A crowded cupboard presents a small spatial puzzle. A suitcase presents another. We look at several objects, estimate their proportions, imagine possible arrangements, and anticipate what will happen if one item turns, slides, or blocks another. The same kind of thinking helps us follow a map, understand a diagram, assemble furniture, read an architectural plan, or judge whether a car can fit through a narrow opening. These are ordinary acts of spatial reasoning.

The experience can feel visual, but spatial reasoning is more than seeing a picture in the mind. It is the capacity to represent locations, shapes, distances, directions, and relationships—and to reason about how those relationships remain stable or change. Sometimes the work happens internally. At other times, the mind leans on gestures, sketches, models, maps, or the physical environment. Spatial thought is therefore both a mental capacity and a way of organizing information for action (National Research Council, 2006).

Spatial reasoning is often treated as a specialized gift: something an architect, engineer, or unusually good navigator possesses. Research describes a more differentiated pattern. People differ in spatial abilities, but the domain contains several partly distinct skills, and many respond to practice. The central question is not simply whether a person is “good at space.” It is how that person represents a spatial problem, which transformations the task requires, and which supports make the structure easier to grasp (Carroll, 1993; McGrew, 2009; Uttal et al., 2013). One way to hold these influences together is to consider the person, task, and representation. This is an editorial organizing lens, not a formal psychological theory. Performance reflects the person’s experience and strategies, the relationships required by the task, and the mental or external representations available for working through them.

Defining Spatial Reasoning

The National Research Council defined spatial thinking as an integration of three elements: concepts of space, tools of representation, and processes of reasoning. Concepts such as distance, direction, scale, containment, and continuity give a problem its structure. Representations—including mental images, diagrams, models, maps, and coordinate systems—make that structure available. Reasoning allows a person to compare, transform, infer, explain, and predict within it (National Research Council, 2006).

This definition is deliberately broader than spatial ability as measured by a single test. A timed mental-rotation task may reveal how efficiently someone transforms an object in imagination. A navigation task may require orientation, route knowledge, landmarks, and decisions made while moving. A graph can spatialize a relationship that is not literally geometric at all. The general domain is connected, but performance in one task should not be assumed to describe every form of spatial thought.

Psychometric theories make a similar distinction. In the Cattell–Horn–Carroll framework, visual processing includes abilities involved in generating, storing, retrieving, and transforming visual patterns. Narrower abilities include visualization, spatial relations, visual memory, closure speed, and perceptual scanning. This classification is useful because it keeps spatial reasoning inside a broader architecture of intelligence without reducing the whole domain to a single score (Carroll, 1993; McGrew, 2009).

Spatial reasoning can therefore be understood as the purposeful use of spatial representations and relationships to solve a problem. It may involve an internal image, but it need not. A person can reason spatially by moving pieces, tracing a route with a finger, drawing a diagram, or comparing landmarks. What matters is the structure of the reasoning, not whether the experience resembles a vivid private picture.

Forms of Spatial Thought

Researchers distinguish types of spatial reasoning by asking what must be represented and what must change. One useful typology poses two questions: Is the task primarily about the properties of a single object or about relations among several objects? And must the relevant spatial information remain stable or be transformed? Crossing these distinctions produces four broad families—intrinsic-static, intrinsic-dynamic, extrinsic-static, and extrinsic-dynamic spatial skills (Newcombe & Shipley, 2015; Uttal et al., 2013).

The categories are not sealed compartments. Reading a mechanical diagram may require recognizing the parts, tracking their arrangement, and anticipating their movement. The typology is best used as a map of task demands, not as a set of labels for people.

Mental Rotation and Object Transformation

Intrinsic spatial tasks concern the shape or configuration of an object itself. Recognizing a familiar object from an unusual angle is comparatively static. Imagining that object rotating, folding, bending, or being cut requires a dynamic transformation. Mental rotation is the classic example, but paper folding, cross-section judgments, and anticipating how components will fit also belong to this family (Newcombe & Shipley, 2015; Uttal et al., 2013).

Roger Shepard and Jacqueline Metzler gave mental rotation one of its most recognizable experimental forms. Adults judged whether pairs of three-dimensional block figures showed the same object at different orientations or a reflected object. Response time increased approximately linearly as the angular difference grew. The pattern suggested that many participants were carrying out an analog-like transformation whose duration reflected the distance rotated (Shepard & Metzler, 1971).

The study did not prove that every person always rotates a picture-like object through an internal space. It used eight adults and a restricted task. Even so, its orderly relation between angle and decision time became important evidence that mental transformations can preserve aspects of physical transformation. Later research has treated the result as a starting point for examining strategy, representation, expertise, and individual difference rather than as a complete theory of imagery (Shepard & Metzler, 1971).

Understanding Spatial Relationships

Extrinsic spatial thought concerns relations among objects or locations: which item is above another, how far two points are separated, whether a path crosses a boundary, or how several parts form a system. A seating plan, subway map, circuit diagram, and molecular model all preserve relations while omitting many surface details.

Some of these relations are static. Others are dynamic: traffic converges, gears turn together, bodies move through a field, and a changing graph displays a system over time. Reasoning well requires selecting the relations that matter and ignoring details that do not. A useful representation is therefore not always the most realistic one. It is the one that makes the relevant structure available for comparison and inference (National Research Council, 2006).

This helps explain why spatial thinking can organize abstract information. A number line gives numerical magnitude a location and direction. A family tree turns kinship into branching position. A scientific graph places change into axes. These tools do more than decorate a verbal explanation; they make relationships visible enough to inspect.

Perspective, Orientation, and Navigation

Navigation adds the problem of a moving observer. A route learned in one direction can feel unfamiliar on the return trip because the visible sequence and left–right relations reverse. Finding a new route may require coordinating an egocentric perspective—where places are relative to the body—with a more survey-like understanding of how locations relate across the larger environment (National Research Council, 2006).

Perspective taking asks a related question: what would the scene look like from another position? The change may be imagined, enacted through movement, or supported by turning a map. These tasks can recruit skills that overlap with object transformation, yet they are not interchangeable with mental rotation. Rotating an object while the observer remains conceptually fixed is different from imagining the observer moving around a stable array (Uttal et al., 2013).

The distinction matters in daily life. Someone may be adept at fitting objects together but struggle to translate a north-oriented map into the turn that must be made from the current position. That uneven profile is not contradictory. It reflects the fact that spatial reasoning is a coordinated family of abilities rather than one all-purpose capacity.

How Spatial Representations Work

Transforming a Spatial Representation

Spatial reasoning begins with a representation, but representations are selective. The mind does not preserve every visible property of an object or environment. It encodes the features that seem relevant to the current goal—perhaps orientation, boundary, distance, or connectivity—and may simplify or distort other features to reduce the burden of the problem (National Research Council, 2006).

A transformation then changes the representation while preserving some relations. When imagining a chair turned toward a window, the imagined view changes, but the chair’s parts must remain connected. When folding a paper shape, edges that begin far apart may become adjacent. Good spatial reasoning depends on tracking which relationships should change and which should remain invariant.

Thinking With Gestures and External Tools

Grounded-cognition theories provide one account of how such thought may be connected to perception and action. Lawrence Barsalou argued that cognition often draws on partial reenactments of perceptual, motor, and introspective states acquired through experience. Deliberate imagery is the familiar conscious case, but simulation can also be partial, automatic, and situated in interaction with the environment (Barsalou, 2008).

This view fits the ordinary usefulness of gesture and external representation. Turning a hand, sketching a line, or manipulating a model can make a transformation easier to track. The external act is not necessarily a crutch replacing “real” thought. It can become part of the reasoning system by stabilizing information that would otherwise have to be maintained and updated internally (Barsalou, 2008; National Research Council, 2006).

Evidence from learning studies supports this practical role. In one experiment, 158 six-year-olds received instruction on a mental-transformation task. Children who produced a movement gesture relevant to the transformation improved more than children who only observed the gesture; pointing movements that did not represent the transformation did not provide the same benefit. The result is task-specific, but it shows that movement can help construct a representation when the gesture carries information needed for the problem (Goldin-Meadow et al., 2012).

How Spatial Reasoning Develops

Spatial understanding develops through repeated coordination of perception and action. Infants and children learn what remains stable when they move, how objects continue behind occlusion, how routes connect, and how a model can stand for a larger place. Historical developmental theories, especially Jean Piaget’s, helped make the construction of object and spatial relations a central question. Modern evidence is less committed to a single stage sequence, but it retains the insight that spatial knowledge is actively built through engagement with the world (Piaget, 1950; National Research Council, 2006).

Language and cultural tools expand that construction. Words such as above, between, behind, near, and parallel direct attention to relationships that might otherwise remain implicit. Blocks, diagrams, maps, coordinate systems, and digital models make structures available for shared inspection. Instruction can therefore change not only what a learner knows, but also what the learner notices as spatially relevant (National Research Council, 2006; Pruden et al., 2011).

Development is neither a smooth ascent nor the unfolding of a single inborn capacity. Opportunities differ. Children encounter different kinds of play, explanation, navigation, drawing, building, and technical instruction. They also bring different patterns of attention, confidence, prior knowledge, visual imagery, working memory, and motor experience to the same task (National Research Council, 2006; Uttal et al., 2013).

For this reason, an observed difference in performance is not self-explanatory. It may reflect experience, strategy, representation, time pressure, familiarity with the task format, or one narrower spatial skill. Assessment can identify a pattern worth understanding; it should not be used to turn a developing capacity into a fixed identity.

Longitudinal studies offer a more concrete view of these everyday influences. Pruden and colleagues followed 52 parent-child pairs from 14 to 46 months and assessed the children’s spatial performance at 54 months. Parents’ spatial-language use predicted children’s own spatial-language production after overall parent language input was controlled, and children who produced more spatial language performed better on later spatial tasks (Pruden et al., 2011). In another study, 53 children and their parents were observed six times between ages two and four. Children observed playing with puzzles later performed better on a two-dimensional spatial-transformation task, controlling for parent education, income, and overall parent word types (Levine et al., 2012).

Neither study establishes a simple causal path. Families were not randomly assigned to use spatial language or play with puzzles, and unmeasured differences could contribute to the associations. Together, however, the findings show why ordinary conversation and play deserve attention as contexts in which children learn to notice, describe, and transform spatial relationships.

Why Spatial Reasoning Matters

Many problems become easier when their structure can be represented in space. This advantage is not confined to occupations with “spatial” in the job description. It appears whenever a person must coordinate parts and wholes, understand a changing configuration, or move between an object and a representation of it (National Research Council, 2006).

Everyday Problem-Solving

Packing, repairing, rearranging, cooking, sewing, driving, and finding one’s way through an unfamiliar building all place demands on spatial thought. These tasks require different combinations of scale, sequence, orientation, and prediction. A recipe may involve imagining volume and arrangement; a repair may require mapping a two-dimensional instruction onto a three-dimensional object.

External representations often make the decisive difference. A rough sketch can preserve a room’s proportions. Turning a map can align it with the viewer’s direction. Laying out parts before assembly can reveal an impossible sequence. These actions reduce the amount of information that must remain active at once and allow the environment to carry part of the cognitive work (National Research Council, 2006).

Everyday competence is also shaped by familiarity. An experienced mover may see packing options that a novice misses; a local resident may navigate with a few landmarks while a visitor needs step-by-step directions. Expertise changes what is noticed and how a scene is represented, so fluent performance should not be mistaken for effortless inborn intuition (Barsalou, 2008; National Research Council, 2006).

Mathematics and Scientific Learning

Mathematics and science repeatedly translate relationships into spatial form. Number lines organize magnitude. Geometry makes transformation explicit. Graphs reveal rate and pattern. Chemistry, anatomy, geology, astronomy, and engineering require learners to coordinate views, scales, models, and processes that cannot always be observed directly (National Research Council, 2006).

Longitudinal evidence shows that spatial ability measured in adolescence is associated with later education and occupations in science, technology, engineering, and mathematics. Project TALENT drew roughly 400,000 students from a stratified random sample of U.S. high schools and followed them for 11 years after graduation. Spatial ability added information beyond mathematical and verbal measures, and many adolescents with especially high spatial scores would have been missed by selection systems focused only on math and language (Wai et al., 2009).

The result is important, but it is not destiny. The study was observational and predictive. It does not show that a spatial score causes a career, nor that other abilities, interests, opportunities, persistence, and social conditions are secondary. Its strongest implication is that systems may overlook a meaningful form of ability when they define talent too narrowly.

Training research offers more qualified evidence. A meta-analysis of 29 controlled pre-post studies found a small positive average effect of spatial training on mathematics performance (Hedges’ g = .28) and a larger average effect on spatial outcomes (g = .49). Mathematics transfer was stronger when the outcome measure was closely aligned with the spatial training, and the mechanisms remained uncertain. Spatializing mathematical ideas may help, but the findings do not support treating generic spatial practice as a guaranteed route to broad mathematical improvement (Hawes et al., 2022).

Design, Skilled Work, and Creative Construction

Design and skilled work make spatial reasoning visible because ideas must survive contact with materials. A carpenter anticipates how cuts alter a piece. A surgeon coordinates anatomy with a changing viewpoint. A technician follows relations in a schematic. An artist manipulates proportion, depth, and composition. An architect moves between plan, elevation, model, and lived environment.

These practices depend on learned representational systems as much as on internal visualization. Professional diagrams omit, exaggerate, or standardize features so that important relations can be seen quickly. Expertise includes knowing what a representation means, when it is misleading, and how to translate it back into action (National Research Council, 2006).

Wai and colleagues found that adolescents with high spatial ability but less exceptional math or verbal scores later entered visual arts as well as STEM fields at rates above the cohort’s base rates. The finding cautions against treating spatial ability as relevant to only one occupational path. It also does not imply that spatial ability alone explains professional skill (Wai et al., 2009).

The Reach and Limits of Spatial Training

The evidence indicates that training can improve spatial performance, although the size and reach of those gains vary. David Uttal and colleagues synthesized 217 training studies and found that spatial performance improved after training. Gains appeared across ages and sexes, remained detectable after delays in the studies that measured them, and transferred beyond the practiced task on average. The review included direct practice, coursework, video games, and other spatially demanding activities; no single training method accounted for the overall pattern (Uttal et al., 2013).

Malleability does not mean unlimited or uniform change. Meta-analytic averages combine people, tasks, durations, and comparison conditions that differ substantially. Improvement on a mental-rotation measure may not produce better navigation. Learning a software model may not automatically improve geometric proof. The National Research Council similarly warned that spatial expertise can remain domain-specific and that transfer often needs guidance and a meaningful context (National Research Council, 2006).

The most defensible training begins with the target problem. If a learner must understand molecular structure, practice should connect rotation and multiple views to molecules. If a child is learning fractions, diagrams and manipulatives should reveal fraction relations rather than merely add visually interesting activity. If an adult struggles with a route, aligned maps, landmarks, and repeated travel may be more relevant than a generic rotation game (Hawes et al., 2022; National Research Council, 2006).

External supports are part of good instruction. Sketches, manipulatives, animations, physical movement, and well-designed diagrams can reduce unnecessary load while keeping the important transformation visible. As fluency grows, support can be varied or gradually removed. The aim is not to force every person to solve the problem in an unaided mental workspace; it is to help the learner build a representation that can be used flexibly (Goldin-Meadow et al., 2012; National Research Council, 2006).

Differences in Spatial Performance

Interpreting Spatial Test Scores

A person can struggle with one spatial task and reason effectively in many other ways. The CHC framework places visual processing among several broad cognitive abilities rather than treating it as a synonym for intelligence. Within visual processing, narrower skills can also differ. A lower score in speeded rotation may coexist with strong visual memory, careful diagram interpretation, verbal reasoning, quantitative knowledge, or creative problem-solving (McGrew, 2009).

Task format matters. Timed tests reward speed as well as accuracy. Unfamiliar block figures may disadvantage someone who performs well with meaningful materials. Anxiety can consume attention. Poorly designed graphics can create difficulty that belongs to the representation rather than the reader (Arrighi & Hausmann, 2022; National Research Council, 2006).

Effort also shapes the experience of a spatial task. In one experiment, participants liked unfamiliar nonsense shapes less when the judgment required 135 degrees of mental rotation than when it required 45 degrees. The result does not show that spatial thinking is inherently unpleasant. It suggests that processing effort can make unfamiliar material slightly less appealing, at least under the study’s conditions (Morsella et al., 2011).

Anxiety and confidence can also shape what a timed test records. Arrighi and Hausmann studied 269 adults completing two mental-rotation tasks of different difficulty. Spatial anxiety and lower task-specific self-confidence statistically mediated sex/gender differences in performance, especially on the more demanding task. Because these variables were measured rather than experimentally manipulated, the analysis cannot establish that anxiety or confidence caused the group differences. It does show that a score may reflect psychological burden and task conditions as well as spatial processing (Arrighi & Hausmann, 2022).

Confidence matters because effort can be misread as evidence of inability. A difficult rotation or perspective shift may feel slow before it becomes organized. Practice is most useful when it provides informative feedback, several ways to represent the problem, and a clear connection between the activity and the skill being learned. Calling a skill trainable should lead to better opportunities for supported practice (Arrighi & Hausmann, 2022; Uttal et al., 2013).

Spatial Reasoning Without Vivid Imagery

Aphantasia offers a direct reason to separate vivid imagery from spatial reasoning. People who report little or no voluntary visual imagery may still solve spatial tasks through analytic or other alternative strategies. The absence of a picture-like experience is therefore not, by itself, evidence that spatial thought is absent (Kay et al., 2024).

Kay and colleagues compared 229 adults reporting aphantasia with 262 controls on two mental-rotation tasks. The aphantasia group was slower but more accurate on average, and both groups showed the familiar increase in response time as angular difference increased. Participants with aphantasia also reported greater use of analytic strategies. Classification relied on self-report in an online study, and group averages do not describe every individual. Even with those limits, the findings challenge the assumption that successful mental rotation requires vivid conscious imagery (Kay et al., 2024).

This distinction has practical consequences. Learners should be offered more than one route into a problem: words paired with diagrams, objects that can be handled, views that can be rotated, and time to explain a strategy. In everyday and professional life, competent spatial thinking routinely includes such tools (National Research Council, 2006).

Differences still matter. They can help educators notice an overlooked strength or identify a task that needs explicit instruction. An assessment becomes harmful when it hardens a partial observation into a global identity. Spatial performance describes how a person responded to a particular task under particular conditions; it is not a final statement about the person.

A Few Words from Psychology Fanatic

Return to the crowded cupboard. What first looked like a private flash of insight is actually a coordinated act: noticing proportions, representing constraints, imagining transformations, and testing a possible arrangement against the physical world. The mind may carry some of that work internally, while the hands and environment carry the rest.

Spatial reasoning gives structure to movement, location, shape, and relation. It also gives visible form to ideas that are not literally spatial. Its varieties explain why one person can navigate easily yet dislike paper-folding problems, while another can design an object but become disoriented in a new neighborhood. Such uneven profiles are expected in a domain composed of several partly distinct abilities.

Taken together, the evidence supports the importance and malleability of spatial skills. Their educational and occupational associations deserve attention, especially when verbal and mathematical measures dominate how ability is recognized. Transfer remains bounded, external tools contribute to performance, and no single score captures the ways a person can learn to think with space. The person–task–representation lens keeps that complexity visible by directing attention to the thinker’s experience and strategy, the demands of the problem, and the representations available for use. A constructive question is which representation allows the relationship to become clear.

Associated Concepts

• Cattell–Horn–Carroll Theory of Intelligence: A hierarchical model that places visual processing alongside other broad and narrow cognitive abilities.
• Working Memory: A limited-capacity system involved in maintaining and updating information during complex reasoning.
• Cognitive Load Theory: A framework for understanding how task design and instructional supports affect limited cognitive resources.
• Barsalou’s Perceptual Symbol Theory: A grounded account in which conceptual processing draws on partial simulations of perception, action, and experience.
• Aphantasia: The experience of little or no voluntary visual imagery, reminding us not to equate imagery vividness with every form of spatial reasoning.
• Law of Least Effort: The tendency to prefer less effortful means when several routes lead toward the same goal.

References

Arrighi, Linda; Hausmann, Markus (2022). Spatial Anxiety and Self-Confidence Mediate Sex/Gender Differences in Mental Rotation. Learning & Memory, 29(9), 312–320. DOI: 10.1101/lm.053596.122.
(Return to Main Text)

Barsalou, Lawrence W. (2008). Grounded Cognition. Annual Review of Psychology, 59, 617–645. DOI: 10.1146/annurev.psych.59.103006.093639.
(Return to Main Text)

Carroll, John B. (1993). Human Cognitive Abilities: A Survey of Factor-Analytic Studies. Cambridge University Press. DOI: 10.1017/CBO9780511571312 ISBN: 978-0-521-38712-5.
(Return to Main Text)

Goldin-Meadow, Susan; Levine, Susan C.; Zinchenko, Elena; Yip, Terina KuangYi; Hemani, Naureen; Factor, Laiah (2012). Doing Gesture Promotes Learning a Mental Transformation Task Better Than Seeing Gesture. Developmental Science, 15(6), 876–884. DOI: 10.1111/j.1467-7687.2012.01185.x.
(Return to Main Text)

Hawes, Zachary C. K.; Gilligan-Lee, Katie A.; Mix, Kelly S. (2022). Effects of Spatial Training on Mathematics Performance: A Meta-Analysis. Developmental Psychology, 58(1), 112–137. DOI: 10.1037/dev0001281.
(Return to Main Text)

Kay, Lachlan; Keogh, Rebecca; Pearson, Joel (2024). Slower but More Accurate Mental Rotation Performance in Aphantasia Linked to Differences in Cognitive Strategies. Consciousness and Cognition, 121, 103694. DOI: 10.1016/j.concog.2024.103694.
(Return to Main Text)

Levine, Susan C.; Ratliff, Kristin R.; Huttenlocher, Janellen; Cannon, Joanna (2012). Early Puzzle Play: A Predictor of Preschoolers’ Spatial Transformation Skill. Developmental Psychology, 48(2), 530–542. DOI: 10.1037/a0025913.
(Return to Main Text)

McGrew, Kevin S. (2009). CHC Theory and the Human Cognitive Abilities Project: Standing on the Shoulders of the Giants of Psychometric Intelligence Research. Intelligence, 37(1), 1–10. DOI: 10.1016/j.intell.2008.08.004.
(Return to Main Text)

Morsella, Ezequiel; Feinberg, Giles H.; Cigarchi, Sepeedeh; Newton, James W.; Williams, Lawrence E. (2011). Sources of Avoidance Motivation: Valence Effects from Physical Effort and Mental Rotation. Motivation and Emotion, 35(3), 296–305. DOI: 10.1007/s11031-010-9172-y.
(Return to Main Text)

National Research Council (2006). Learning to Think Spatially: GIS as a Support System in the K–12 Curriculum. The National Academies Press. DOI: 10.17226/11019 ISBN: 978-0-309-09208-1.
(Return to Main Text)

Newcombe, Nora S.; Shipley, Thomas F. (2015). Thinking About Spatial Thinking: New Typology, New Assessments. In John S. Gero (Ed.), Studying Visual and Spatial Reasoning for Design Creativity (pp. 179–192). Springer Dordrecht. DOI: 10.1007/978-94-017-9297-4_10.
(Return to Main Text)

Piaget, Jean (1950). The Psychology of Intelligence. Routledge & Kegan Paul. Translated by Malcolm Piercy and David E. Berlyne.
(Return to Main Text)

Pruden, Shannon M.; Levine, Susan C.; Huttenlocher, Janellen (2011). Children’s Spatial Thinking: Does Talk About the Spatial World Matter? Developmental Science, 14(6), 1417–1430. DOI: 10.1111/j.1467-7687.2011.01088.x.
(Return to Main Text)

Shepard, Roger N.; Metzler, Jacqueline (1971). Mental Rotation of Three-Dimensional Objects. Science, 171(3972), 701–703. DOI: 10.1126/science.171.3972.701.
(Return to Main Text)

Uttal, David H.; Meadow, Nathaniel G.; Tipton, Elizabeth; Hand, Linda L.; Alden, Alison R.; Warren, Christopher; Newcombe, Nora S. (2013). The Malleability of Spatial Skills: A Meta-Analysis of Training Studies. Psychological Bulletin, 139(2), 352–402. DOI: 10.1037/a0028446.
(Return to Main Text)

Wai, Jonathan; Lubinski, David; Benbow, Camilla P. (2009). Spatial Ability for STEM Domains: Aligning Over 50 Years of Cumulative Psychological Knowledge Solidifies Its Importance. Journal of Educational Psychology, 101(4), 817–835. DOI: 10.1037/a0016127.
(Return to Main Text)