ArticlePDF Available

Abstract

Here we show that if an adult demonstrates a new way to execute a task to a group of infants aged 14 months, the children will use this action to achieve the same goal only if they consider it to be the most rational alternative. Our results indicate that imitation of goal-directed action by preverbal infants is a selective, interpretative process, rather than a simple re-enactment of the means used by a demonstrator, as was previously thought.
H
ere we show that if an adult demon-
strates a new way to execute a task to a
group of infants aged 14 months, the
children will use this action to achieve the
same goal only if they consider it to be
the most rational alternative. Our results
indicate that imitation of goal-directed
action by preverbal infants is a selective,
interpretative process, rather than a simple
re-enactment of the means used by a
demonstrator, as was previously thought
1–3
.
In Meltzoffs seminal study
1
, a group of
14-month-old subjects watched a demon-
strator illuminate a light-box by leaning
forwards and touching its top with her fore-
head
1,2
. One week later, two-thirds of them
re-enacted this head action to achieve the
same outcome, although none of the control
group used it spontaneously. This was taken
as evidence that infants separate the goal
from the means, automatically imitating the
means as demonstrated
2
. Such imitative
learning is thought to be specific to humans,
as primates do not imitate new strategies to
achieve goals, relying instead on motor
actions already in their repertoire (emula-
tion)
3
. If this were also the case in infants,
they would be expected to touch the box
with their hands, rather than imitating the
unfamiliar head action. (Meltzoff, however,
did not report such hand actions
1,2
.)
The readiness of infants to re-enact the
head action is surprising, given that 1-year-
old babies can evaluate the rationality of the
means in relation to the goal and the con-
straints of the situation
4,5
. When constraints
change, these infants are able to work out the
most effective action that the demonstrator
should use to achieve the goal (the principle
of rational action
6,7
). Infants would therefore
be expected to re-enact an action only if it
seemed to them to be the most effective
means to achieve the goal (see also ref. 8).
So why did Meltzoffs subjects re-enact
the head action, when they could just have
touched the box with their hands? If infants
noticed that the demonstrator declined to
use her hands despite the fact that they were
free, they may have inferred that the head
action must offer some advantage in turning
on the light. They therefore used the same
action themselves in the same situation.
To test this idea, we replicated Meltzoffs
study
1
with one modification: in one condi-
tion, the subjects could see that the demon-
strator’s hands were occupied while she
executed the head action (pretending to be
cold, she had wrapped a blanket around
herself which she held onto with both hands;
‘hands occupied’, Fig. 1a). After witnessing
the same head action when the adult’s hands
were free (Fig. 1b), 69% of infants re-enacted
the head action, replicating Meltzoffs
results
1
. However, after watching the adult
turn on the light with her head when her
hands were occupied, the number of chil-
dren who imitated the head action dropped
significantly to only 21% (P*0.02; Fig. 1c).
It must therefore have seemed sensible to the
infants that the demonstrator should use the
head action when her hands were occupied
— nevertheless, 79% of them chose not to
imitate her because their own hands were
free, presumably concluding that the head
action was not the most rational.
Whether they re-enacted the head action
or not, all infants who watched the adult
perform under both conditions still used the
hand action. This suggests that 14-month-
old infants are still subject to an automatic,
emulation-like process whereby the memory
of the effect (illumination by touch) acti-
vates the response that is most strongly
associated with establishing contact (hand
action). But the re-enactment of the head
action, when inferred to be rational by the
infant, indicates that imitation by 14-
month-olds goes beyond emulation. We
conclude that the early imitation of goal-
directed actions is a selective, inferential
process that involves evaluation of the
rationality of the means in relation to the
constraints of the situation.
György Gergely*, Harold Bekkering†‡,
Ildikó Király*
*Institute for Psychology, Hungarian Academy of
Sciences, 1132 Budapest, Hungary
e-mail: gergelyg@mtapi.hu
Max Planck Institute for Psychological Research,
Amalienstrasse 33, 80799 Munich, Germany
Present address: Department of Experimental and
Work Psychology, University of Groningen,
9712 TS Groningen, The Netherlands
1. Meltzoff, A. N. Dev. Psychol. 24, 470–476 (1988).
2. Meltzoff, A. N. J. Exp. Child Psychol. 59, 497–515 (1995).
3. Tomasello, M. The Cultural Origins of Human Cognition
(Harvard Univ. Press, Cambridge, Massachusetts, 1999).
4. Gergely, G. et al. Cognition 56, 165–193 (1995).
5. Csibra, G. et al. Cognition 72, 237–267 (1999).
6. Gergely, G. & Csibra, G. Cognition 63, 227–233 (1997).
7. Csibra, G. & Gergely, G. Dev. Sci. 1, 255–259 (1998).
8. Bekkering, H., Wohlschlager, A. & Gattis, M.
Q. J. Exp. Psychol. Human Exp. Psych. 53A, 153–164 (2000).
brief communications
NATURE
|
VOL 415
|
14 FEBRUARY 2002
|
www.nature.com 755
Rational imitation in preverbal infants
Babies may opt for a simpler way to turn on a light after watching an adult do it.
Figure 1 Comparison of the methods used by 14-month-old infants to switch on a light-box 1 week after watching how an adult executed
the same task under two different conditions. a, b, Adult switching on the light by touching the lamp with her forehead in the hands-occupied
condition (a,
n
414) or the hands-free condition (b,
n
413). c, Methods used by infants to switch on the light-box after watching the head
action used by the demonstrator under these two conditions (left bar, adult had hands occupied; right bar, adult had hands free), recorded
over a 20-s period. Blue, head action was re-enacted; green, only manual touch was used. Further details are available from the authors.
0%
20%
40%
60%
80%
100%
Hands
occupied
Hands
free
a
bc
Evolutionary biology
Performance constraints
in decathletes
P
hysical performance by vertebrates is
thought to be constrained by trade-
offs between antagonistic pairs of
ecologically relevant traits and between
conflicting specialist and generalist pheno-
types
1,2
, but there is surprisingly little
evidence to support this reasoning
3–5
. Here
we analyse the performance of world-class
athletes in standardized decathlon events
and find that it is subject to both types of
trade-off, after correction has been made
for differences between athletes in general
ability across all 10 events. These trade-offs
may have imposed important constraints
on the evolution of physical performance in
humans and other vertebrates.
Decathletes compete in 10 different
track and field events over two consecutive
days. We constructed a data set of the
performance of 600 world-class decathletes
across all events from sources available on
the Internet (see Fig. 1 legend). Individual
performance for any pair of disciplines was
positively correlated for the entire data set.
This was unexpected, as physiological and
biomechanical theory predicts that there
should be trade-offs between certain pairs
© 2002 Macmillan Magazines Ltd
tion coefficients for each pair of perfor-
mance traits and found evidence of some
performance trade-offs (Table 1). For
example, good sprinters performed relative-
ly poorly over 1,500 metres, but did well in
the long jump, the 400 metres and the
110-metre hurdles. Likewise, there was a
negative correlation between performance
in shot-putting (which calls for explosive
power in the upper limbs) and in the
1,500 metres (which depends on endurance
in the lower limbs).
These results seem to support the idea of
trade-offs generated by conflicting anat-
omical (such as relative body weight and
limb proportions) and muscle-fibre-type
requirements. Negative correlations may
also reflect differences in training, for
example if athletes focus on certain disci-
plines to the detriment of others. However,
the consistency with which particular
performance traits co-vary suggests that it
is more difficult to combine excellence in
some pairs of events than in others.
We calculated measures of excellence
and average performance for each decath-
lete across all disciplines to test the assump-
tion that there must be a performance
trade-off between specialist and generalist
phenotypes. The principle of allocation, a
hypothesis that underlies much of ecologi-
cal and evolutionary theory, predicts that
excellence in one task can only be attained
at the expense of average performance in all
other tasks, and vice versa
1
.
For each athlete, we selected the highest
standardized score from his set of 10 events
and defined this as that athletes ‘degree of
excellence. We also calculated the average
of the 10 scores for each athlete to estimate
their overall average performance. The
degree of excellence was negatively correlat-
ed with average performance across all
disciplines (r410.37, P*0.00001; Fig. 1),
which is consistent with the principle of
allocation.
We conclude that in an environment in
which the selection criterion is combined
high performance across multiple tasks,
increased performance in one function may
impede performance in others. The rel-
evance of our findings for the evolution of
performance traits will depend on whether
these correlations have a genetic basis —
studies of mice
9
and insects
10
indicate that
this may be the case. The influence of an
individual athlete’s training schedule or
career development on these apparent
trade-offs also needs to be determined.
Raoul Van Damme, Robbie S. Wilson,
Bieke Vanhooydonck, Peter Aerts
Department of Biology, University of Antwerp,
Universiteitsplein 1, 2610 Wilrijk, Belgium
e-mail: rwilson@uia.ac.be
1. Levins, R. Evolution in Changing Environments (Princeton
Univ. Press, New Jersey, 1968).
2. Futuyma, D. J. & Moreno, G. Annu. Rev. Ecol. Syst. 19,
207–233 (1988).
3. Garland, T. Jr in Quantitative Genetic Studies of Behavioral
Evolution (ed. Boake, C. R. B.) 251–277 (Univ. Chicago Press,
Chicago, 1994).
4. McPeek, M. A. Am. Nat. 148 (suppl.), 124–138 (1996).
5. Whitlock, M. C. Am. Nat. 148 (suppl.), 65–77 (1996).
6. Vanhooydonck, B., Van Damme, R. & Aerts, P. Evolution 55,
1040–1048 (2001).
7. Esbjörnsson, M., Sylvén, C., Holm, I. & Jansson, E. Int. J. Sports
Med. 14, 257–263 (1993).
8. Rivero, J. L., Serrano, A. L., Henckel, P. & Agüera, E.
J. Appl. Physiol. 75, 1758–1766 (1993).
9. Dohm, M. R., Hayes, J. P. & Garland, T. Jr Evolution 50,
1688–1701 (1996).
10.Gilchrist, G. W. Evolution 50, 1560–1572 (1996).
Competing financial interests: declared none.
756 NATURE
|
VOL 415
|
14 FEBRUARY 2002
|
www.nature.com
of performance traits — for example, speed
depends on the athlete having a high
proportion of fast, fatigue-sensitive muscle
fibres, whereas endurance relies on a higher
proportion of slower fibres that are more
resistant to fatigue
3,6–8
.
However, when the analysis was restrict-
ed to athletes of comparable ranking (on
the basis of International Amateur Athletic
Federation scores), trade-offs became evi-
dent between certain traits. For example,
performance in the 100 metres is negatively
correlated with speed over 1,500 metres for
athletes who score above 8,000 points
(n4133; r410.21, P40.016). We suspect
that the apparent contradiction in our
results is caused by differences in ‘general
quality’ between athletes, which effectively
mask any trade-off effects when the entire
population is examined.
To account for these differences in gen-
eral quality, we calculated partial correla-
brief communications
Figure 1 Trade-off between excellence in a single discipline (specialist) and average performance across all 10 events (generalist) for
600 world-class decathletes. Excellence in a particular discipline (represented by the highest residual score for a particular athlete) is
negatively correlated with average performance (expressed as the average of all 10 residual scores for that athlete). To ensure that scores
in different events received equal weighting, an athlete’s score was standardized for each event by subtracting it from the population
mean and dividing it by the population standard deviation. The decathlete data set was constructed from the following websites:
www.iaaf.org; www.gbrathletics.com; www.athletics.org.au; www.dlv-sport.de; www.athle.org.
Maximal excellence in a particular discipline ('specialist')
01234
Overall performance ('generalist')
–4
–3
–2
–1
0
1
2
Table 1 Correlation between 10 performance measures for 600 world-class decathletes
Shot putt Javelin Pole vault High jump Long jump 100 m 110-m hurdles 400 m 1,500 m
Discus &&NS NS NS NS NS NS NS
Shot putt & NS NS NS NS NS NS 1
Javelin NS NS NS NS NS NS NS
Pole vault & NS NS & NS NS
High jump & NS & NS NS
Long jump &&NS NS
100 m &&1
110-m hurdles & NS
400 m &
Decathlon competitions consist of: first day, 100 metres, long jump, shot putt, high jump and 400 metres; second day, 110-metre hurdles, discus, pole vault,
javelin and 1,500 metres. Partial correlation coefficients were used to adjust correlations between performance pairs for an athlete’s overall ability on the
basis of the other events (regardless of physiological similarity between traits). For each decathlete, all performance results are taken from their best
decathlon score. All data (units in m or in m s
11
) were log-transformed. Significant positive (&) and negative (1) correlations (
P
*0.0001) were intact
after Bonferroni correction (d.f., 590); NS, not significant.
brief communications is intended to provide a forum
for brief, topical reports of general scientific interest and
for technical discussion of recently published material of
particular interest to non-specialist readers. Priority will
be given to contributions that have fewer than 500
words, 10 references and only one figure. Detailed
guidelines are available on
Nature
s website
(www.nature.com) or on request from
nature@nature.com
© 2002 Macmillan Magazines Ltd
... Understanding the behavior of others Attribution theory (Weiner, 2018) Interpreting emotions from faces (Todorov, Olivola, Dotsch, & Mende-Siedlecki, 2015) Implicit processes in evaluating others (Greenwald & Lai, 2020) Obedience (Grzyb, Dolinski, Trojanowski, & Bar-Tal, 2018) Conformity (Schultz, Nolan, Cialdini, Goldstein, & Griskevicius, 2007) Mechanisms of persuasion (Cialdini, 2001) Bystander effects (Fischer et al., 2011) Automatic imitation (Cracco et al., 2018) Emotional contagion (Hatfield, Cacioppo, & Rapson, 1993) Transmission of beliefs and stories (Heath et al., 2001) Role-play experiments (Haney, Banks, & Zimbardo, 1973) Impact of shared attention on cognition, emotion and affiliation (Haj-Mohamadi et al., 2018) Social and cognitive development "Theory of mind" tasks (Wimmer & Perner, 1983) Inferring goals (Gergely, Nadasdy, Csibra, & Biro, 1995;Hamlin, Wynn, & Bloom, 2007) Socialization of moral norms (Grusec & Goodnow, 1994) Children enforcing social norms on others (Schmidt & Tomasello, 2012) Imitation of facial expressions in neonates (Meltzoff & Moore, 1977;Oostenbroek et al., 2016) Imitating aggressive behavior (Bandura, Ross, & Ross, 1961) Rational imitation in infants (Gergely, Bekkering, & Kiraly, 2002) Early shared intentionality (Tomasello & Carpenter, 2007) Helping behavior in infants (Warneken & Tomasello, 2007) ...
Article
Social interaction is both ubiquitous and central to understanding human behavior. Such interactions depend, we argue, on shared intentionality: the parties must form a common understanding of an ambiguous interaction (e.g., one person giving a present to another requires that both parties appreciate that a voluntary transfer of ownership is intended). Yet how can shared intentionality arise? Many well-known accounts of social cognition, including those involving "mind-reading," typically fall into circularity and/or regress. For example, A's beliefs and behavior may depend on her prediction of B's beliefs and behavior, but B's beliefs and behavior depend in turn on her prediction of A's beliefs and behavior. One possibility is to embrace circularity and take shared intentionality as imposing consistency conditions on beliefs and behavior, but typically there are many possible solutions and no clear criteria for choosing between them. We argue that addressing these challenges requires some form of we-reasoning, but that this raises the puzzle of how the collective agent (the "we") arises from the individual agents. This puzzle can be solved by proposing that the will of the collective agent arises from a simulated process of bargaining: agents must infer what they would agree, were they able to communicate. This model explains how, and which, shared intentions are formed. We also propose that such "virtual bargaining" may be fundamental to understanding social interactions. (PsycInfo Database Record (c) 2022 APA, all rights reserved).
Article
Things afford positive, neutral, or negative long-run effects on the replicative probability of the focal individual's genes. At the most general level, values are internal estimates of those effects. Value information steers physiology and behavior in the right direction: approach apple, avoid lion. Thus, value computation is of paramount biological importance. Task analysis suggests there are many prerequisites for valuing things aptly. Here, I focus on two: the need to compute value accurately, and the need to properly feed and integrate value information into the various systems that use value information (e.g., emotion systems). For example, the subjective food value imputed to an apple needs to reflect the nutrient content of the apple (accuracy); the intensity of gratitude aroused if someone gave you an apple needs to reflect the food value imputed to the apple (integration). Here, I evaluate these hypotheses with two preregistered studies. Consistent with the integration hypothesis, there are close correspondences between (i) the food values that participants impute to each of 40 food items (Study 1; goods) and (ii) the social values and the social emotions (including: gratitude, anger, shame, and pride) that result when those food items occur as constituents of broader social events. Similar correspondences are observed when participants evaluate each of 28 diseases and injuries (Study 2; bads). Consistent with the accuracy hypothesis, exploratory analyses indicate that the food values, the social values, and the social emotions elicited by the food items all track the nutrient content of those food items. Valuation is inherently a computational process. For this reason, a computational–functionalist perspective is distinctively suited to spur progress in our understanding of human values.
Article
Few issues have garnered as much attention as that of understanding mechanisms of developmental change. Understanding mechanisms of developmental change is important because it allows researchers to go beyond studying at what age an ability emerges to understanding the processes by which those abilities develop in the first place. Despite the clear importance of mechanisms, the notion of a developmental mechanism or mechanism of developmental change remains largely undefined and there exists no clear guidance on how to study these mechanisms systematically in the developmental literature. Given these outstanding questions, this paper has two main aims. The first aim was to provide a clear definition for mechanism and a discussion about what is meant by development. I argue that the definition of mechanism that is provided better aligns with how most, if not all, developmental scientists think about them. The second goal was to provide concrete suggestions for how developmental scientists might study and test different kinds of mechanisms of developmental change. One of the main arguments of the paper is that there is no one-size-fits-all approach to studying and testing mechanisms of developmental change and that how developmental researchers study them depends on key features of the mechanism in question.
Article
Imagination is a cognitive process used to generate new ideas from old, not just in the service of creativity and fantasy, but also in our ordinary thoughts about alternatives to current reality. In this article, I argue for the central function of imagination in the development of social cognition in infancy and childhood. In Section 1, I review a work showing that even in the first year of life, social cognition can be viewed through a nascent ability to imagine the physical possibilities and physical limits on action. In Section 2, I discuss how imagination of what should happen is appropriately constrained by what can happen, and how this influences children's moral evaluations. In the final section, I suggest developmental changes in imagination—especially the ability to imagine improbable events—may have implications for social inference, leading children to learn that inner motives can conflict. These examples point to a flexible and domain‐general process that operates on knowledge to make social meaning. This article is categorized under: Psychology > Development and Aging Cognitive Biology > Cognitive Development Philosophy > Knowledge and Belief Children's social cognitive development depends on general and increasingly flexible ability to imagine alternative possibilities for action.
Article
Infants have sophisticated knowledge about the physical world, and show enhanced learning about objects that violate physical principles. However, it is unknown whether infants also preferentially learn from the individual who produces an outcome that violates expectations. We investigated whether 15‐month‐old infants (N = 48) selectively imitate individuals who produce surprising outcomes. In Experiment 1, infants watched an experimenter hide a ball and produce an expected outcome in which the ball was revealed where it was hidden, or a surprising outcome in which the ball was revealed in a different location. The experimenter then demonstrated a novel action: using her head to activate a light while her hands were free. Infants imitated that novel action more if the experimenter had previously produced a surprising than an expected outcome. In Experiment 2, infants witnessed the experimenter produce the surprising outcome, then use her head to activate the light while her hands were occupied. Infants did not differentially imitate the head‐touch action relative to either condition in Experiment 1, perhaps indicating a tension between surprise‐induced learning and rational imitation. These experiments show that surprising events are pedagogical opportunities: infants selectively learn from surprising individuals, but also may account for rationality in their surprise‐induced social learning.
Article
Full-text available
Nativist and empiricist approaches require foundationalism because they cannot account for the emergence of representation. Foundationalism is the assumption of an innate representational base. In turn, foundationalism places limits on the nature of learning as a constructivist process. In contrast, action-based approaches can account for the emergence of representation through (inter)action. In so doing, action-based approaches can pursue an emergent constructivism for learning and development. Despite the theoretical symmetry between nativism and empiricism with respect to foundationalism, there is an asymmetry in nativist and empiricist research programs. Nativism generally ignores constructivist complexity that non-nativist approaches assume needs to be investigated empirically. In practice, this means that the plethora of nativist looking time studies do not provide adequate control conditions for the rich interpretations drawn from such research. Instead, it is the a priori assumptions of nativism doing the justification. Without such assumptions, the meaning of the data is unclear at best. Importantly, the problem of a priori assumptions driving rich interpretations is not specific to nativism or looking methodologies. Mindreading as a research program also engages in rich interpretations for studies that concern social-cognition from infancy through preschool. Similarly, these studies do not include the types of control conditions motivated from more constructivist thinking. To the extent that empiricist research programs incorporate constructivist thinking into research, they converge with action-based approaches. This creates a sort of methodological bridge between lean-empiricist research programs and action-based approaches. However, this bridge has limitations that we illustrate through an example concerning maternal mental-state discourse and theory of mind development. The ultimate conclusions are threefold: (1) Action-based approaches are the best theoretical framework for understanding learning and development; (2) constructivist methodology is multiply motivated; (3) there are varying degrees of methodological commensurability between empiricism and action-based approaches
Article
Beginning in infancy, children expect individuals in a group to care for and be loyal to in-group members. One prominent cue that children use to infer that individuals belong to the same group is similarity. Does any salient similarity among individuals elicit an expectation of in-group preference, or does contextual information modulate these expectations? In Experiments 1 and 2, 12-month-old infants expected in-group preference between two individuals who wore the same novel outfit, but they dismissed this similarity if one of the outfits was used to fulfill an instrumental purpose. In Experiment 3, 26-month-old toddlers expected in-group preference between two individuals who uttered the same novel labels, but they dismissed this similarity if the labels were used to convey incidental as opposed to categorical information about the individuals. Together, the results of these experiments ( N = 96) provide converging evidence that from early in life, children possess a context-sensitive mechanism for determining whether similarities mark groups.
Article
Pretense is generally thought to constitutively involve imagination. We argue that this is a mistake. Although pretense often involves imagination, it need not; nor is it a kind of imagination. The core nature of pretense is closer to imitation than it is to imagination, and likely shares some of its motivation with the former. Three main strands of argument are presented. One is from the best explanation of cross‐cultural data. Another is from task‐analysis of instances of pretend play. And the third concerns the different ways in which pretense (especially childhood pretense) and imagination impact one's evaluative/affective systems.
Article
A large body of research has shown that engaging in self-explanation improves learning across a range of tasks. It has been proposed that the act of explaining draws attention and cognitive resources towards evidence that supports good explanations—information that is broad, abstract, and consistent with prior knowledge—which in turn aids discovery and promotes generalization. However, it remains unclear whether explanation impacts the learning process via improved hypothesis generation, increasing the probability that the most generalizable hypotheses are considered in the first place, or hypothesis evaluation, the appraisal of such hypotheses in light of observed evidence. In two experiments with adults, we address this question by separating hypothesis generation and evaluation in a novel category learning task and quantifying the effect of explaining on each process independently. We find that explanation supports learners' generation of broad and abstract hypotheses but does not impact their evaluation of them. These results provide a more precise account of the process by which explanation impacts learning and offer additional support for the claim that hypothesis generation and evaluation play distinct roles in problem solving.
Article
Theory of mind is the human conceptual capacity to understand other people as agents who have subjective mental states such as beliefs, desires, and intentions. It is the basis of distinctively human forms of social understanding and interaction that are essential for communication, cooperation, and culture. In this Review, I summarize the current state of research about the emergence and development of theory of mind in early childhood. I describe the typical developmental trajectory and review findings about the cognitive, linguistic, social and neural foundations of theory of mind development. Finally, I review an ongoing debate regarding whether there are different — implicit versus explicit — forms of theory of mind that develop independently, and conclude by providing an outlook on future challenges and perspectives for research in this area. Theory of mind is the capacity to understand that other people have their own beliefs, desires, and intentions. In this Review, Rakoczy describes the developmental trajectory of theory of mind and discusses the distinction between implicit and explicit forms of this capacity.
Article
In this paper we shall argue that mentalistic action explanations, which form an essential component of a mature theory of mind, are conceptually and developmentally derived from an earlier and purely teleological interpretational system present in infancy. First we summarize our evidence demonstrating teleological action explanations in one-year-olds. Then we shall briefly contrast the structure of teleological vs. causal mentalistic action explanations and outline four logical possibilities concerning the nature of the developmental relationship between them. We shall argue for the view that causal mentalistic action explanations are constructed as useful theoretical extensions of the earlier, purely teleological, nonmentalistic interpretational stance.
Article
This paper reports a habituation study indicating that 12-month-old infants can take the "intentional stance" in interpreting the goal-directed spatial behavior of a rational agent. First, we examine previous empirical claims suggesting that the ability to attribute intentions to others emerges during the second half of the first year. It is argued that neither the perceptual evidence (concerning the early ability to discriminate agents), nor the behavioral data (indicating the use of communicative gestures for instrumental purposes) are sufficient to support such claims about the early appearance of a theory of mind, as there are alternative explanations for these phenomena in terms of simpler psychological processes. It is then suggested that to show that an infant indeed attributes an intention to interpret the goal-directed behavior of a rational agent, one needs to demonstrate that the baby can generate an expectation about the most rational future means action that the agent will perform in a new situation to achieve its goal. We then describe a visual habituation study that meets this requirement. The results demonstrate that based on the equifinal structure of an agent's spatial behavior, 12-month-old infants can identify the agent's goal and interpret its actions causally in relation to it. Furthermore, our study indicates that infants of this age are able to evaluate the rationality of the agent's goal-directed actions, which is a necessary requirement for applying the intentional stance. In closing, we discuss some of the theoretical and methodological implications of our study.
Article
Long-term recall memory was assessed using a nonverbal method requiring subjects to reenact a past event from memory (deferred imitation). A large sample of infants (N = 192), evenly divided between 14- and 16-months old, was tested across two experiments. A delay of 2 months was used in Experiment 1 and a delay of 4 months in Experiment 2. In both experiments two treatment groups were used. In one treatment group, motor practice (immediate imitation) was allowed before the delay was imposed; in the other group, subjects were prevented from motor practice before the delay. Age-matched control groups were used to assess the spontaneous production of the target acts in the absence of exposure to the model in both experiments. The results demonstrated significant deferred imitation for both treatment groups at both delay intervals, and moreover showed that infants retained and imitated multiple acts. These findings suggest that infants have a nonverbal declarative memory system that supports the recall of past events across long-term delays. The implications of these findings for the multiple memory system debate in cognitive science and neuroscience and for theories of infantile amnesia are considered.
Article
We argue that Premack and Premack's criticism of our demonstration (Gergely et al., 1995) of interpreting goal-directed action in one year-olds in terms of the principle of rationality are ill-founded, and their suggested alternative test for goal-attribution is open to lower level interpretations. We show that the alterative model they propose for our data in terms of 'appropriate' change of means action is but a somewhat imprecise restatement of our account of the infant's naive theory of rational action. Finally, we elaborate and clarify our model of the teleological stance in infancy which we suggest is an as yet nonmentalistic precursor of the young child's later emerging causal theory of mind.
Article
The proper domain of naive psychological reasoning is human action and human mental states but such reasoning is frequently applied to non-human phenomena as well. The studies reported in this paper test the validity of the currently widespread belief that this tendency is rooted in the fact that naive psychological reasoning is initially restricted to, and triggered by, the perception of self-initiated movement of agents. We report three habituation experiments which examine the necessary conditions under which infants invoke a psychological principle, namely the principle of rational action, to interpret behaviour as goal directed action. Experiment 1 revealed that the principle of rational action already operates at 9 (but not yet at 6) months of age. Experiment 2 demonstrated that perceptual cues indicating agency, such as self-propulsion, are not necessary prerequisites for interpreting behaviour in terms of the principle of rational action. Experiment 3 confirmed that this effect cannot be attributed to generalisation of agentive properties from one object to another. These results suggest that the domain of naive psychology is initially defined only by the applicability of its core principles and its ontology is not restricted to (featurally identified) object kinds such as persons, animates, or agents. We argue that in its initial state naive psychological reasoning is not a cue-based but a principle-based theory.
  • A N J Meltzoff
  • Exp
Meltzoff, A. N. J. Exp. Child Psychol. 59, 497–515 (1995).
  • J L Rivero
  • A L Serrano
  • P Henckel
  • E Agüera
Rivero, J. L., Serrano, A. L., Henckel, P. & Agüera, E. J. Appl. Physiol. 75, 1758–1766 (1993).
Competing financial interests: declared none
  • G W Gilchrist
Gilchrist, G. W. Evolution 50, 1560–1572 (1996). Competing financial interests: declared none.
  • D J Futuyma
  • G Moreno
Futuyma, D. J. & Moreno, G. Annu. Rev. Ecol. Syst. 19, 207–233 (1988).