Trends in Cognitive Sciences
ReviewDevelopment of mental transformation abilities
Introduction
One of the most impressive skills of the human species is the ability to represent and mentally transform the shapes of objects. People can generate mental images of 2D shapes or 3D objects and can transform them in various ways (e.g., rotating, bending, or folding them) [1]. Such flexible representations are vital for making predictions regarding the positions of moving objects; for example, to avoid collisions when crossing a street. They also allow anticipation of the effects of actions when manipulating objects or using tools. Furthermore, the ability to perform mental spatial transformations predicts number and mathematics skills 2, 3. Thus, determining the origins and development of mental transformations is a central and topical problem in cognitive science, with translational implications for intervention. However, research on this issue has led to paradoxical findings, with infants showing remarkable abilities but young children failing on seemingly similar tasks. In this review we put these contrasting results in context with similar findings in other domains and suggest an underlying mechanism.
Section snippets
Age discrepancies in mental transformation
Much of the previous research on mental transformation has focused on a specific kind of spatial transformation termed mental rotation (MR), which refers to imagining a rotational movement of an object (or array of objects) in 2D or 3D space. In a classic MR task [4], participants must decide whether a rotated image is the same as a comparison image or its mirror image. In developmental research, this paradigm has been adapted for use with children and even infants, with oddly contrasting
Studies with preschoolers and young children
Although Marmor 5, 6 found that 4–5-year-olds were able to perform MR, there has been controversy about her conclusions [7]. A follow-up study employing the same procedure with different stimuli failed to replicate Marmor's results [8]. Other studies showed that, at 4–5 years, many children performed at chance level and only few showed signs of MR 9, 10, 11. Even efforts to simplify the tasks by using a touch screen or presenting simple and engaging stimuli have failed to demonstrate MR in
Infant studies
In sharp contrast to research with preschoolers, recent research has shown that infants can discriminate mirror images despite differences in orientation 14, 15, 16, 17, 18, 19, 20, 21. For example, in two such studies 14, 15, infants saw an asymmetrical object being moved straight down behind an occluder. When the occluder was lowered, it revealed either the same object (possible event) or its mirror image (impossible event) in one of five different orientations (Figure 1). Infants had been
Presentation of motion or multiple views
In MR paradigms used with adults [4] and children 5, 6, 9, 25, 26, 27, stimuli are typically static images in single orientations. By contrast, infant often see objects in actual physical rotation before the test 16, 17, 20, 21, 23, 24 or in multiple static orientations 18, 19. This may allow infants to extrapolate motion or to interpolate between familiar views [28]. Indeed, research has demonstrated that infants are better at recognizing objects that are presented in motion compared with
Perspective taking
Perspective taking refers to the ability to adopt someone else's spatial perspective. It is logically akin to MR, as it involves mentally rotating oneself into another vantage point. However, despite this similarity, perspective taking and MR are dissociated in various behavioral and neural ways 33, 34, 35. Developmental research has shown that perspective taking emerges at around 4 or 5 years but improves considerably through the age of 8 years [36]. Although preschoolers perform better on
Conceptualizing development
How can these paradoxical age discrepancies be explained? One possible interpretation is to assume a U-shaped developmental trajectory (Figure 2A), in which infants possess an early ability that is temporarily lost and reacquired later. However, as discussed above regarding MR, infant paradigms (Paradigm1) and paradigms used with older children (Paradigm2) differ in many ways. Hence, we should also consider an alternative trajectory (Figure 2B) with two possible versions. In one scenario,
Common mechanism?
Taken together, contrasting results of seemingly sophisticated competencies in infancy and surprisingly low performance in preschoolers can be found in several cognitive domains. Some of these domains appear closely related, such as perspective taking and theory of mind; others do not seem to have much in common, such as knowing how a ball rolls and reasoning about another person's beliefs. Furthermore, it is remarkable that similar age dissociations are found not only in areas such as
Concluding remarks
Developmental research has shown apparently sophisticated abilities in infants across multiple domains, whereas older children struggle with seemingly similar tasks. Recent infant studies in the domain of MR have demonstrated similar patterns of paradoxical age discrepancies, providing new support for the claim that mental simulation may be key to successful performance in many cognitive domains. Whereas content, modality, and simulated perceptual inputs may vary across domains, the ability to
Acknowledgments
This research was supported by research grants from the Swiss National Science Foundation # PZ00P1_131866 and from the US National Science Foundation #SBE-0541957 and SBE-1041707. The authors thank two anonymous reviewers for their helpful comments.
References (98)
Development of kinetic images: when does the child first represent movement in mental images?
Cogn. Psychol.
(1975)Development of mental rotation in 3- to 5-year-old children
Cogn. Dev.
(2013)- et al.
Mental object rotation and motor development in 8- and 10-month-old infants
J. Exp. Child Psychol.
(2013) - et al.
Dynamic mental representation in infancy
Cognition
(1997) - et al.
Tracking and anticipation of invisible spatial transformation by 4- to 8-month-old infants
Cogn. Dev.
(1996) - et al.
Spilling thoughts: configuring attentional resources in infants’ goal-directed actions
Infant Behav. Dev.
(2000) - et al.
A dissociation between mental rotation and perspective-taking spatial abilities
Intelligence
(2004) The development of spatial perspective taking
- et al.
From infants’ to children's appreciation of belief
Trends Cogn. Sci.
(2012) - et al.
Beliefs about beliefs: representation and constraining function of wrong beliefs in young children's understanding of deception
Cognition
(1983)