
respect to the display (in a protocol study, only one out of eight participants
reported rotating the array), suggesting that it is more effective than the
Guilford–Zimmerman at measuring spatial orientation ability. This con-
clusion was also supported by systematic errors in which participants con-
fused left/right as well as back/front pointing directions, suggesting that
they encoded the locations of the objects with respect to body coordinates.
A confirmatory factor analysis indicated that this measure of perspective
taking is dissociable from measures of spatial visualization ability, while
the Guilford–Zimmerman test is not (see also Hegarty & Waller, 2004).
If the factor analytic tradition has garnered only weak evidence for spa-
tial orientation because of ineffective assessments of this ability, it has all but

neglected assessing spatial abilities in at least two other relevant domains:

dynamic spatial abilities, which are involved in reasoning about moving
stimuli, and large-scale environmental spatial abilities, which are involved
in acquiring knowledge as one moves around a large-scale environment.
Much of this neglect stems from the fact that these factors can be difficult to
assess by means of paper-and-pencil tests. We will discuss environmental
spatial abilities in some detail later. For now, we focus briefly on what is
known about dynamic spatial abilities.
Dynamic spatial abilities are those that are required to reason about
moving stimuli. The advent of computer testing has led to renewed in-
terest in the ability to reason about motion and the ability to integrate
spatial information over time. Even before computer testing, though, Seib-
ert (1970) reported an ability to “bridge small gaps in time and to merge
intellectually . . . time-segmented fragments” and dubbed it serial integra-
tion (SI). This factor was revealed in tests in which the examinee identi-
fied a stimulus, parts of which were successively revealed and concealed
over time. Gibson (1947) conducted similar tests on his “successive per-
ception” ability using motion pictures as stimuli. More recently, Jones,
Dunlap, and Bilodeau (1986) showed that performance on computer video
games required different abilities from those assessed by paper-and-pencil
tests. Similar conclusions were made by Hunt, Pellegrino, Frick, Farr, and
Alderton (1988), who reported evidence for an ability to reason about mov-
ing stimuli (see also Fischer, Hickey, Pellegrino, & Law, 1994; Law et al.,
1993).
Hunt et al. gave participants several traditional paper-and-pencil tests
of spatial ability, several computer-administered tests of spatial ability in-
volving static stimuli, and several computer-administered tests involving
dynamic stimuli. The static computerized tests required participants to
make perceptual comparisons of shapes quickly, to perform mental ro-
tation, to image pieces combining to form a larger object (i.e., a form-
board test), to add detail to mental images, and to compare folded and
unfolded versions of cubes (i.e., surface development). The dynamic tests
measured people’s ability to remember and extrapolate movement tra-
jectories, judge absolute and relative velocities, and combine judgments
of velocity and path extrapolation. In one task, for example, participants
watched a target move horizontally across the computer screen. Partici-
pants pressed a button to initiate the vertical motion of another object on
the screen. The task required participants to time their response in order to
make the object intersect the moving target. Hunt et al. found that, in gen-
eral, the static computer-administered tests correlated more highly with

the paper-and-pencil tests than with the computer-administered dynamic

tests. This pattern of results, as well as results from confirmatory factor
analyses, led them to conclude that the ability to reason about dynamic
spatial relations represented a distinct factor of human spatial aptitude.
Despite the appeal of this conclusion, the precise characterization of
performance differences on dynamic versus static tests is still an open
question. For example, Larson (1996) has questioned the degree to which
performance differences on dynamic versus static tasks are due to stim-
ulus motion alone, and pointed out that most of the dynamic tasks used
by Hunt et al. require entirely different mental operations than the static
tasks require. When these operations are matched between static and dy-
namic stimuli, the difference between the two may be greatly reduced.
Larson (1996) gave participants dynamic and static versions of a men-
tal rotations task and found that performance on the two tasks corre-
lated extremely highly. He concluded that more evidence is required to
establish conclusively that there is a distinct visuospatial factor associ-
ated with the ability to reason about moving stimuli. Clearly, however,
evidence for a dynamic spatial ability cannot be derived from tasks that
require the same mental processes as static tests. The fact that Hunt et
al. designed their tasks to encompass a broad range of mental processes
and still found that these tasks correlate relatively highly with each other
provides fairly compelling evidence for a general ability to reason about
motion.

cognitive analyses of performance on

spatial abilities tests
The factor-analytic tradition has generally focused on the discovery and
description of the factors that underlie spatial ability. Perhaps more in-
formative for cognitive psychologists is an analysis of the perceptual and
cognitive processing demands made by these factors. Performance on tests
of spatial abilities depends on execution of basic cognitive processes such
as encoding a visual stimulus, constructing a visual image, retaining an im-
age in working memory, transforming an image, and comparing a visual
stimulus to an image in working memory. In recent years, cognitive psy-
chologists have attempted to understand spatial abilities at a deeper level
by analyzing the basic cognitive processes involved in solving items from
tests of spatial abilities and examining the extent to which individual dif-
ferences in these basic processes can be related to theories of imagery and
working memory. These studies often combine the use of experimental

methods with examination of individual differences to study aptitude-

treatment interactions, as advocated by Cronbach (1957). For example, one
method is known as the “componential approach” to understanding cog-
nitive abilities (Sternberg, 1977). In this approach, researchers first conduct
task analyses to identify the basic cognitive processes involved in per-
forming items from psychometric tests. They then measure performance
on the psychometric tests and on laboratory tests designed to isolate the
constituent cognitive processes. By analyzing errors and reaction times on
the laboratory tasks, researchers examine the extent to which individual
differences in performance on the psychometric tests can be accounted for
by speed and accuracy of the basic cognitive processes.

The Contribution of Speed of Processing

Tests of spatial ability usually involve items that must be solved within a
time limit. Thus, the speed of basic cognitive processes is clearly a possible
source of individual differences in test performance. Mumaw, Pellegrino,
Kail, and Carter (1984) used a componential approach to study speed of
processing in mental rotation tests. Their task analysis revealed that mental
rotation involves processes of (1) encoding the stimuli, (2) rotation of one
of the stimuli, (3) comparison of the rotated image to the other stimulus,
and (4) executing a response. The researchers then assessed the perfor-
mance of individuals on the Primary Mental Abilities (PMA) Space test
(Thurstone, 1958), which is a test of spatial relations (speeded rotation),
and also measured their speed and accuracy in laboratory mental rotation
tasks involving alphanumeric characters and characters from the PMA test.
For the laboratory mental rotation tasks, they calculated the intercept and
slopes of the functions relating reaction time to rotation angle, arguing
that the intercept represented the time to encode and compare the stim-
uli and to execute a response, whereas the slopes represented the time to
mentally rotate the figures. Performance on the PMA was correlated with
the slopes of the reaction time functions for both alphanumeric and PMA
characters. It was also correlated with the intercept for PMA characters
(but not alphanumeric characters) and was not correlated with accuracy
on the mental rotation trials. Mumaw et al. concluded that individual dif-
ferences in tests of spatial relations are largely due to differences in speed of
mental rotation and speed of encoding and comparing unfamiliar stimuli.
Differences in rate of rotation between high- and low-spatial individuals
have also been reported in several other studies (Just & Carpenter, 1985;
Pellegrino & Kail, 1982).

Although speed of processing can account for individual differences in

simple spatial-relations tests, this is not the case for more complex tests that
load on the spatial visualizations factor. These tests (e.g., paper-folding and
form-board tests) typically require a sequence of spatial transformations
and have relatively liberal time limits. To examine the effects of task com-
plexity, Lohman (1988) constructed a faceted spatial test in which subjects
had to retain, synthesize, rotate, and compare polygons. Items differed in
the number of these basic processes that were required. For example, the
simplest items merely required participants to maintain the representation
of a polygon and match it against another, whereas more complex items
involved synthesizing the representations of three different polygons and
rotating the resulting representation. By varying the number of basic oper-
ations required by each item, and measuring reaction time on these items,
Lohman could estimate the time taken for each basic operation. An impor-
tant result of this study was that time to execute the basic processes was a
poor predictor of the complexity of items that an individual could solve.
Mumaw and Pellegrino (1984) reached the same conclusion in a study in
which they varied the complexity of items in a form-board test.
To examine the relative contributions of speed and complexity to per-
formance on a complex mental rotation task (rotation of three-dimensional
shapes of the type used by Shepard & Metzler, 1971) Lohman (1986)
constructed a task in which he varied the amount of presentation time
(exposure) allowed for individual test items and the angle of rotation.
Lohman then computed accuracy at these different levels of exposure and
rotation, creating speed-accuracy curves for high- and low-spatial individ-
uals. These curves showed that accuracy reached an asymptote at a lower
level for low-spatial participants. That is, for any level of complexity (say a
rotation of 120 degrees) there was a maximum level of accuracy reached by
participants that did not increase with more time on the task. This level of
accuracy was lower for low-spatial participants than high-spatial partici-
pants and the difference was greater for trials that required more rotation.
Thus, for imagined rotation of three-dimensional figures, high- and low-
spatial individuals differ in the complexity of items that they can rotate,
and not merely in the speed of rotation.
In conclusion, componential analyses of spatial tests have identified a
“speed – power” continuum along which tests can be placed (Lohman,
1988). Items on simpler tests, such as those loading on the perceptual
speed or spatial relations factor, can be easily solved by most people. A
speed limit must be imposed in order to observe individual differences,
and consequently, these differences reflect speed of processing (Pellegrino

& Kail, 1982). In contrast, items loading on the spatial visualization factor,
including three-dimensional rotation items, paper-folding and form-board
tests are more complex, and cannot be solved by all individuals, even when
given a liberal time limit. Differences in performance in these tests are char-
acterized by differences in “power,” or the complexity of items that a given
individual can solve. It is generally assumed that these tests measure the
ability to construct a high-quality representation of a novel stimulus, and
to maintain the quality of that representation after transformations, such
as stimulus rotation and synthesis (Lohman, 1979; Just & Carpenter, 1985;
Mumaw & Pellegrino, 1984; Pellegrino & Kail, 1982).
On the other hand, research on cognitive aging (Salthouse, 1996) has
suggested how individual differences in speed of processing might lead to
the pattern of performance observed on power tests. Salthouse has noted
that “power” tests involve a series of transformations, and the products
of these transformations must be in memory at the same time to accom-
plish the task. Differences in speed of processing might lead to failure to
complete the task because the products of early processes are no longer
available by the time that later processes are completed. Salthouse has re-
ferred to this as the “simultaneity mechanism.” This mechanism explains
how differences in speed of processing might lead to an inability to solve
items from complex spatial tasks such as form board and paper folding,
even with a liberal time limit. Research reviewed earlier indicates that
speed of elementary processing is not always correlated with accuracy in
power tests, so speed is probably not the only factor accounting for indi-
vidual differences in performance of these tests. Nevertheless, the possible
contribution of the simultaneity mechanism needs to be more carefully
examined in the individual differences literature.

The Contribution of Strategies

Because spatial visualization tests include more complex transformations,
they are amenable to a range of strategies, and strategy differences can be
another source of individual differences in performance. We already ob-
served this in the discussion of performance on the Guilford–Zimmerman
spatial orientations test for which Barratt (1953) found that the majority of
individuals reported a mental rotation strategy rather than a true spatial
orientation strategy.
Just and Carpenter (1985) used eye-fixation and verbal protocols to
analyze individual differences in the Cube Comparisons test, shown in
Figure 4.1b. They identified three strategies used to perform these items: a

mental-rotation strategy, a perspective-taking strategy in which the solver

imagines moving with respect to a fixed cube, and a strategy of comparing
orientation-free descriptions of the cubes (e.g., the relative orientations of
the letters on the right and left cubes). Most participants used a mental ro-
tation strategy, but an important difference between high- and low-spatial
individuals was that in using this strategy, low-spatial individuals rotated
around axes that were orthogonal to the faces of the cubes. In contrast,
high-spatial individuals rotated around noncanonical axes. Therefore, a
trial that might involve two or more orthogonal rotations for a low-spatial
individual could be accomplished in a single rotation by a high-spatial
participant.
In addition to differences among individuals in solution strategy, indi-
viduals do not always use the same strategy to solve all items in a spatial
visualization test. Kyllonen, Lohman, and Woltz (1984) modeled reaction
times on Lohman’s (1979) faceted spatial task (described earlier) and found
that the reaction times for many participants were best fit by models assum-
ing that they shifted strategies depending on practice and complexity of the
item presented. Furthermore, the optimal strategies for solving complex
spatial items did not necessarily require imaging a sequence of transfor-
mations to a figure. Rather, individuals sometimes depended on analytic
strategies. For example, in the paper-folding test, they often applied the
analytical principle that if the paper is folded down the middle before the
hole is punched, the pattern of holes should be symmetrical around that
fold. Kyllonen, Lohman, and Snow (1984) trained people to solve items on
the paper-folding test using both visualization and analytical strategies.
The effectiveness of this training depended on a complex interaction be-
tween the spatial and verbal aptitudes of the learner and characteristics of
the problems. For example, for high-spatial individuals, the effectiveness
of training depended on the difficulty of the problems (analytic training
was preferable for the most difficult items whereas visualization training
was preferable for other items). In the case of low-spatial individuals, this
effectiveness depended on their verbal abilities (analytic training was more
effective for high-verbal individuals, visualization training was more ef-
fective for those with low-verbal ability). Thus, it appears that to some
extent spatial visualization tests measure the ability to choose the opti-
mal strategy for solving a particular item, given one’s abilities. The fact
that spatial visualization items are often best solved by analytical strate-
gies has been cited as an explanation of the fact that spatial visualization
tests often load on the G factor, measuring general intelligence (Lohman,
1988).

Spatial Abilities and Working Memory

A major conclusion of cognitive research on spatial abilities, especially
spatial visualization ability, is that high- and low-spatial individuals differ
in the quality of the spatial representations that they construct and their
ability to maintain this quality after transforming the representations in
different ways (Carpenter & Just, 1986; Just & Carpenter, 1985; Lohman,
1988; Mumaw & Pellegrino 1984). Several difficulties experienced by low-
spatial individuals suggest that they are less able to maintain a complex
image after transforming it. For example, in the Cube-Comparisons task,
individuals often “lose” information about the letter on one side of a cube
after that side has been rotated “out of view” in their mental image (Just &
Carpenter, 1985; Carpenter & Just, 1986). Similarly, when rotating Shepard-
Metzler figures, low-spatial individuals often have to make several rota-
tion attempts, suggesting that they lose information about a figure while
attempting to rotate it (Carpenter & Just, 1986; Lohman, 1988). Individ-
ual differences in spatial visualization have therefore been modeled as
differences in working memory resources for storage and processing of
spatial information (Carpenter, Just, Keller, Eddy, & Thulborn, 1999; Just &
Carpenter, 1985). In this view, a high-spatial individual might have more
resources for storing and processing spatial information than a low-spatial
individual.
Current theories conceptualize working memory as a system specialized
for maintenance of information in an activated state as necessary for task
performance, and for executive control of attentional resources in order to
maintain task goals, inhibit distracting information, and schedule differ-
ent subprocesses required to accomplish a complex task (Baddeley, 1986;
Engle, Kane, & Tuholski, 1999; Miyake & Shah, 1999). These theories make
a distinction between “short-term memory” tasks (such as digit span) that
merely require the maintenance of information, and “working memory”
tasks (such as the Reading Span Test, Daneman & Carpenter, 1980) that
require storage in the face of competing processing requirements (Cowan,
1995). Furthermore, there is considerable evidence that maintenance of
verbal and spatial information is carried out by different systems, which
Baddeley (1986) referred to as the phonological loop and the visual-spatial
sketchpad, respectively.
A number of recent studies have addressed the extent to which in-
dividual differences in spatial ability can be explained by differences in
the functioning of working memory. Shah and Miyake (1996) examined
the degree to which individual differences in spatial visualization can be

accounted for in terms of differences in a spatial working memory sys-

tem that is distinct from a verbal working memory system. They devel-
oped two tests of spatial working memory. The first, called the Arrow
Span test, was a spatial “short-term memory” task that requires individ-
uals to maintain spatial information (the direction of arrows) in work-
ing memory. The second, called the Rotation-Arrow Span, was a spatial
“working memory” task that requires the maintenance of this information
in the face of interference from an unrelated task (mental rotation). Both
tasks were highly correlated with tests of spatial visualization. In con-
trast, spatial visualization tests had low correlations with a test of verbal
working memory (the reading span test) and a test of verbal short-term
memory (digit span). These results add support to the view that tests of
spatial visualization rely on a spatial working memory system that is dis-
sociable from verbal working memory. This conclusion was supported
by a later study (Hegarty, Shah, & Miyake, 2000) showing that a spatial
concurrent task (spatial tapping) interferes with performance on spatial
abilities tests while a verbal concurrent task (articulatory suppression)
does not.
Another issue, addressed by Miyake, Rettinger, Friedman, Shah, and
Hegarty (2001) is the extent to which different spatial abilities factors can be
explained by differential involvement of the storage and executive control
components of the working memory system. The three factors examined
by Miyake et al. (2001) were spatial visualization (SV), spatial relations
(SR), and perceptual speed (P). The authors reasoned that tasks loading on
all three factors seem to require the temporary maintenance of visuospa-
tial information and thus should all involve the visuospatial component of
working memory. In contrast, the three factors seem to differ in terms of the
demand they place on the executive component of working memory. Tests
of spatial visualization appear to be the most demanding of the executive
component, because they require a sequence of internal spatial transfor-
mations to be performed on the stimulus, involving the management of
task-specific goals and subgoals as well as the scheduling and coordina-
tion of different cognitive processes (Lohman, 1996; Marshalek, Lohman,
& Snow, 1983). The spatial relations factor should involve the executive to
a lesser extent. Although tasks that load on this factor also necessitate men-
tal transformations, only a single transformation is needed for each item,
and this requires less executive control. Finally, the visuospatial perceptual
speed factor should require the least executive involvement because tasks
that load on this factor merely require the maintenance of a visuospatial
representation of a simple figure in memory and do not require any spatial
transformation or extensive goal management.

Miyake et al. (2001) measured performance on tests of the three spatial

factors and tests of spatial short-term memory (storage of spatial informa-
tion), spatial working memory (storage and processing of spatial informa-
tion), and executive working memory. First, confirmatory factor analysis
of the working memory tasks did not indicate a clear distinction between
spatial short-term memory and spatial working memory, so these tests
were combined into a single visuospatial working memory factor, which
was separable from the executive WM factor. Second, Miyake et al. exam-
ined the relation between the executive and spatial components of working
memory and the three spatial ability factors (i.e., spatial visualization, spa-
tial relations, and perceptual speed) using structural equation modeling.
This is a multivariate technique that allows one to measure the degree of
relation between latent variables, (i.e., the underlying factors of factor anal-
ysis), rather than that between manifest variables (i.e., individual tasks).
In this particular study, structural equation modeling was used as a ver-
sion of multiple regression analysis in which the predictor and dependent
variables are all latent (as opposed to manifest) variables. By comparing
the standardized coefficients for the paths from the executive and visu-
ospatial WM and STM variables to the spatial ability factors, we evaluated
our hypothesis regarding the contributions of executive and visuospatial
aspects of working memory to spatial abilities. This analysis indicated that
as predicted, the contribution of the executive variable was the strongest
for the spatial visualization factor (.91) and the weakest for the perceptual
speed factor (.43). The contribution of the visuospatial working memory
factor was important across all three spatial ability factors.
These results suggest a close link between individual differences in spa-
tial working memory and those in spatial abilities. They also shed new
light on why the spatial ability subfactors (especially spatial visualization
and spatial relations) are not completely independent and are usually cor-
related with one another. These factors are similar in as much as they rely
on both executive control mechanisms and spatial storage. On the other
hand, they can show some separability because the relative importance of
these two components of working memory is different. Thus, depending
on the difficulty levels of the psychometric tests or the ability levels of
the test takers, it might not be always possible to clearly distinguish these
factors.

Spatial Abilities and Mental Imagery

Analyses of the cognitive components of spatial ability often refer to men-
tal imagery processes such as maintenance and transformation of mental

images. There is also some overlap between the tasks included in the mental
imagery literature and those on spatial abilities tests, the most prominent
example being mental rotation. However, the psychological investigations
of mental imagery and spatial abilities have generally been quite separate,
with studies of imagery tending to use experimental methods rather than
measuring individual differences. Consequently, there have been only a
few attempts to study the relation between individual differences in spa-
tial test performance and individual differences in imagery ability.
Historically, imagery ability has referred to differences in self-reported
vividness of imagery. This tradition goes back to Galton’s (1883) classic
study in which he asked people to imagine their breakfast table on a given
day and to rate the vividness of this image. Based on this tradition, re-
searchers developed self-report measures such as the Questionnaire upon
Mental Imagery (QMI) developed by Betts (1909) and the Vividness of Vi-
sual Imagery Questionnaire (VVIQ) developed by Marks (1973). In these
measures, people read descriptions of perceptual experiences (scenes, etc.),
constructed images of these experiences, and then rated the vividness of
their mental images. This type of imagery ability has been consistently
found to be unrelated to measures of spatial ability (Poltrock & Agnoli,
1986).
However, research in cognitive neuroscience and working memory sug-
gests that there are different dimensions of imagery ability, some of which
may be more closely related to spatial abilities (e.g., Kosslyn, 1995). In this
literature, a distinction has been made between primarily visual imagery
and primarily spatial imagery. Visual imagery refers to a representation
of the visual appearance of an object, such as its shape, color, or bright-
ness. A typical visual imagery task might be to state the color of an object
that has a characteristic color that is not verbally associated with the object
(e.g., a football). Spatial imagery refers to a representation of the spatial
relationships between parts of an object, the location of objects in space or
their movement, and is not limited to the visual modality (i.e., one could
have an auditory or haptic spatial image). A typical spatial imagery task
is mental rotation. For example, Farah, Hammond, Levine, and Calvanio
(1988) demonstrated that following brain lesions, patients can be extremely
impaired in tasks that require visual aspects of imagery yet show normal
performance in tests of spatial imagery. In the working-memory literature,
dual-task studies have also shown that visual imagery tasks are impaired
by concurrently viewing irrelevant pictures but not by arm movements,
whereas spatial imagery tasks are impaired by arm movements but not by
irrelevant pictures (Logie, 1995).

Another conception of individual differences in imagery is based on

Kosslyn’s computational model of mental imagery (Kosslyn, 1980; Kosslyn
& Schwartz, 1977). According to this theory, imagery takes place in a “visual
buffer” with limited resolution and extent. The theory postulates several
processing modules that are responsible for creating images from various
sources, maintaining these images in memory, and transforming the images
in various ways. For example, one module called PICTURE creates a visual
image of an object from a long-term memory representation. Another called
REGENERATE is responsible for refreshing images so that they remain
activated in memory. A third called TRANSLATE moves part of an image
relative to other parts.
Kosslyn, Brunn, Cave, and Wallach (1984) examined individual dif-
ferences in the Vividness of Visual Imagery Questionaire (VVIQ), a test
of spatial visualization (form board), and ten imagery tasks related to
Kosslyn’s (1980) theory of imagery. The imagery tasks included measures
of the resolution of the visual buffer, ability to generate images from ver-
bal descriptions, reorganization of mental images and transformation pro-
cesses such as mental rotation. Kosslyn et al. (1984) first performed a task
analysis of each imagery task to determine which of the processing mod-
ules in his theory of mental imagery (PICTURE, REGENERATE, etc.) were
used in performing the task, and computed a measure of similarity be-
tween pairs of tasks based on the number of these processing modules
that they shared. They then examined the extent to which task similarity
(based on shared processing modules) was related to the correlations be-
tween pairs of tasks. The correlations between the tasks showed a large
range, from −.44 to .79, indicating that the different imagery tasks did not
measure a single undifferentiated ability, but instead measured a number
of different abilities. The strength of these correlations was highly related
to the number of processing modules that they shared according to the
task analysis. Cluster analysis of the data suggested that the tasks could
be grouped according to a small number of processing modules. One clus-
ter of tasks primarily measured the ability to transform and find specific
patterns in images, another primarily involved the ability to maintain im-
ages over time, and a third cluster primarily involved measures of image
resolution.
We now turn to the question of the relation between different dimen-
sions of imagery and spatial abilities. Although this question was not di-
rectly addressed by Kosslyn et al. (1984), one relevant result from that
study was that the Form-Board test (a measure of spatial visualization)
was included in the first cluster of tasks, which measured the ability to

transform and find specific patterns in images. A study by Poltrock and

Brown (1984) assessed the relation between imagery tasks and spatial visu-
alization more directly. They administered nine imagery tasks of the type
studied by Kosslyn (1980) and six tests of spatial visualization ability. Again
the correlations between the imagery tasks were not high, suggesting that
they reflected several different abilities, in contrast with measures of spa-
tial visualization, which defined a single factor. In a regression model, the
imagery tasks that were most predictive of spatial visualization were mea-
sures of speed and accuracy of image rotation, integration of image parts,
and adding detail to images. These are strikingly similar to the measures of
elementary processes proposed in cognitive-components studies of spatial
ability (summarized by Pellegrino & Kail, 1982), suggesting that ability
to accomplish these imagery processes might be the basis of individual
differences in spatial abilities. In contrast, a self-report measure of image
generation ability had a negative regression coefficient, indicating that the
unique variance captured by that variable (the variance that it did not share
with tests of imagery performance) was in fact negatively related to spatial
visualization ability.
In related research, Kozhevnikov and colleagues studied the “visualizer-
verbalizer” cognitive style, a measure of the extent to which people tend
to use visual imagery versus nonvisual methods in cognitive tasks such as
problem solving. Kozhevnikov, Hegarty, and Mayer (2002) identified two
types of visualizers, those with high-spatial ability and those with low-
spatial ability. Kozhevnikov and Kosslyn (2000) found that high-spatial vi-
sualizers are good at spatial imagery tasks such as mental rotation, but poor
at visual imagery tasks such as interpreting degraded pictures, whereas
low-spatial visualizers show the opposite pattern.
In summary, research on the relation between imagery and spatial abil-
ities has suggested that there may be several different aspects of imagery
ability that are differentially related to what is measured by spatial abilities
tests. Although the components of mental imagery ability have been char-
acterized in different ways, there seems to be consensus that the ability to
construct vivid mental images (typically measured by self-report scales) is
distinct from the ability to transform and detect patterns in images (mea-
sured by tests of performance). Vividness of mental imagery appears to be
either unrelated or negatively related to measures of spatial ability, whereas
ability to transform images accurately is positively related to spatial ability.
Other aspects of imagery ability that may or may not be distinct from these
are the ability to maintain images over time and resolution of the imagery
buffer.

Evaluation of Cognitive Analyses of Spatial Abilities

Cognitive psychologists have studied individual differences in spatial abil-
ities within several different research traditions: research on process anal-
ysis of spatial ability test performance, research on working memory, and
research on mental imagery. We have seen that each of these research tra-
ditions has provided important insights into the nature of human spatial
abilities. However, perhaps more striking is the degree of overlap between
the concepts discussed in the different traditions. For example, research
in mental imagery has been concerned with the maintenance and regen-
eration of images over time, which is clearly a working memory function.
In addition, some of the cognitive components of spatial ability perfor-
mance identified in the spatial abilities literature are strikingly similar to
the processing modules identified in Kosslyn’s theory of mental imagery.
Furthermore, research on strategic variation in performance of spatial visu-
alization tasks is consistent with results in the working memory tradition
suggesting that executive processes are implicated in complex spatial vi-
sualization tasks.
An examination of any introductory textbook in cognitive psychology
reveals that spatial abilities, spatial working memory, and visual-spatial
imagery are largely treated as separate topics. The research reviewed in this
section suggests that we can make more rapid progress in understanding
these constructs if we combine insights from the different research tradi-
tions. For example, Kosslyn’s theory of mental imagery might be used to
specify the storage and processing demands made by different strategies
for accomplishing spatial abilities tests. This analysis might provide in-
formation about the extent to which spatial tasks and strategies depend
on image resolution, image maintenance, speed and accuracy of image
transformations, and the interactions between these different factors (e.g.,
a tradeoff between resolution and speed of processing a mental image). It
might also suggest when a task or strategy requires imagery processes to
be augmented by other cognitive processes, such as goal maintenance. In
this way, we might continue to develop a general theory of how individual
differences in visual imagery and other cognitive processes account for
individual differences in spatial cognitive tasks.

environmental spatial abilities

Informal observations suggest that people differ enormously in their abili-
ties to find their way to places, to navigate efficiently, and to form accurate

mental representations of large-scale environments such as buildings, cam-

puses, or cities. We use the term environmental spatial abilities to refer collec-
tively to abilities that require integrating spatial information over time and
across the viewpoint changes associated with self-motion. Unlike abilities
involving stimuli that can be apprehended from a single viewpoint (e.g.,
stimuli on a paper-and-pencil test), environmental knowledge is acquired
over time, as a person explores or navigates an environment. As noted
earlier, psychometricians have historically not assessed these abilities and
as a result, the factor structure of environmental spatial abilities remains
largely unexplored (for more general overview of environmental spatial
skills, see Montello, Chapter 7 on navigation, and Taylor, Chapter 8 on
understanding maps).
In 1975, Siegel and White synthesized much of the then-existing re-
search on large-scale spatial knowledge acquisition and developed what is
currently the dominant model of environmental knowledge. Their model
posits a “main sequence” of representational changes that occur with more
experience in an environment, from knowledge of landmarks, to knowl-
edge of routes, and ultimately to knowledge of configurations. When
viewed as an inevitable developmental sequence, the Siegel and White
model has garnered much criticism (Colle & Reid, 1998; Foley & Cohen,
1984; Hanley & Levine, 1983; McDonald & Pellegrino, 1993; Montello, 1993;
Schmitz, 1997). Yet as a description of different types of mental represen-
tations of space it remains very influential. Although Siegel and White did
not directly address individual differences among adults, it is possible that
these variations among people’s ability to represent or process environ-
mental information might be characterized by abilities to adopt mental
representations of space with more or less sophistication or flexibility.
It is somewhat surprising, then, that of the few multivariate studies of
environmental cognition, none has conclusively upheld the distinctions
made by Siegel and White. In 1986, Lorenz and Neisser reported tentative
evidence for three dimensions of environmental spatial abilities that ap-
peared to be related to the distinctions made by Siegel and White: memory
for landmarks, memory for routes, and awareness of geographical direc-
tions. However, Lorenz was unable to replicate these findings in a subse-
quent larger multivaritate study (Lorenz, 1988). Another large study by
Allen, Kirasic, Dobson, Long, and Beck (1996) showed that environmental
abilities (those required to learn a small model of a town) were distin-
guished primarily by the degree of metric knowledge required to perform
various tasks. Allen et al.’s factor analysis of environmental spatial abil-
ities revealed a “topological knowledge” factor requiring only relational

judgments such as proximity and connectedness. Measures loading on

this factor included scene recognition, scene sequencing, map placement,
intraroute distance error, and route reversal, which reflect landmark and
route knowledge. This factor was distinct from the “Euclidean” factors
that represented the ability to judge metric distances and directions, which
reflect knowledge of configuration.
More recently, Hegarty, Montello, Richardson, and Ishikawa (2004) have
shown that separate environmental spatial abilities may be associated more
with the medium through which an environment is learned (e.g., real
world, video, computer-simulated), or the sensory inputs to that learn-
ing, than with the characteristics or processing demands of the task used
to measure spatial knowledge. This finding makes clear the need to opera-
tionalize the domain of behaviors one wishes to account for, a point echoed
by Allen (1999). It also calls into question the degree to which conclusions
about the factor structure of real-world environmental spatial abilities can
be assessed with tasks involving models or simulations of the environment
(see also McDonald & Pellegrino, 1993). The current status of research on
environmental spatial ability is not unlike that of visuospatial abilities with
researchers in the early 20th century – there is probably too little empirical
evidence to support unequivocal statements about the factor structure of
this ability.
Although the factor structure of environmental spatial abilities remains
poorly understood, investigators are beginning to understand better the re-
lationship between environmental spatial abilities and those abilities mea-
sured by paper-and-pencil psychometric tests. On one hand, the processing
demands of paper-and-pencil tests are clearly relevant to those required
for acquiring large-scale environmental knowledge. Acquiring knowledge
of the environment typically requires a person to encode visual stimuli and
to compare mental images with visual stimuli. Likely, it also requires trans-
formations of spatial information in memory. However, large-scale spatial
abilities require more than this. They require the ability to integrate spatial
information over time and across many views while using several sensory
modalities. Environmental spatial abilities may also be tied more closely to
processes associated with long-term than with short-term memory (Chase
& Chi, 1981). Whereas items on psychometric tests are typically performed
in seconds and place their demands almost exclusively on the capacity or
processing power of working memory, environmental knowledge is typi-
cally acquired over days, weeks, or years (Moeser, 1988). This suggests that
storage and retrieval processes in long-term memory may be relevant to
understanding environmental spatial ability. Because of these differences,

one might expect psychometric measures of spatial ability to be only mildly

predictive of complex environmental abilities like wayfinding, navigation,
and the ability to form a configural representation.
Early studies that focused on this relationship, however, tended to re-
port significant associations between visuospatial abilities (as measured by
psychometric tests) and environmental abilities. In one study, Thorndyke
and Goldin (1983) selected twelve people who performed well on tests of
procedural and survey knowledge of Los Angeles and compared them to
a similar group of twelve who performed poorly on tests of their environ-
mental knowledge. Among a host of other differences, the “good cognitive
mappers” scored significantly higher on tests of spatial visualization, spa-
tial orientation, and visual memory than the “poor cognitive mappers.”
Importantly, scores of verbal abilities were not different between the two
groups, indicating that the spatial elements of the tests were critical com-
ponents of their ability to distinguish between good and poor cognitive
mappers. Thorndyke and Goldin concluded that paper-and-pencil tests of
spatial abilities can be used effectively for the recruitment and selection
of personnel for which cognitive mapping skill is important. These results
have led some to believe that there is a significant relationship between
spatial ability and environmental knowledge.
However, more recent studies that examine this issue closely generally
find tenuous links between large- and small-scale abilities. Table 4.2 sum-
marizes many of the studies that have compared performance on psycho-
metric tests of spatial ability and performance in a large-scale environment.
With very few exceptions, these studies have generally found very little
correlation between the two types of abilities. Correlations as high as .5
were obtained by Walsh et al. (1981); however, these authors do not spec-
ify how many correlations they computed in order to arrive at the two
sizable correlations that they report. Rovine and Weisman (1989) also ob-
tained notable correlations (ca. .4) between an embedded figures test and
various measures of wayfinding behavior. For all other studies, the corre-
lation is closer to .2. In one particularly impressive study, Allen et al. (1996)
used structural equations modeling to show that “higher-level” cognitive
abilities such as spatial-sequential memory and perspective-taking skill
mediate the relationship between laboratory assessments of spatial ability
and real-world assessments of environmental knowledge, but that there
is nearly no direct correlation between spatial ability and environmental
knowledge. The weight of the evidence presented in Table 4.2 leads one
to question whether performance on large-scale and small-scale tasks are
substantively based on any common spatial abilities.

table 4.2. Summary of correlational studies of the relationship between spatial

abilities and environmental knowledge in adults.

Number of Correlations Range of Median

Investigators (year) participants reported correlations correlation
Walsh, Krauss, & Regnier 31 2 .47 to .57 .52
(1981)
Bryant (1982) 85 2 .26 to .39 .33
Goldin & Thorndyke 94 18 ?? to .3 Not
(1982) reported
Meld (1985/1986) 61 8 .03 to .24 .09
Pearson & Ialongo (1986) 353 8 .08 to .30 .18
Lorenz (1988) – (Study 1) 56 24 .01 to .34 .09
Lorenz (1988) – (Study 2) 109 15 .00 to .44 .10
Sholl (1988) – (Study 1) 28 4 .07 to .19 .14
Rovine & Weisman (1989) 45 18 .01 to .43 .21
Allen et al. (1996) – 100 42 .02 to .37 .16
(Original sample)
Allen et al. (1996) – 103 42 .00 to .41 .19
(Validation sample)
Juan-Espinosa et al. (2000) 111 4 .20 to .30 .27
Waller (2000) 151 9 .01 to .34 .20
Kirasic (2001) 240 30 .05 to .39 .22
Hegarty et al. (submitted) 221 12 .11 to .41 .26

Recently, however, Hegarty et al. (2004) have shown a closer relation-

ship between paper-and-pencil assessments of spatial ability and mea-
sures of environmental spatial knowledge than previous research might
suggest. These investigators gave participants a large battery of tests, in-
cluding paper-and-pencil spatial tests, tests of general intelligence, and
self-reported sense of direction. Subsequently, they measured participants’
ability to learn the spatial characteristics of buildings experienced either
in the real world, from a video, or from a desktop computer simulation.
Although zero-order correlations between small-scale and environmen-
tal spatial abilities were generally only around .25, Hegarty et al. used
structural equations modeling to show that when other relevant abilities
were controlled for, and when error variance associated with specific tests
was partialed out, the unique contribution of small-scale spatial ability to
predicting environmental spatial ability was actually quite high (the path
coefficient was .50). Hegarty et al. suggested three reasons for the stronger-
than-expected relationship between paper-and-pencil and environmental
abilities: (1) their measures of environmental spatial abilities represented

ability to acquire knowledge of previously unfamiliar environments; (2)

their measures of environmental spatial abilities were focused primarily
on assessing configural knowledge – the most sophisticated form of envi-
ronmental knowledge; and (3) their measures of small-scale spatial ability
were broader than those typically used by other researchers, including
measures of spatial working memory. These results suggest that the rela-
tionship between psychometrically assessed spatial abilities and environ-
mental spatial abilities may be relatively strong when people are required
to form a sophisticated spatial representation in a relatively short period
of time.
If, in general, paper-and-pencil assessments of visuospatial ability are
not highly predictive of abilities to learn and represent large real-world
spaces, then they may be more closely related to abilities in an increasingly
relevant domain: learning the spatial characteristics of computer-simulated
(“virtual”) environments. Recent findings tend to support a conclusion that
psychometrically assessed spatial abilities are more closely associated with
wayfinding and navigating in virtual environments than with similar ac-
tivities in real-world environments (Hegarty et al. 2004; Juan-Espinosa,
Abad, Colom, & Fernandez-Truchaud, 2000; Moffat, Hampson, & Hatzi-
pantelis, 1998; Waller, 1999; see Wickens, Vincow, & Yeh, Chapter 10, for
more general discussion of navigation in virtual environments). These
studies are summarized in Table 4.3. Although correlations between visu-
ospatial abilities and spatial knowledge of virtual environments rarely ex-
ceed .4, they are typically slightly higher than analogous relationships with
abilities and knowledge of real environments. In one study, for example,

table 4.3. Summary of correlational studies of the relationship between spatial

abilities and spatial knowledge acquisition in virtual environments.

Number of Correlations Range of Median abs.

Investigators (year) participants reported correlations correlation
Bailey (1994) 64 14 .04 to .36 .14
Moffat et al. (1998) 74 6 .34 to .62 .49
Waller (1999) – Darken 10 520 .00 to .81 .24
(1996) data set
Juan-Espinosa et al. 111 2 .26 to .31 .29
(2000)
Waller (2000) 151 9 .19 to .40 .28
Waller et al. (2001) 24 4 .05 to .16 .09
Hegarty et al. 228 12 .17 to .45 .29
(submitted)

Waller (2000) gave people several tests of psychometrically assessed spatial

ability, and then examined their ability to form an accurate representation
of both a large real-world (campus) environment, and a room-sized maze
that they learned in a desktop virtual environment. Performance on the
paper-and-pencil tests was not significantly related to people’s ability to
estimate distances and directions in the real world; however, it did pre-
dict people’s ability to do so in the virtual environment. Waller concluded
that for many people, computer-simulated environments – particularly
those portrayed in desktop systems – are probably perceived more as a
small-scale stimulus with which they interact than as a large-scale envi-
ronment in which they navigate. This may be because desktop displays
do not involve the users’ whole body in the same way that environmen-
tal learning does. Recent studies have shown that performance on spatial
tasks is less accurate when vestibular and kinesthetic information is not
present (Chance, Gaunet, Beall, & Loomis, 1998; Klatzky, Loomis, Beall,
Chance, & Golledge, 1998). In the absence of these cue sources, the users
of desktop virtual environments may need to rely more on the processes
associated with small-scale visuospatial abilities.

functions of spatial abilities

It is undeniable that visuospatial abilities are required for many common
activities. A traveler packing a suitcase for a trip may need to perform an
informal version of a form board task. Telling a lost traveler which direc-
tion to turn may involve the ability to imagine alternative perspectives.
In order to keep track of the locations of his teammates and opponents,
a basketball player needs the ability to reason about movement in the vi-
sual field. While spatial abilities are clearly involved in such activities, and
this conclusion has been supported by correlations of spatial abilities tests
with self-reports of such activities (Groth-Marnat & Teal, 2000; Lunneborg
& Lunneborg, 1984, 1986), the degree to which individual differences in
visuospatial abilities can account for differences in such behaviors is in
general not well known.
Spatial ability is also important in predicting success in various occupa-
tions. This is probably not surprising because, as mentioned earlier, many
tests of spatial ability were developed primarily for the purpose of per-
sonnel selection. It has also been strongly linked to academic success in
mathematics and science. For example sex differences in mathematics per-
formance are often associated with corresponding differences in spatial
abilities (see Halpern & Collaer, Chapter 5). However, what is the basis for

these associations? Do these occupations and academic disciplines require

spatial ability per se, or are the associations based on other factors common
to spatial abilities and occupational or academic success? In this section
we first review some of the findings pertaining to the functions of spatial
ability in occupational and academic success, based on predictive validity
studies. We then explore more precisely the relation between spatial ability
and complex cognition, based on recent work in cognitive psychology. In
this research, cognitive psychologists analyze complex cognitive skills that
characterize occupational and academic success and examine the contribu-
tions of spatial abilities to these skills. We will describe such studies in the
domains of surgery, mechanical reasoning, and mathematics performance.

Spatial Abilities and Occupational Success

For most of the 20th century, the United States Armed Services have se-
lected and classified personnel based on batteries of standardized tests that
include assessments of visuospatial ability. During the middle of the 20th
century, spatial tests were found to be especially predictive of success in
aviation and piloting (Humphreys & Lubinski, 1996). Although the mili-
tary’s use of spatial testing for predicting success in piloting diminished
during the latter half of the last century, subsequent experimental work
has confirmed the association between spatial ability and piloting skill
(Dror, Kosslyn, & Waag, 1993; Gordon & Leighty, 1988). Dror et al. (1993),
for example, showed that pilots were faster than nonpilots at both mental
rotation and at making precise distance judgments.
Spatial ability has also been linked to success in technical training pro-
grams and to job performance in occupations such as engineering drawing,
shop mechanics, auto mechanics, drafting, and designing (Ghiselli, 1973;
Likert & Quasha, 1970; McGee, 1979a). Ghiselli (1973) summarized much
of the research on the associations between spatial ability, job proficiency,
and success in training programs for a number of these occupations and
found that validity coefficients were between .18 and .28 for job perfor-
mance and between .35 and .46 for success in training programs. Although
these correlations would not usually be classified as high, Ghiselli argued
that they suggest moderate to substantial predictive validity, given that the
criterion measures in most cases were supervisors’ ratings, which typically
have low reliability (in the .6 to .8 range). Furthermore, tests with low va-
lidity coefficients can be useful in applied settings, especially when used
to select relatively few valid cases from a large pool of candidates (Hunt,
1995; Taylor & Russell, 1939). Until relatively recently, spatial tests were

regarded as valuable for selection and classification only for occupations

below the professional level (Humphreys & Lubinski, 1996). However, it is
clear that spatial abilities are required for many professional occupations.
As early as 1957, the United States Employment Service listed 84 profes-
sional occupations for which spatial abilities in the top 10 percent of the
working population were required. These occupations included various
types of engineers, scientists, draftsmen, designers, and medical doctors.
For one kind of medical doctor, the surgeon, spatial ability may be in-
creasingly important. Performance in surgery is strongly dependent on
spatial skills. Anatomical environments can be extremely complex, with
intricate 3-D relationships between deformable structures that can vary
from patient to patient. A good surgeon must be able to visualize relation-
ships among tissues and organs that cannot be seen, as well as to construct
3-D mental models from 2-D images (X-ray, CT, and MRI images). Several
studies have shown moderate to high correlations between standardized
tests of spatial ability and performance ratings on a variety of surgical
tasks in open surgery (Gibbons, Gudas, & Gibbons, 1983; Gibbons, Baker, &
Skinner, 1986; Murdoch, Bainbridge, Fisher, & Webster, 1994; Schueneman,
Pickelman, Hesslein, & Freeark, 1984).
While the advent of minimally invasive surgical techniques has im-
proved the quality of care for patients, it has also brought new challenges
to the surgeon. Minimally invasive surgery is performed through cannu-
las, typically 5–10 mm in diameter, inserted through the skin. Long thin
instruments are then inserted through the cannula. This cannula creates a
fulcrum and reduces the degrees of freedom of movement. The surgeon
performs the operation by watching a video image from an endoscope in-
serted through one of the cannulas. This surgical situation gives rise to sev-
eral obvious spatial problems. First, the surgeon must adapt to the changing
orientations between camera and instruments. Second, the camera must
be placed and tissue exposed so that key structures are not obscured and
so that instruments will be effective with their limited degrees of freedom.
Third, complex surgical procedures are carried out by a team of surgeons,
(i.e., a camera operator and assistant in addition to the primary surgeon),
and the team must be able to communicate spatial plans that depend on
shared spatial representations of anatomy.
Tendick et al. (2000) describe a program of research to develop cognitive
models of surgical skills and train these skills using VE simulations. One
representative surgical skill studied by Tendick and his colleagues is the use
of an angled laparoscope, that is, a scope that has the objective lens angled
with respect to its axis. Angled laparoscopes are used in many surgical

procedures because they expand the range of viewing directions for the
surgeon. However, they are difficult to use. In order to view a particular part
of the anatomy, users must use spatial reasoning to place the laparoscope in
the correct location. Tendick et al. developed a VE simulation of an angled
laparoscope. Learning rates of novices using this simulation for the first
time were correlated .39 to .58 with different measures of spatial abilities
(Eyal & Tendick, 2001).

Spatial Abilities and Mechanical Reasoning

As we have seen, the earliest spatial abilities tests were developed to mea-
sure mechanical reasoning and to predict performance in professions such
as machine design, assembly, and repair (Smith, 1964). For many of these
activities, the ability to infer the behavior of a mechanical system from a
representation of the shape and connectivity of the components is critical.
For example, a machine assembler must configure components correctly so
that they will move together to achieve the desired effect. Similarly, some-
one who is repairing a mechanical system might have to first envision how
the components should move in order to locate a faulty component that is,
for example, jamming the system.
Hegarty (1992) used the term mental animation to refer to the ability to
infer the behavior of a mechanical system from its configuration. A task
analysis of the most popular test of mechanical reasoning (Bennett, 1969)
revealed that a large proportion of items on this test involve mental anima-
tion (Hegarty & Sims, 1994). For example, Figure 4.5 shows a typical item
from a test of mechanical comprehension. The item shows the configura-
tion of components in pulley system. The task is to infer which direction
the lower pulley will turn, given that the rope is being pulled.
Hegarty and her colleagues have studied the cognitive processes in-
volved in solving mental animation items. First, Hegarty (1992) showed
that whenever possible, people decompose the depicted mechanical sys-
tem into components and infer the motion of these components piecemeal,
in the order of the causal chain of events in the system. This conclusion
was supported by error data, reaction time data, and measurement of eye
fixations. Further research indicated that mental animation is highly cor-
related with spatial visualization ability (Hegarty & Sims, 1994; Hegarty &
Steinhoff, 1997; Hegarty & Kozhevnikov, 1999) but not with verbal ability,
and that it relies on spatial rather than verbal working memory (Sims &
Hegarty 1997). As predicted, differences in performance between high- and
low-spatial individuals were greater for items in which more mechanical

A B

figure 4.5. Item from a test of mental animation. Examinees are asked the follow-
ing question: “When the rope is pulled, which direction (A or B) will the lower
pulley turn?”

components had to be mentally animated. Low-spatial-individuals made

more errors, but did not differ in their reaction times to solve problems.
Like tests of spatial visualization, mental animation appears to depend
on both strategic differences among individuals and the ability to visualize
complex spatial relationships. For example, one strategy that can be used
on mental animation problems is to offload information onto the external
display, for example by drawing an arrow on each mechanical component
in a diagram as its direction of motion is inferred. This relieves people of
the necessity of maintaining this information in working memory. Hegarty
and Steinhoff (1997) found that when given the opportunity to make notes
on diagrams, only some students did so. These students were able to use a
strategy to compensate for limited spatial working memory resources and
showed improved performance. Other students did not use this strategy
and had poor performance. However, this strategy is not applicable to
all mental animation problems because sometimes several components
constrain each other’s motions and therefore have to be mentally animated
in parallel rather than serially (Hegarty & Kozhevnikov, 1999). Mental
animation of problems that cannot be decomposed is particularly highly

related to spatial ability, presumably because these problems are more

dependent on spatial working memory resources.

Spatial Abilities and Mathematics

From almost the earliest days of intelligence testing, spatial ability has
been considered to be closely related to academic achievement, particularly
to success in mathematics. In addition to general intelligence, mathemat-
ical reasoning is typically thought to require abilities associated with vi-
sual imagery, as well as the ability to perceive, number, and space con-
figurations, and to maintain these representations in memory (Hamley,
1935). Many investigators have assumed that spatial ability is more pre-
dictive of mathematical performance than is verbal ability. There is ex-
tensive research reporting correlations between spatial ability and math-
ematical performance (e.g., Battista, 1990; McGee, 1979b; Sherman, 1979;
Smith, 1964). In a meta-analysis that included 75 studies, Friedman (1995)
found that correlations between spatial and mathematical ability gener-
ally ranged between .3 and .45. Although moderate in size, these correla-
tions suggest a substantial relationship between spatial and mathematical
abilities. However, it is notable that Friedman’s analyses showed a closer
relationship between mathematical and verbal ability, with correlations in
the .4 to .5 range.
Perhaps more interesting are findings that relate the specific processes
required for success in mathematics with spatial ability. For example, re-
cent research has studied the relation between people’s cognitive style and
how they represent problems in physics and mathematics. One impor-
tant distinction from this literature is the visualizer-verbalizer cognitive
style (Hegarty & Kozhevnikov, 1999; Kozhevnikov, Hegarty, & Mayer, 2002;
Lean & Clements, 1981; Presmeg, 1992). This reflects the extent to which
people use visual-spatial representations (images or diagrams) while solv-
ing problems that can be solved by either visual-spatial or more abstract
analytical methods (Krutetskii, 1976). A puzzling result in the educational
literature is that although spatial ability is correlated with performance
in mathematics and physics, visualizers are not more successful at solv-
ing problems in these domains (Lean & Clements, 1981; Presmeg, 1992).
Kozhevnikov, Hegarty, and Mayer (2002) showed that there are actually
two types of visualizers; those with high-spatial ability and those with
low-spatial ability. In one study conducted with elementary school stu-
dents, high-spatial visualizers were more likely to construct diagrams or
schematic spatial representations of the spatial relations between objects

described in a problem, whereas low-spatial visualizers tended to con-

struct pictorial images of the objects described in the problems (Hegarty
& Kozhevnikov, 1999). For example, one problem described a man plant-
ing trees 5 meters apart on a path. Students were given the total length
of the path and had to find the number of trees planted. High-spatial vi-
sualizers drew a diagram of the path, marking the position of trees on
the path and the distance between them, whereas low-spatial visualizers
reported a mental image of the man planting trees. Consequently, high-
spatial visualizers were more successful in problem solving. In another
study, Kozhevnikov, Hegarty, and Mayer (2002) presented college students
with graphs of motion showing either the position, velocity, or accelera-
tion of objects over time. Low-spatial visualizers tended to interpret the
graphs as pictures, whereas high-spatial visualizers interpreted them cor-
rectly as abstract representations. These findings show that spatial ability is
important for both constructing and comprehending abstract spatial repre-
sentations in mathematical problem solving. Further discussion of the role
of spatial thinking and mathematics can be found in Halpern and Collaer,
Chapter 5.

conclusions
The research that we have reviewed in this chapter has provided strong ev-
idence that spatial ability is differentiated from general intelligence. It has
also shown that spatial ability is not a single, undifferentiated construct,
but instead is composed of several somewhat separate abilities, such as
spatial visualization, flexibility of closure, spatial memory, and perceptual
speed. These distinctions have evolved over decades of ongoing research.
Unfortunately during much of the history of this research, the treatment
of individual differences, the development of spatial abilities tests, and
the design of factor analysis studies have not been systematic. Tests were
developed primarily as predictors of real-world cognitive tasks and un-
til recently, vitually all factor-analytic studies have been exploratory. As a
result, research on the structure of spatial abilities has been largely atheo-
retical and has all but ignored individual differences in important aspects
of spatial cognition, such as the processing of dynamic spatial displays and
navigation in large-scale space. There is a need for a theoretical framework
that outlines the space of different cognitive abilities associated with repre-
senting and processing spatial information. This framework would enable
us to examine the status of current spatial abilities research, and the types
of spatial abilities tests that need to be developed.

Research on the cognitive analysis of existing spatial tests can provide

important input to this theoretical framework. Our review of this litera-
ture has shown that different spatial abilities may depend more or less on
speed of processing, strategies, quality of spatial images, active mainte-
nance of spatial information, and central executive processes. Cognitive
research on spatial abilities has suffered from the fact that constructs such
as working memory and imagery have typically been treated separately
in the cognitive literature. Despite these constructs belonging to separate
research traditions, we have seen that they are actually closely related to
spatial processing. These constructs might be related in defining a theoret-
ical framework for characterizing spatial abilities.
Most research on spatial ability to date has focused on tasks that re-
quire the mental manipulaton of small objects and that can be presented
in paper-and-pencil tests. Research on large-scale spatial abilities is in its
infancy. It is important that this relatively new research endeavor does not
repeat the mistakes we have seen in the history of research on small-scale
spatial abilities. For example, it is important to guide research by theory.
This means considering up front what the dimensions are that define the
space of large-scale spatial abilities and testing these models (see also Allen,
1999). Even more important is adherence to a set of principles and analytic
methods that allow theories to be disconfirmed. Studies to date have sug-
gested some of these dimensions, for example (1) a distinction between
ability to learn a new environment versus wayfinding and navigation in a
known environment, (2) a distinction between tasks that measure knowl-
edge of landmarks versus routes or configurations, and (3) the nature of the
processes and representations associated with different forms of learning.
Research on spatial ability has traditionally been very influenced by
practical concerns, such as personnel selection, and there is a large body
of literature showing strong relations between spatial abilities and vari-
ous professions such as engineering, aircraft piloting, and surgery. Recent
research has also begun to analyze complex tasks involved in these pro-
fessions in terms of their demand on spatial skills. This is an important
research direction that can elucidate how spatial information processing is
involved in complex cognition and may be important in training complex
spatial skills.

acknowledgments
This research was conducted while David Waller was at the Univer-
sity of California, Santa Barbara, and was supported by grants 9873432

and 9980122 from the National Science Foundation and N00014-96-10525

from the Office of Naval Research.

Suggestions for Further Readings

The following papers provide additional information on factor analytic studies of
spatial ability.
This is an excellent historical introduction to spatial abilities testing and early factor
analytic studies:
Smith, I. F. (1964). Spatial ability: Its educational and social significance. San Diego:
Knapp.
Chapter 4: of this book presents a reanalysis of over 90 data sets that bear on the
factor structure of spatial ability:
Carroll, J. (1993). Human cognitive abilities: A survey of factor-analytical studies. New
York: Cambridge University Press.
This book also offers a thorough review of the factor analytic literature, and con-
siders the environmental and biological determinants of individual differences
in spatial abilities, with particular reference to those associated with sex diff-
erences:
McGee, M. G. (1979a). Human spatial abilities: Sources of sex differences. New York:
Praeger.
The following papers offer excellent summaries of the literature on cognitive anal-
yses of performance on spatial abilities tests. Lohman (1988) also summarizes
some preliminary studies concerned with the training of spatial abilities:
Lohman, D. F. (1988). Spatial abilities as traits, processes, and knowledge. In R.
J. Sternberg (Ed.), Advances in the psychology of human intelligence (pp. 181–248).
Hillsdale, NJ: Erlbaum.
Pellegrino, J. W., & Kail, R. V. (1982). Process analyses of spatial aptitude. In R. J.
Sternberg (Ed.), Advances in the psychology of human intelligence (Vol 1, pp. 311–
365). Hillsdale, NJ: Erlbaum.
The following papers are good introductions to the study of environmental spatial
cognition from individual differences and other perspectives:
Allen, G. L. (1999). Spatial abilities, cognitive maps and wayfinding: Bases for in-
dividual differences in spatial cognition and behavior. In R. G. Golledge (Ed.),
Wayfinding behavior: Cognitive mapping and other spatial processes (pp. 46–80). Bal-
timore, MD: Johns Hopkins Press.
McDonald, T. P., & Pellegrino, J. W. (1993). Psychological perspectives on spatial
cognition. In T. Garling & R. G. Golledge (Eds.), Behavior and environment: Psy-
chological and geographical approaches. Amsterdam: Elsevier Science Publishers.
This short paper summarizes much of the literature on the relations of spatial
abilities to performance in technical occupations and training programs:
Ghiselli, E. E. (1973). The validity of aptitude tests in personnel selection. Personnel
Psychology, 26, 461–477.

References
Allen, G. L. (1999). Spatial abilities, cognitive maps and wayfinding: Bases for indi-
vidual differences in spatial cognition and behavior. In R. G. Golledge (Ed.),
Wayfinding behavior: Cognitive mapping and other spatial processes (pp. 46–80).
Baltimore, MD: Johns Hopkins Press.
Allen, G. L., Kirasic, K., Dobson, S. H., Long, R. G., & Beck, S. (1996). Predicting
environmental learning from spatial abilities: An indirect route. Intelligence, 22,
327–355.
Baddeley, A. D. (1986). Working memory. New York: Oxford University Press.
Bailey, J. H. (1994). Spatial knowledge acquisition in a virtual environment (Doc-
toral dissertation, University of Central Florida). Dissertation Abstracts Interna-
tional, 55(6-B), 2421.
Barratt, E. S. (1953). An analysis of verbal reports of solving spatial problems as an
aid in defining spatial factors. Journal of Psychology, 36, 17–25.
Battista, M. T. (1990). Spatial visualization and gender differences in high school
geometry. Journal for Research in Mathematics Education, 21, 47–60.
Bennett, C. K. (1969). Bennett mechanical comprehension test. San Antonio, TX: The
Psychological Corporation.
Betts, G. H. (1909). The distribution and functions of mental imagery. New York:
Columbia University Press.
Binet, A., & Simon, T. (1905). Methodes nouvelles put le diagnostic du niveau
intellectuel des anormaux. L’Annee’ Psychologique, 11, 191–244.
Borich, G. D., & Bauman, P. M. (1972). Convergent and discriminant validation of
the French and Guilford–Zimmerman spatial orientation and spatial visualiza-
tion factors. In Proceedings of the Annual Convention of the American Psychological
Association. Vol. 7 (pt. 1) (pp. 9–10).
Bryant, K. J. (1982). Personality correlates of sense of direction and geographical
orientation. Journal of Personality and Social Psychology, 43, 1318–1324.
Carpenter, P. A., & Just, M. A. (1986). Spatial ability: An information-processing
approach to psychometrics. In R. J. Sternberg (Ed.), Advances in the psychology of
human intelligence (Vol. 3, pp. 221–252). Hillsdale, NJ: Erlbaum.
Carpenter, P. A., Just, M. A., Keller, T. A., Eddy, W. F., & Thulborn, K. R. (1999).
Graded functional activation in the visuospatial system with the amount of task
demand. Journal of Cognitive Neuroscience, 11, 9–24.
Carroll, J. (1993). Human cognitive abilities: A survey of factor-analytical studies. New
York: Cambridge University Press.
Chance, S. S., Gaunet, F., Beall, A. C., & Loomis, J. M. (1998). Locomotion mode
affects the updating of objects encountered during travel: The contribution of
vestibular and proprioceptive inputs to path integration. Presence: Teleoperators
and Virtual Environments, 7(2), 168–17.
Chase, W. G., & Chi, M. T. H. (1981). Cognitive skill: Implications for spatial skill
in large-scale environments. In J. H. Harvey (Ed.), Cognition, social behavior, and
the environment (pp. 111–135). Hillsdale, NJ: Erlbaum.
Colle, H. A., & Reid, G. B. (1998). The room effect: Metric spatial knowledge of local
and separated regions. Presence: Teleoperators and Virtual Environments, 7, 116–128.
Cowan, N. (1995). Attention and memory: An integrated framework. New York: Oxford
University Press.

Cronbach, L. J. (1957). The two disciplines of scientific psychology. American Psy-

chologist, 12, 671–684.
Daneman, M., & Carpenter, P. A. (1980). Individual differences in working memory
and reading. Journal of Verbal Learning and Verbal Behavior, 19, 450–466.
Darken, R. P. (1996). Wayfinding in large-scale virtual worlds. Unpublished doctoral
dissertation, George Washington University, Department of Electrical Engineer-
ing and Computer Science, Washington, DC.
Dror, I. E., Kosslyn, S. M., & Waag, W. L. (1993). Visual-spatial abilities of pilots.
Journal of Applied Psychology, 78, 763–773.
Ekstrom, R. B., French, J. W., Harman, H. H., & Dermen, D. (1976). Kit of Factor
Referenced Cognitive Tests. Princeton, NJ: Educational Testing Service.
El Koussy, A. A. H. (1935), An investigation into the factors involving the visual
perception of space. British Journal of Psychology, 20, 92.
Engle, R. W., Kane, M. J., & Tuholski, S. W. (1999). Individual differences in working
memory capacity and what they tell us about controlled attention, general fluid
intelligence and functions of the prefrontal cortex. In A. Miyake & P. Shah (Eds.),
Models of working memory: Mechanisms of active maintenance and executive control
(pp. 102–134). New York: Cambridge University Press.
Eyal, R., & Tendick, F. (January, 2001). Spatial ability and learning the use of an
angled laparoscope in a virtual environment. In J. D. Westwood et al., (Eds.),
Medicine Meets Virtual Reality 2001, IOS Press, Amsterdam, pp. 146–152.
Eysenk, H. J. (1939). Review of “primary mental abilities” by L. L. Thurstone. British
Journal of Psychology, 9, 270–275.
Farah, M. J., Hammond, K. M., Levine, D. N., & Calvanio, R. (1988). Visual and
spatial memory: Dissociable systems of representation. Cognitive Psychology, 20,
439–462.
Fischer, S. C., Hickey, D. T., Pellegrino, J. W., & Law, D. J. (1994). Strategic processing
in dynamic spatial reasoning tasks. Learning and Individual Differences, 6, 65–105.
Foley, J. E., & Cohen, A. J. (1984). Mental mapping of a megastructure. Canadian
Journal of Psychology, 38, 440–453.
French, J. W. (1951). Description of aptitude and achievement tests in terms of
rotated factors. Psychometric Monograph, 5.
Friedman, L. (1995). The space factor in mathematics: Gender differences. Review
of Educational Research, 65, 22–50.
Galton, F. (1883). Inquiries into human faculty and its development. London:
Macmillan.
Ghiselli, E. E. (1973). The validity of aptitude tests in personnel selection. Personnel
Psychology, 26, 461–477.
Gibbons, R., Baker, R., & Skinner, D. (1986). Field articulation testing: A predictor
of technical skills in surgical residents. Journal of Surgical Research, 41, 53–57.
Gibbons, R., Gudas, C., & Gibbons, S. (1983). A study of the relationship between
flexibility of closure and surgical skill. Journal of the American Podiatric Medical
Association, 73, 12–16.
Gibson, J. J. (1947). Army air forces aviation psychology program, report no. 7: Motion
picture testing and research. Washington, DC: U.S. Government Printing Office.
Goldberg, J., & Meredith, W. (1975). A longitudinal study of spatial ability. Behavior
Genetics, 5, 127–135.

Goldin, S. E., & Thorndyke, P. W. (1982). Simulating navigation for spatial knowl-
edge acquisition. Human Factors, 24, 457–471.
Gordon, H. W., & Leighty, R. (1988). Importance of specialized cognitive function
in the selection of military pilots. Journal of Applied Psychology, 73, 38–45.
Groth-Marnat, G., & Teal, M. (2000). Block design as a measure of everyday spatial
ability: A study of ecological validity. Perceptual and Motor Skills, 90, 522–526.
Guilford, J. P., & Lacey, J. I. (1947). Printed Classification Tests, A.A.F. Aviation Psycho-
logical Progress Research Report, No. 5, Washington, DC: U.S. Government Printing
Office.
Guilford, J. P., & Zimmerman, W. S. (1948). The Guilford-Zimmerman Aptitude
Survey. Journal of Applied Psychology, 32, 24–34.
Gustafsson, J. E. (1984). A unifying model for the structure of intellectual abilities.
Intelligence, 8, 179–203.
Hamley, H. R. (1935). The testing of intelligence. London: Evans.
Hanley, G. L., & Levine, M. (1983). Spatial problem solving: The integration of
independently learned cognitive maps. Memory and Cognition, 11, 415–522.
Hegarty, M. (1992). Mental animation: Inferring motion from static diagrams of
mechanical systems. Journal of Experimental Psychology: Learning, Memory and Cog-
nition, 18, 1084–1102.
Hegarty, M. Montello, D. R., Richardson, A. E., Ishikawa T., & Lovelace, K. (2004).
Spatial abilities at different scales: Individual differences in aptitude-test perfor-
mance and Spatial-layout learning. Manuscript submitted for publication.
Hegarty M., & Kozhevnikov, M. (1999). Spatial abilities, working memory and
mechanical reasoning. In J. Gero & B. Tversky (Eds.), Visual and spatial reasoning
in design. Sydney, Australia: Key Centre of Design and Cognition.
Hegarty, M., Shah, P., & Miyake, A. (2000). Constraints on using the dual-task
methodology to specify the degree of central executive involvement in cognitive
tasks. Memory and Cognition, 28, 376–385.
Hegarty, M., & Sims, V. K. (1994). Individual differences in mental animation during
mechanical reasoning. Memory and Cognition, 22, 411–430.
Hegarty, M., & Steinhoff, K. (1997). Individual differences in use of diagrams as
external memory in mechanical reasoning. Learning and Individual Differences, 9,
19–42.
Hegarty, M., & Waller, D. (2004). A dissociation between mental Rotation and
perspective-taking abilities. Intelligence, 32, 175–191.
Humphreys, L. G., & Lubinski, D. (1996). Assessing spatial visualization: An un-
derappreciated ability for many school and work settings. In C. P. Benbow &
D. Lubinski (Eds.), Intellectual talent: Psychometric and social issues, (pp. 116–140).
Baltimore, MD: John Hopkins Press.
Hunt, E. (1995). Will we be smart enough? New York: Sage.
Hunt, E., Pellegrino, J. W., Frick, R. W., Farr, S. A., & Alderton, D. (1988). The ability
to reason about movement in the visual field. Intelligence, 12, 77–100.
Jones, M. B., Dunlap, W. P., & Bilodeau, I. M. (1986). Comparison of video game
and conventional test performance. Simulation and Games, 17, 435–446.
Juan-Espinosa, M., Abad, F. J., Colom, R., & Fernandez-Truchaud, M. (2000). In-
dividual differences in large-spaces orientation: g and beyond? Personality and
Individual Differences, 29, 85–98.

Just, M. A., & Carpenter, P. A. (1985). Cognitive coordinate systems: Accounts of

mental rotation and individual differences in spatial ability. Psychological Review,
92, 137–172.
Kelly, T. L. (1928). Crossroads in the mind of man. Stanford: Stanford University
Press.
Kirasic, K. C. (2001). Age differences in adults’ spatial abilities, learning environ-
mental layout, and wayfinding behavior. Spatial Cognition and Computation, 2,
117–134.
Klatzky, R. L., Loomis, J. M., Beall, A. C., Chance, S. S., & Golledge, R. G. (1998). Spa-
tial updating of self-position and orientation during real, imagined, and virtual
locomotion. Psychological Science, 9, 293–298.
Kosslyn, S. M. (1980). Image and mind. Cambridge, MA: Harvard University
Press.
Kosslyn, S. M. (1995). Mental imagery. In S. M. Kosslyn & D. N. Osherson (Eds.),
Visual cognition: An invitation to cognitive science, Vol. 2 (pp. 267–296), Cambridge,
MA: MIT Press.
Kosslyn, S. M., Brunn, J., Cave, K. R., & Wallach, R. W. (1984). Individual differences
in mental imagery ability: A computational analysis. In S. Pinker (Ed.), Visual
cognition (pp. 195–244), Cambridge, MA: MIT Press.
Kosslyn, S. M., & Schwartz, S. P. (1977). A simulation of visual imagery, Cognitive
Science, 1, 262–295.
El Koussy, A. A. H. (1935). An investigation in to the factors involving the visual
perception of space. British Journal of Psychology, 20, 92.
Kozhevnikov, M., Hegarty, M. & Mayer (2002). Revising the visualizer/verbalizer
dimension: Evidence for two types of visualizers. Cognition and Instruction, 20,
47–77.
Kozhevnikov, M., & Hegarty, M. (2001). A dissociation between object-
manipulation spatial ability and spatial orientation abilities. Memory and Cog-
nition, 29, 745–756.
Kozhevnikov, M., & Kosslyn, S. M. (2000). Revising cognitive style dimension:
Evidence for two types of visualizers. Abstracts of the Psychonomic Society, 5, 242.
Krutetskii, V. A. (1976). The psychology of mathematical abilities in schoolchildren.
Chicago: University of Chicago Press.
Kyllonen, P. C., Lohman, D. F., & Snow, R. E. (1984). Effects of aptitudes, strat-
egy training and task facets on spatial task performance. Journal of Educational
Psychology, 74, 130–145.
Kyllonen, P. C., Lohman, D. F., & Woltz, D. J. (1984). Componential modeling of al-
ternative strategies for performing spatial tasks. Journal of Educational Psychology,
6, 1325–1345.
Larson, G. E. (1996). Mental rotation of static and dynamic figures. Perception and
Psychophysics, 58, 153–159.
Law, D. J., Pellegrino, J. W., Mitchell, S. R., Fischer, S. C., McDonald, T. P., & Hunt,
E. B. (1993). Perceptual and cognitive factors governing performance in compara-
tive arrival-time judgments. Journal of Experimental Psychology: Human Perception
and Performance, 19, 1183–1199.
Lean, C., & Clements, M. A. (1981). Spatial ability, visual imagery, and mathematical
performance. Educational Studies in Mathematics, 12, 267–299.

Likert, R., & Quasha, W. H. (1970). Revised Minnesota paper form board test. New
York: The Psychological Corporation.
Logie, R. H. (1995). Visuospatial working memory. Hove, UK: Erlbaum.
Lohman, D. F. (1979). Spatial ability: Review and re-analysis of the correlational
literature. Tech. Rep. No. 8, Stanford University.
Lohman, D. F. (1986). The effect of speed-accuracy tradeoff on sex differences in
mental rotation. Perception and Psychophysics, 39, 427–436.
Lohman, D. F. (1988). Spatial abilities as traits, processes, and knowledge. In
R. J. Sternberg (Ed.), Advances in the psychology of human intelligence (pp. 181–
248). Hillsdale, NJ: Erlbaum.
Lohman, D. F. (1996). Spatial ability and g. In I. Dennis & P. Tapsfield (Eds.),
Human abilities: Their nature and measurement (pp. 97–116). Mahwah, NJ:
Erlbaum.
Lorenz, C. A. (1988). The structure of the spatial domain: An analysis of indi-
vidual differences. (Doctoral dissertation, Cornell University, 1988). Dissertation
Abstracts International, 49(06 – B), 2400.
Lorenz, C. A., & Neisser, U. (1986). Ecological and psychometric dimensions of
spatial ability Tech. Rep. No. 10, Emory Cognition Project, Emory University,
Atlanta, GA.
Lunneborg, C. E., & Lunneborg, P. W. (1984). Contribution of sex-differentiated
experiences to spatial and mechanical reasoning abilities. Perceptual and Motor
Skills, 59, 107–113.
Lunneborg, C. E., & Lunneborg, P. W. (1986). Everyday spatial activities test for
studying differential spatial experience and vocational behavior. Journal of Voca-
tional Behavior, 28, 135–141.
Marks, D. F. (1973). Visual imagery differences and eye movements in the recall of
pictures. Perception and Psychophysics, 14, 407–412.
Marshalek, B., Lohman, D. F., & Snow, R. E. (1983). The complexity continuum in
the radex and hierarchical models of intelligence. Intelligence, 7, 107–127.
McDonald, T. P., & Pellegrino, J. W. (1993). Psychological perspectives on spatial
cognition. In T. Garling & R. G. Golledge (Eds.), Behavior and environment: Psy-
chological and geographical approaches (pp. 47–82). Amsterdam: Elsevier Science
Publishers.
McGee, M. G. (1979a). Human spatial abilities: Sources of sex differences. New York:
Praeger.
McGee, M. G. (1979b). Human spatial abilities: Psychometric studies and environ-
mental, genetic, hormonal, and neurological influences. Psychological Bulletin, 86,
889–918.
Meld, A. (1985). Map- and route-learning strategies in spatial cognition: A compar-
ison of survey and procedural knowledge. (Doctoral dissertation, University of
Washington, 1986). Dissertation Abstracts International, 46(7-A), 1880.
Michael, W. B., Guilford, J. P., Fruchter, B., & Zimmerman, W. S. (1957). The descrip-
tion of spatial-visualization abilities. Education and Psychological Measurement, 17,
185–199.
Miyake, A., & Shah, P. (Eds.) (1999). Models of working memory. Cambridge: Cam-
bridge University Press.

Miyake, A., Rettinger, D. A., Friedman, N. P., Shah, P., & Hegarty, M. (2001). Visu-
ospatial working memory, executive functioning and spatial abilities. How are
they related? Journal of Experimental Psychology: General, 130, 621–640.
Moeser, S. D. (1988). Cognitive mapping in a complex building. Environment and
Behavior, 20, 21–49.
Moffat, S. D., Hampson, E., & Hatzipantelis, M. (1998). Navigation in a “virtual”
maze: Sex differences and correlation with psychometric measures of spatial
ability in humans. Evolution and Human Behavior, 19, 73–87.
Montello, D. R. (1993). Scale and multiple psychologies of space. In A. U. Frank &
I. Campari (Eds.), Spatial information theory: A theoretical basis for GIS (pp. 312–
321). Proceedings of COSIT ’93. Berlin: Springer-Verlag.
Montello, D. R., Lovelace, K. L., Golledge, R. G., & Self, C. M. (1999). Sex-related
differences and similarities in geographic and environmental spatial abilities.
Annals of the Association of American Geographers, 89, 515–534.
Mumaw, R. J., & Pellegrino, J. W. (1984). Individual differences in complex spatial
processing. Journal of Educational Psychology, 76, 920–939.
Mumaw, R. J., Pellegrino, J. W., Kail, R. V., & Carter, P. (1984). Different slopes for
different folks: Process analysis of spatial aptitude. Memory and Cognition, 12,
515–521.
Murdoch, J. R., Bainbridge, L. C., Fisher, S. G., & Webster, M. H. C. (1994). Can
a simple test of visual motor skill predict the performance of microsurgeons?
Journal of the Royal College of Surgeons, Edinburgh, 39, 150–152.
Pearson, J. L., & Ialongo, N. S. (1986). The relationship between spatial ability and
environmental knowledge. Journal of Environmental Psychology, 6, 299–304.
Pellegrino, J. W., & Kail, R. V. (1982). Process analyses of spatial aptitude. In R. J.
Sternberg (Ed.), Advances in the psychology of human intelligence (Vol 1, pp. 311–
365). Hillsdale, NJ: Erlbaum.
Poltrock, S. E., & Agnoli, F. (1986). Are spatial visualization ability and visual im-
agery ability equivalent? In R. J. Stenberg (Ed.), Advances in the psychology of
human intelligence (pp. 255–296). Hillsdale, NJ: Erlbaum.
Poltrock, S. E., & Brown, P. (1984). Individual differences in visual imagery and
spatial ability. Intelligence, 8, 93–138.
Presmeg, N. C. (1992). Prototypes, metaphors, metonymies, and imaginative ratio-
nality in high school mathematics. Educational Studies in Mathematics, 23, 595–610.
Robert, M., & Ecuyer-Dab, I. (November, 2000). Do women excel in remembering the
locations of both stationary and repositioned objects? Poster session presented at the
41st annual meeting of the Psychonomics Society, New Orleans, LA.
Roff, M. (1952). A factorial study of tests in the perceptual area. Psychometric
Monograph No. 8.
Rovine, M. J., & Weisman, G. D. (1989). Sketch-map variables as predictors of way-
finding performance. Journal of Environmental Psychology, 9, 217–232.
Salthouse, T. A. (1996). The processing-speed theory of adult age differences in
cognition. Psychological Review, 103, 403–428.
Schmitz, S. (1997). Gender-related strategies in environmental development: Effects
of anxiety on wayfinding in and representation of a three-dimensional maze.
Journal of Environmental Psychology, 17, 215–228.

Schueneman, A., Pickelman, J., Hesslein, R., & Freeark, R. (1984). Neuropsychologic
predictors of operative skill among general surgery residents. Surgery, 96, 288–
295.
Seibert, W. F. (1970). The motion picture as a psychological testing medium: A look
at cine-psychometrics. Viewpoints, 46(5), 91–102.
Shah, P., & Miyake, A. (1996). The separability of working memory resources for
spatial thinking and language processing: An individual differences approach.
Journal of Experimental Psychology: General, 125, 4–27.
Shepard, R. N., & Metzler, J. (1971). Mental rotation of three-dimensional objects.
Science, 171, 701–703.
Sherman, J. (1979). Predicting mathematics performance in high school girls and
boys. Journal of Educational Psychology, 71, 242–249.
Sholl, M. J. (1988). The relationship between sense of direction and mental geo-
graphic updating. Intelligence, 12, 299–314.
Siegel, A. W., & White, S. H. (1975). The development of spatial representations of
large-scale spatial environments. In H. W. Reese (Ed.), Advances in child develop-
ment and behavior. Vol. 10, pp. 9–55. New York: Academic Press.
Silverman, I., & Eals, M. (1992). Differences in spatial abilities: Evolutionary theory
and data. In J. H. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evo-
lutionary psychology and the generation of culture (pp. 533–549). New York: Oxford
University Press.
Sims, V. K., & Hegarty, M. (1997). Mental animation in the visual-spatial sketchpad:
Evidence from dual-task studies. Memory and Cognition, 25, 321–332.
Smith, I. F. (1964). Spatial ability: Its educational and social significance. San Diego:
Knapp.
Sternberg, R. J. (1977). Intelligence, information processing and analogical reasoning.
Hillsdale, NJ: Erlbaum.
Sternberg, R. J. (1981). Nothing fails like success: The search for an intelligent
paradigm for studying intelligence. Journal of Educational Psychology, 73, 142–155.
Taylor, H. C., & Russell, J. T. (1939). The relationship of validity coefficients to
the practical effectiveness of tests in selection: Discussions and tables. Journal of
Applied Psychology, 23, 565–578.
Tendick, F., Downes, M., Gogtekin, T., Cavusoglu, M. C., Feygin, D., Wu, X., Eyal,
R., Hegarty, M., & Way, L. W. (2000). A virtual testbed for training laparoscopic
surgical skills. Presence, 9, 236–255.
Thorndyke, P. W., & Goldin, S. E. (1983). Spatial learning and reasoning skill. In
H. L. Pick & L. P. Acredolo (Eds.), Spatial orientation: Theory, research, and application
(pp. 195–217). New York: Plenum.
Thurstone, L. L. (1938). Primary mental abilities. Chicago: University of Chicago
Press.
Thurstone, L. L. (1950). Some primary abilities in visual thinking (Tech. Rep. No. 59).
Chicago: University of Chicago, Psychometric Laboratory.
Thurstone, T. G. (1958). Manual for the SRA primary mental abilities. Chicago: Science
Research Associates.
Vincent, W. J., & Allmandinger, M. F. (1971). Relationships among selected tests of
spatial orientation ability. Journal of Motor Behavior, 3(3), 259–264.

Waller, D. (1999). An assessment of individual differences in spatial knowledge of

real and virtual environments (Doctoral dissertation, University of Washington,
1999). Dissertation Abstracts International, 60, 1882B.
Waller, D. (2000). Individual differences in spatial learning from computer-
simulated environments. Journal of Experimental Psychology: Applied, 6, 307–321.
Waller, D., Knapp, D., & Hunt, E. (2001). Spatial representations of virtual mazes:
The role of visual fidelity and individual differences. Human Factors, 43, 147–158.
Walsh, D. A., Krauss, I. K., & Regnier, V. A. (1981). Spatial ability, environmental
knowledge, and environmental use. In L. Liben, A. Patterson, & N. Newcombe
(Eds.), Spatial representation and behavior across the life span: Theory and application
(pp. 321–357). New York: Academic Press.
Zimmerman, W. S. (1954). Hypotheses concerning the nature of the spatial factors.
Educational and Psychological Measurement, 14, 396–400.

Spatial Ability: A Neglected Talent in Educational and Occupational Settings
No ratings yet
Spatial Ability: A Neglected Talent in Educational and Occupational Settings
12 pages
10 21449-Ijate 1102435-2370110
No ratings yet
10 21449-Ijate 1102435-2370110
22 pages
Ramful2016 3 Indikator Spasial
No ratings yet
Ramful2016 3 Indikator Spasial
19 pages
Spatial Cognition Development Analysis
No ratings yet
Spatial Cognition Development Analysis
11 pages
Measurement of Spatial Abilities History and Background
No ratings yet
Measurement of Spatial Abilities History and Background
9 pages
Bibliography: Intelligences. New York: Basic Books, 1985
No ratings yet
Bibliography: Intelligences. New York: Basic Books, 1985
6 pages
Spatial Visualization in Physics Problem Solving: Maria Kozhevnikov, Michael A. Motes, Mary Hegarty
No ratings yet
Spatial Visualization in Physics Problem Solving: Maria Kozhevnikov, Michael A. Motes, Mary Hegarty
31 pages
978 1 5275 3310 3 Sample
No ratings yet
978 1 5275 3310 3 Sample
30 pages
Coutrot Et Al 2018 Current Biology
No ratings yet
Coutrot Et Al 2018 Current Biology
11 pages
Psychology 8a Module III
No ratings yet
Psychology 8a Module III
14 pages
Which Cognitive Abilities Make The Difference? Predicting Academic Achievements in Advanced STEM Studies
No ratings yet
Which Cognitive Abilities Make The Difference? Predicting Academic Achievements in Advanced STEM Studies
24 pages
s10648 012 9215 X
No ratings yet
s10648 012 9215 X
26 pages
Linn & Petersen (1985) PDF
No ratings yet
Linn & Petersen (1985) PDF
21 pages
Applications of Test
No ratings yet
Applications of Test
53 pages
Vehicles of Spatial Ability Colom 2001
No ratings yet
Vehicles of Spatial Ability Colom 2001
10 pages
Developmental Science - 2014 - Tosto - Why Do Spatial Abilities Predict Mathematical Performance
No ratings yet
Developmental Science - 2014 - Tosto - Why Do Spatial Abilities Predict Mathematical Performance
9 pages
Spatial Skills in Architecture Exams
No ratings yet
Spatial Skills in Architecture Exams
14 pages
Environmental Spatial Ability Scale
No ratings yet
Environmental Spatial Ability Scale
23 pages
A Systemic Literature Riview of Spatial Abilities and Mathmatics Problem Solving
No ratings yet
A Systemic Literature Riview of Spatial Abilities and Mathmatics Problem Solving
4 pages
Bon Accio 2006
No ratings yet
Bon Accio 2006
15 pages
Improving Spatial Ability Using A Web-Based Virtual Environment (Wbve)
No ratings yet
Improving Spatial Ability Using A Web-Based Virtual Environment (Wbve)
9 pages
Visual-Spatial Ability Important in STEM, Ignored in Gifted Education
No ratings yet
Visual-Spatial Ability Important in STEM, Ignored in Gifted Education
24 pages
Psychology Unitiiiaptitude 200504123026
No ratings yet
Psychology Unitiiiaptitude 200504123026
31 pages
Monographs Society Res Child - 2017 - Verdine - V Results Individual Difference Factors in Spatial and Mathematical Skills
No ratings yet
Monographs Society Res Child - 2017 - Verdine - V Results Individual Difference Factors in Spatial and Mathematical Skills
8 pages
Presentation 9
No ratings yet
Presentation 9
20 pages
Gender Differences in Spatial Orientation:a Review: Emanuele Coluccia, Giorgia Louse
No ratings yet
Gender Differences in Spatial Orientation:a Review: Emanuele Coluccia, Giorgia Louse
12 pages
Ability and Sex Differences in Spatial Thinking WH
No ratings yet
Ability and Sex Differences in Spatial Thinking WH
9 pages
A Study On The Comparison of Geometrical-Mechanica
No ratings yet
A Study On The Comparison of Geometrical-Mechanica
20 pages
Spatial Ability in Radiologists: A Necessary Prerequisite?: Commentary
No ratings yet
Spatial Ability in Radiologists: A Necessary Prerequisite?: Commentary
3 pages
John B. Carroll - Human Cognitive Abilities - A Survey of Factor-Analytic Studies (1993) PDF
100% (2)
John B. Carroll - Human Cognitive Abilities - A Survey of Factor-Analytic Studies (1993) PDF
824 pages
A Study On The Spatial Abilities of Prospective Social Studies Teachers - A Mixed Method Research
No ratings yet
A Study On The Spatial Abilities of Prospective Social Studies Teachers - A Mixed Method Research
22 pages
Week 4
No ratings yet
Week 4
52 pages
Lee Bednarz 2012 Components of Spatial Thinking - Evidence From A Spatial Thinking Ability Test
No ratings yet
Lee Bednarz 2012 Components of Spatial Thinking - Evidence From A Spatial Thinking Ability Test
13 pages
What Improves Spatial Thinking Evidence From The Spatial Thinking Abilities Test
0% (1)
What Improves Spatial Thinking Evidence From The Spatial Thinking Abilities Test
20 pages
DBDA Manual
No ratings yet
DBDA Manual
29 pages
PR 2 - Dbda
No ratings yet
PR 2 - Dbda
17 pages
Cognitive Functions
No ratings yet
Cognitive Functions
13 pages
Deary Penke Johnson 2010 - Neuroscience of Intelligence Review
No ratings yet
Deary Penke Johnson 2010 - Neuroscience of Intelligence Review
12 pages
Genetic Foundations of Human Intelligence
No ratings yet
Genetic Foundations of Human Intelligence
44 pages
Ni Hms 72545
No ratings yet
Ni Hms 72545
9 pages
Dbda Introduction - Group 4
No ratings yet
Dbda Introduction - Group 4
6 pages
Mohler Dissertation
No ratings yet
Mohler Dissertation
383 pages
Artigo - Variations in Cognitive Maps
No ratings yet
Artigo - Variations in Cognitive Maps
14 pages
Noble 2007
No ratings yet
Noble 2007
18 pages
Chapter 1
No ratings yet
Chapter 1
9 pages
Assignment No 1 Intro To Psychology
No ratings yet
Assignment No 1 Intro To Psychology
45 pages
Chapter 1 - Variations in Psychological Attributes-4
No ratings yet
Chapter 1 - Variations in Psychological Attributes-4
25 pages
DBDA Manual
No ratings yet
DBDA Manual
25 pages
THE DESCRIPTION OF SPATIAL-VISUALIZATION Abilities
No ratings yet
THE DESCRIPTION OF SPATIAL-VISUALIZATION Abilities
15 pages
Ujian Aptitud
No ratings yet
Ujian Aptitud
24 pages
Spatial Thinking Ability Test
No ratings yet
Spatial Thinking Ability Test
13 pages
This Is 2nd Post in Our Smart Workplace Series. in Case You Missed Our First Post About Cultural Intelligence, You Can Find It
No ratings yet
This Is 2nd Post in Our Smart Workplace Series. in Case You Missed Our First Post About Cultural Intelligence, You Can Find It
4 pages
Wai Lakin 2024 The Importance of Identifying Spatial Skills in Talented Students
No ratings yet
Wai Lakin 2024 The Importance of Identifying Spatial Skills in Talented Students
8 pages
Class 12 Psychology
No ratings yet
Class 12 Psychology
199 pages
Chapter 5 Intelligence and Achievement
No ratings yet
Chapter 5 Intelligence and Achievement
124 pages
What Determines Our Navigational Abilities?: Thomas Wolbers and Mary Hegarty
100% (1)
What Determines Our Navigational Abilities?: Thomas Wolbers and Mary Hegarty
9 pages
Creyos Health Sample Report
No ratings yet
Creyos Health Sample Report
3 pages
J Exper Analysis Behavior - 2013 - Williams - Individual Differences Intelligence and Behavior Analysis
No ratings yet
J Exper Analysis Behavior - 2013 - Williams - Individual Differences Intelligence and Behavior Analysis
13 pages
After Reading This Chapter, You Would Be Able To:: Some Misuses of Intelligence Tests
No ratings yet
After Reading This Chapter, You Would Be Able To:: Some Misuses of Intelligence Tests
22 pages
Special Article: Biology and The Future of Psychoanalysis: A New Intellectual Framework For Psychiatry Revisited
100% (1)
Special Article: Biology and The Future of Psychoanalysis: A New Intellectual Framework For Psychiatry Revisited
20 pages
Personality Insights for Managers
No ratings yet
Personality Insights for Managers
57 pages
School of Psychology....
No ratings yet
School of Psychology....
17 pages
Sensory Deprivation
100% (2)
Sensory Deprivation
24 pages
Usingslipperysnakes
No ratings yet
Usingslipperysnakes
2 pages
5 Visualization Techniques for Success
100% (3)
5 Visualization Techniques for Success
10 pages
An Investigation of English Listening Strategies: Riri Fauzana
No ratings yet
An Investigation of English Listening Strategies: Riri Fauzana
6 pages
801 English Question Paper
No ratings yet
801 English Question Paper
4 pages
Understanding Divine
No ratings yet
Understanding Divine
27 pages
Komuniti Pembelajaran Profesional: PLC: Peer Coaching
No ratings yet
Komuniti Pembelajaran Profesional: PLC: Peer Coaching
43 pages
Abacus Brochure
No ratings yet
Abacus Brochure
2 pages
A Psychology of The Film
No ratings yet
A Psychology of The Film
20 pages
ReMeth - Chap 1
No ratings yet
ReMeth - Chap 1
29 pages
Self Hypnosis Techniques Guide
90% (10)
Self Hypnosis Techniques Guide
12 pages
Well Done!: Threshold Achievement Test For Information Literacy
No ratings yet
Well Done!: Threshold Achievement Test For Information Literacy
1 page
Tumbang ICF ADHD
No ratings yet
Tumbang ICF ADHD
1 page
MBA Organizational Behavior Insights
No ratings yet
MBA Organizational Behavior Insights
19 pages
Bookhoover Ebook - Speed Reading Guide - English
No ratings yet
Bookhoover Ebook - Speed Reading Guide - English
20 pages
Child Development Table
No ratings yet
Child Development Table
3 pages
The Program For Better Vision
100% (2)
The Program For Better Vision
223 pages
Ambient Noise and Virtual Learning
No ratings yet
Ambient Noise and Virtual Learning
3 pages
Ideas Pertaining To A Pure Phenomenology and To A Phenomenological Philosophy - Second Book Studies in The Phenomenology of Constitution (PDFDrive)
No ratings yet
Ideas Pertaining To A Pure Phenomenology and To A Phenomenological Philosophy - Second Book Studies in The Phenomenology of Constitution (PDFDrive)
454 pages
Understanding the Psychological Self
No ratings yet
Understanding the Psychological Self
24 pages
Sparse Distributed Memory: The Science of Computing
No ratings yet
Sparse Distributed Memory: The Science of Computing
5 pages
Arts Integration and 21st Century Skills
No ratings yet
Arts Integration and 21st Century Skills
3 pages
Binocular Vision Lab Guide
No ratings yet
Binocular Vision Lab Guide
11 pages
The Brain's Default Mode Network: Further
No ratings yet
The Brain's Default Mode Network: Further
17 pages
PSYC 1001 Seminar Questions 2020revised
100% (1)
PSYC 1001 Seminar Questions 2020revised
100 pages
Two-Dimensional Blooms Cognitive Domain
No ratings yet
Two-Dimensional Blooms Cognitive Domain
4 pages
Chapter 3 - BPTNC
No ratings yet
Chapter 3 - BPTNC
24 pages

Individual Differences in Spatial Abilities

Uploaded by

Individual Differences in Spatial Abilities

Uploaded by

4

Individual Differences in Spatial Abilities

Mary Hegarty and David A. Waller

differential methods in the same study, in order to examine aptitude-

the factor-analytic approach to visuospatial abilities

to quantitatively scored multiple-choice questions that could be answered

Tests cited as typical

3. Spatial Orientation - Guilford–Zimmerman

(c) Kinesthetic imagery, which was believed to be associated with left-right

Cambridge Books Online © Cambridge University Press, 2010

simple (two-dimensional) mental rotation items (such as the Card Rota-

Flexibility of closure is thus more closely associated with the ability to

1948; McGee, 1979b; Thurstone, 1950) and it continues to appear promi-

Evaluation of the Factor-Analytic Literature

neglected assessing spatial abilities in at least two other relevant domains:

the paper-and-pencil tests than with the computer-administered dynamic

cognitive analyses of performance on

methods with examination of individual differences to study aptitude-

The Contribution of Speed of Processing

Although speed of processing can account for individual differences in

The Contribution of Strategies

mental-rotation strategy, a perspective-taking strategy in which the solver

Spatial Abilities and Working Memory

accounted for in terms of differences in a spatial working memory sys-

Miyake et al. (2001) measured performance on tests of the three spatial

Spatial Abilities and Mental Imagery

Another conception of individual differences in imagery is based on

transform and find specific patterns in images. A study by Poltrock and

Evaluation of Cognitive Analyses of Spatial Abilities

environmental spatial abilities

mental representations of large-scale environments such as buildings, cam-

judgments such as proximity and connectedness. Measures loading on

one might expect psychometric measures of spatial ability to be only mildly

table 4.2. Summary of correlational studies of the relationship between spatial

Number of Correlations Range of Median

Recently, however, Hegarty et al. (2004) have shown a closer relation-

ability to acquire knowledge of previously unfamiliar environments; (2)

table 4.3. Summary of correlational studies of the relationship between spatial

Number of Correlations Range of Median abs.

Waller (2000) gave people several tests of psychometrically assessed spatial

functions of spatial abilities

these associations? Do these occupations and academic disciplines require

Spatial Abilities and Occupational Success

regarded as valuable for selection and classification only for occupations

Spatial Abilities and Mechanical Reasoning

components had to be mentally animated. Low-spatial-individuals made

related to spatial ability, presumably because these problems are more

Spatial Abilities and Mathematics

described in a problem, whereas low-spatial visualizers tended to con-

Research on the cognitive analysis of existing spatial tests can provide

and 9980122 from the National Science Foundation and N00014-96-10525

Suggestions for Further Readings

Cronbach, L. J. (1957). The two disciplines of scientific psychology. American Psy-

Just, M. A., & Carpenter, P. A. (1985). Cognitive coordinate systems: Accounts of

Waller, D. (1999). An assessment of individual differences in spatial knowledge of

You might also like