- Email: [email protected]
The relation between mirror self-image reactions and imitation in 14- and 18-month-old infants
Infant Behavior & Development 36 (2013) 809–816
Contents lists available at ScienceDirect
Infant Behavior and Development
The relation between mirror self-image reactions and imitation in 14- and 18-month-old infants Norbert Zmyj a,∗ , Wolfgang Prinz b , Moritz M. Daum c a b c
Ruhr-Universität Bochum, Bochum, Germany Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany University of Zurich, Zurich, Switzerland
a r t i c l e
i n f o
Article history: Received 26 September 2012 Received in revised form 29 July 2013 Accepted 11 September 2013 Available online 10 October 2013 Keywords: Imitation Mirror image Testing behavior Self-recognition Contingency
a b s t r a c t Previous research suggests that sensitivity to aspects of the self and others develop in tandem. We tested 14- and 18-month-olds’ imitative abilities and mirror self-image reactions (i.e., testing behavior and passing the mark test). Results showed that 14-month-olds’ imitation was closely related to the occurrence of testing behavior in front of the mirror, where they checked whether they could control the movements of the mirror image. Eighteenmonth-olds, however, no longer showed this relation. Furthermore, in 18-month-olds, we found a high association between imitation and passing the mark test. These correlations suggest that infants’ mirror self-image reactions and imitation share the ability to detect and produce visual-motor contingencies. © 2013 Elsevier Inc. All rights reserved.
1. Introduction Imitation plays a fundamental role in the evolution of human culture (Tomasello, 1999). Very early in ontogeny, infants learn and copy what they observe in their social environment (Barr, Dowden, & Hayne, 1996; Meltzoff, 1988). The extent to which they learn from others depends on a variety of context factors, for example a model’s characteristics, such as age (e.g., Seehagen & Herbert, 2011; Zmyj, Daum, Prinz, Nielsen, & Aschersleben, 2012), a model’s intentionality (Carpenter, Akhtar, & Tomasello, 1998), and the relevance of actions for achieving an external result (Brugger, Lariviere, Mumme, & Bushnell, 2007). It is taken for granted in this line of research that infants are able to imitate observed actions. However, it is still a matter of debate as to how an observed action is transferred into a motor signal (correspondence problem, Heyes, 2001). Since the mechanism for this transformation remains unclear, we refer to imitation as the reproduction of an observed action (Paulus, 2011). The term action refers to object-directed bodily movements as well as to bodily movements that are not directed to an object. In the present study, we investigated infants’ imitation and mirror self-image reactions as two instances of observing and producing actions in order to explore whether they are related in development. 1.1. Visual-motor contingency and imitation Many researchers have suggested that the correspondence problem can be solved by the human cognitive system, as action perception and action production share common representations in that perceived actions are represented in the same format as planned and performed actions (Prinz, 1997). Accordingly, the perception of an action primes its execution.
∗ Corresponding author. Tel.: +49 234 3227783. E-mail address: [email protected] (N. Zmyj). 0163-6383/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.infbeh.2013.09.002
810
N. Zmyj et al. / Infant Behavior & Development 36 (2013) 809–816
This connection is already present in the first years of life (Daum, Prinz, & Aschersleben, 2011; Elsner, 2007; Hauf, Elsner, & Aschersleben, 2004). There are at least two accounts of how this link emerges so early in life. First, the nativist account of imitation suggests that a common link between action perception and action production is present from birth (Meltzoff & Moore, 1997). According to this idea, imitation builds upon the core human ability to represent actions from different modalities in the same supramodal format. This makes the correspondence problem obsolete. Newborns are thought to be able to imitate based on this mechanism (Meltzoff & Moore, 1977). However, there is still controversy as to whether a neonate’s execution of a facial gesture (e.g., tongue protrusion), after having observed this behavior, is based on a direct link between perception and production of these actions (as suggested by Meltzoff & Moore, 1997) or whether this behavior is based on simpler mechanisms (Anisfeld, 2005). For example, infants protrude the tongue when they observe something interesting, and observing a protruded tongue is more interesting than observing an open mouth (Jones, 1996). Second, the empiristic account of imitation suggests that imitation is the result of associative learning (Ray & Heyes, 2011). The basis for this associative learning is that the infant’s execution of an action often coincides with the observation of the same action. Sources for this visual-motor contingency are watching one’s own body moving directly or in reflecting surfaces, which additionally provide visual feedback of one’s own actions that are not directly observable (e.g., facial gestures), and the imitative feedback of parents (Papouˇsek & Papouˇsek, 1979). Infants are able to detect contingency (i.e., the occurrence of one event increases the likelihood that another event will occur) very early in life (Bahrick & Watson, 1985; Rochat & Morgan, 1998; Striano, Henning, & Stahl, 2005). According to the idea of associative learning, imitation builds upon the experience of seeing and doing the same action at the same time (Ray & Heyes, 2011). Despite the diverging viewpoints on the basis of imitation, both accounts suggest that learning is a viable foundation for how infants link self-generated actions to the corresponding visual and proprioceptive feedback. According to the nativist account of imitation, infants learn the relation between self-generated actions and the posture of different body parts via a so-called “body babbling”, according to which infants randomly generate actions and monitor the corresponding afferent feedback (Meltzoff & Moore, 1997). Similarly, the empiristic account of imitation suggests that self-observation is a potential source for linking seeing and doing the same action (Ray & Heyes, 2011). An infant’s ability to detect different forms of contingencies develops gradually over the first year of life for different modalities. At around 5 months of age, infants detect visual-motor contingencies (Bahrick & Watson, 1985; Schmuckler & Fairhall, 2001; Zmyj, Hauf, & Striano, 2009), and at around 10 months of age, they detect visual-tactile contingencies (Zmyj, Jank, Schütz-Bosbach, & Daum, 2011). Imitative abilities likewise develop gradually over the first years of life (Barr et al., 1996; Jones, 2007; Nielsen & Dissanayake, 2004). However, so far, the relation between contingency detection and imitation early in life has not been explored. 1.2. Mirror self-image reactions and imitation When infants observe themselves in the mirror, they are able to experience the direct relation between self-generated actions and the corresponding contingent visual feedback. Before the age of 18 months, infants primarily observe the mirror self-image as they move their body or parts of it (Amsterdam, 1972) and engage in testing behavior by intentionally speeding up or suddenly stopping their movement in order to test whether the mirror self-image displays the same behavior (BischofKöhler, 1989). This testing behavior is an example of how infants monitor the visual consequences of self-generated actions. At around the age of 18 months, infants start to locate a mark on their cheek by using the mirror self-image in the socalled mark test (Amsterdam, 1972). The cognitive prerequisites for passing the mark test are still a matter of debate (for an overview, see Bard, Todd, Bernier, Love, & Leavens, 2006). Some researchers have suggested a high-level interpretation for this ability in the sense of a qualitative improvement in representational development. In other words, infants become capable of forming secondary representations in addition to forming primary representations (Bischof, 1996; Perner, 1991; Suddendorf & Whiten, 2001). Primary representations of objects are experienced as coming from the outside world and carry the tag of “reality” (Leslie, 1987). By contrast, a secondary representation of objects is conceived of as an imagination of objects which are automatically experienced as a mere copy of a real object (Perner, 1991). To succeed in the mark test, infants have to link their mirror image (i.e., primary representation) to their imagined self (i.e., secondary representation). Hence, the two types of representation need to be coordinated in order to recognize the fact that the mirror image reflects their own appearance. This view, however, has been criticized by some researchers, who have suggested a low-level explanation for passing the mark test. For example, Heyes (1994) suggested that what has been described as mirror self-recognition is better described by the term “mirror-guided body inspection”. It merely requires a differentiation between sensory input from the infant’s own body and sensory input originating elsewhere. According to this explanation, individuals use the mirror to inspect the body without necessarily having a self-concept. Interestingly, even accounts that use a high-level interpretation of the mark test do consider this low-level explanation as a basis. It has been suggested that detecting contingencies between self-generated actions and the visual and proprioceptive feedback is the foundation for the later passing of the mark test (e.g., Rochat, 2003). Hence, testing behavior in front of the mirror and passing the mark test might lie along the same line of developmental progression. It should be noted, however, that this assumption still lacks empirical evidence. Despite the controversy regarding the cognitive underpinnings of passing the mark test, it is assumed that advances in using the mirror image are related to advances in imitation of others (Ray & Heyes, 2011; Suddendorf & Whiten, 2001). Some empirical evidence has been obtained for this view. Passing the mark test emerges in tandem with imitation of actions (e.g., Asendorpf & Baudonniere, 1993; Asendorpf, Warkentin, & Baudonniere, 1996; Herold & Akhtar, 2008; Nielsen & Dissanayake,
N. Zmyj et al. / Infant Behavior & Development 36 (2013) 809–816
811
2004). For example, 18-month-olds who engaged in longer periods of synchronic imitation (i.e., observing an experimenter’s action and copying this action at the same time) were also more likely to pass the mark test than infants who did not imitate in this context (Asendorpf et al., 1996). An uncharted territory, however, is the relation between the occurrence of testing behavior in front of the mirror and imitation of others’ actions. According to the empiristic account of imitation, these two behaviors are merely two sides of the same coin. Testing behavior is evidence of the ability to relate self-generated actions to the visual feedback of one’s own actions. Likewise, imitation of others’ actions is evidence of the ability to relate the visual input of others’ actions and self-generated actions. 1.3. The present study The empiristic account of imitation emphasizes that imitation is a learned mechanism to transform observed actions into one’s own actions. Learning about the contingency between an efferent motor signal for a self-generated action and the afferent visual feedback of this action is a possible source for acquiring the ability to generate an efferent motor signal for a self-generated action that is similar to the afferent feedback of an other-generated action. Accordingly, this account would predict a relation between the ability to detect the contingency between self-generated actions and visual feedback on the one hand and the ability to imitate others’ actions on the other hand. By contrast, the nativist account of imitation only suggests a learning mechanism for detecting the contingency between efferent motor signals for self-generated actions to the corresponding afferent feedback. However, this account assumes that there is an innate matching between observing actions of others and performing the same actions, without any contribution from advances in contingency detection between afferent and efferent information of the self. Accordingly, the nativist account of imitation would not predict a relation between the ability to detect the contingency between self-generated actions and the visual feedback on the one hand and the ability to imitate on the other hand. Imitation and self-perception in the mirror can be considered as two instances of seeing and doing the same action. Advances in detecting contingent behavior (i.e., testing behavior and passing the mark test) are then likely to coincide with advances in producing contingent behavior (i.e., imitation). We tested this hypothesis by presenting 14- and 18-month-olds with the mark test (analogous to Amsterdam, 1972) and an imitation task in which the infants were required to imitate a model’s salient component of the behavior (Carpenter, Call, & Tomasello, 2005). In this imitation task, infants observed how a model made a toy animal hop or slide toward one of two locations. The two locations were either salient, with one box present at each location, or not salient, without any boxes present. We tested these age groups because around half of 18-month-olds are expected to pass the mark test (Amsterdam, 1972). By contrast, 14-month-olds are not yet expected to pass the mark test, but half of this age group is expected to show testing behavior in front of the mirror (Amsterdam, 1972). The task, first introduced by Carpenter and colleagues, is ideal because children within the age range of 12–18 months performed at similarly moderate levels on this task, which makes ceiling effects less likely. 2. Method 2.1. Participants Participants were thirty 14-month-olds (mean age = 13 months, 29 days; range = 13 months, 15 days–14 months, 13 days; 14 girls) and twenty-seven 18-month-olds (mean age = 17 months, 26 days; range = 17 months, 13 days – 18 months, 15 days; 14 girls). An additional nineteen 14-month-olds and eighteen 18-month-olds were tested but not included in the final data analysis. As in the original study by Carpenter et al. (2005), the majority of the infants excluded from data analysis consistently responded in irrelevant ways in all trials of the imitation task; for example, they did not touch the toy animal or returned it to the experimenter (nine 14-month-olds and twelve 18-month-olds). Other infants were not included in the final sample due to fussiness (two 14-month-olds and five 18-month-olds), procedural errors (five 14-month-olds and one 18-month-old), or because they refused to look in the mirror (three 14-month-olds). This high attrition rate is analogous to the attrition rate reported by Carpenter et al. (2005). 2.2. Materials In the imitation task, we used a small pink rubber dragon (height 8 cm) and two rectangular boxes (10 cm × 10 cm × 10 cm) of different colors (yellow, green) that were open at the front and at the back. The boxes were fixed to a blue rectangular board (50 cm × 15 cm) 20 cm apart from each other. In the mirror task, we used a full-length mirror (120 cm × 40 cm), red lipstick, a baby doll, a tissue, three multi-color soft cubes, multi-color cups, and an earphone for the parent’s left ear. 2.3. Design and procedure Each participant and parent was first escorted to a reception room. For approximately 10 min, the infant was allowed to explore the room while the research assistant described the test procedure to the parent. All infants were tested individually
812
N. Zmyj et al. / Infant Behavior & Development 36 (2013) 809–816
with one parent present. The test room was unfurnished and contained only a table, two chairs and a mirror, which was hidden behind white curtains during the imitation task. 2.3.1. Imitation task The imitation task was similar to the one described by Carpenter et al. (2005). Infants sat on their parent’s lap at the table, facing the experimenter who sat at the opposite side of the table. Each infant was randomly assigned to one of two conditions (boxes present vs. boxes absent). In the boxes present condition, the boxes were located on the table during the whole task, whereas in the boxes absent condition, the boxes were not used at all for the task. Each condition consisted of four trials with the same demonstration of a dragon being moved from a center position to either the left or the right to a final location. The final location (left or right) and action style (hopping or sliding) of the action demonstrated was counterbalanced in a 2 × 2 design. The final locations had identical coordinates on the table in both the boxes present condition and the boxes absent condition. For the hopping action style, the experimenter made the dragon hop to one location in a straightforward manner. The series of approximately eight jumps was accompanied by the experimenter saying “bee” for each jump (i.e., “beebeebee. . .”). For the sliding action style, the experimenter slid the dragon along the surface of the table to the final location. The sliding was accompanied by the experimenter saying a continuous “beeeeeeeeee” for the duration of the slide action. Each condition consisted of four trials with the same demonstration. Whether the condition was boxes present or absent, with hopping or sliding action style to either a left or right final location, after demonstrating the action using the dragon, the experimenter picked up the dragon and placed it in front of the infant, equidistant to both final locations and told the infant, “Now it’s your turn!” The same sequence was repeated for a total of 4 times and the experimenter waited for 30 s before demonstrating the next action combination. 2.3.2. Mirror task The procedure of the mirror task was an adaptation of the tasks used by Asendorpf and Baudonniere (1993) and Amsterdam (1972). After the imitation task was finished, the table was removed from the room and the experimenter left the room. Parents were given instructions on how to administer the mirror task in the warm-up phase. During the experiment, the experimenter gave instructions to the parent step by step via an earphone from an adjacent room. First, the parent drew back the curtains that hid the mirror. When the infant had taken a look at him/herself in the mirror, the parent took the infant aside and showed the child a baby doll that was marked with a red lipstick dot on its cheek. The parent was asked to give the infant a tissue in order to clean the face of the doll. If the infant did not respond, the parent demonstrated how to remove the dot. Then, under the pretext of blowing the infant’s nose, the parent marked the infant’s cheek with a red dot that was attached to the tissue and the parent was asked to play with the infant out of view of the mirror for 30 s. After that, the parent placed the infant in front of the mirror and put multi-color soft cubes, multi-color cups, and the baby doll in front of the mirror in order to increase the interest of the area in front of the mirror. Infants had 3 min to explore themselves in the mirror. During this phase, the parents sat beside the mirror and were asked to only intervene if the infants left the area in front of the mirror, for example, hiding themselves behind the curtain. In this case, the parents were instructed to place their infants in front of the mirror again. 2.4. Coding 2.4.1. Imitation task The coding was analogous to Carpenter et al. (2005). We coded whether infants matched the demonstrated action in terms of location (left and right) and action style (hopping and sliding). Hopping was coded if the infants made the dragon break contact with the table at least once and sliding was coded when infants made the dragon move on the table without breaking contact with the table. Left and right location were coded when infants put the dragon either in one of the two boxes in the boxes present condition or at one of the corresponding locations on the table in the boxes absent condition. Any other behavior, for example, returning the toy to the experimenter or holding the toy in the hand, was not considered for coding. Only the infants’ first task-related performance (i.e., putting the object in a certain location or reproducing one of the two action styles) on every trial was used for the further analyses. We did not analyze the imitation of sound effects because this variable was not used for comparison with the mirror task. Carpenter et al. (2005) noted that infants copied the experimenter with respect to her perceived goal. That is, in a condition analogous to the boxes present condition, infants usually ignored the action style and copied the location. In contrast, in a condition analogous to the boxes absent condition, infants showed the opposite behavior: They ignored the location and copied the action style. Accordingly, each trial was scored with 1 when the infants correctly copied the location in the boxes present condition. Similarly, each trial was scored with 1 when the infants correctly copied the action style in the boxes absent condition. Each infant received a score between 0 and 4 in the imitation task. 2.4.2. Mirror task As in Asendorpf and Baudonniere (1993), infants were coded as having passed the task when they touched their cheek while looking at their reflection in the mirror or at the parent or when they verbally referred to the self (the child’s name or “I”). Furthermore, the number of testing behaviors in front of the mirror was coded. Testing behavior was defined as a
N. Zmyj et al. / Infant Behavior & Development 36 (2013) 809–816
813
Table 1 Mean number of imitative acts for each age group, condition, and dependant variable. Standard deviations are given in parentheses. Action style
14-month-olds 18-month-olds
Location
Boxes absent
Boxes present
Boxes absent
Boxes present
2.0 (1.4) 1.6 (1.2)
0.8 (1.4) 1.1 (1.1)
0.1 (0.3) 0.1 (0.3)
2.0 (1.2) 1.4 (1.3)
repetitive movement of body parts or a sudden freezing of a movement (as in Meltzoff, 1990) that was accompanied by the infants looking at their reflection in the mirror. 2.5. Reliability A trained observer who was unaware of the hypothesis and of the infants’ performance in the mirror task rated the task performance in the imitation task from videos. A second trained observer recoded 30% of infants (n = 17; nine 14-month-olds and eight 18-month-olds). Good levels of interrater reliability were obtained for action style and location (r = .94 and r = .91, respectively, intraclass correlations). The same coding procedure was applied for the mirror task with different independent coders and different infants. Again, good levels of reliability were obtained for passing the mirror task ( = .88, Cohen’s kappa), number of testing behaviors, and the time spent looking in the mirror (r = .72 and r = .97, respectively, intraclass correlations). The analyses were conducted parametrically (i.e., analysis of variance [ANOVA] and t-tests) when all test criteria were met. In all other cases (e.g., small sample size), we used non-parametrical tests (i.e., Mann–Whitney U-test or Chi-square test). All statistical tests were two-tailed with a p-value of .05. 3. Results 3.1. Imitation task An ANOVA with age and condition as between-subjects factors and the imitation score as dependent variable revealed that infants matched action styles significantly more often in the boxes absent condition than in the boxes present condition, F(2, 54) = 6.5, p < .05, with no interaction between age and condition (see Table 1). In contrast, infants matched the location more often in the box present condition than in the box absent condition, F(2, 54) = 53.0, p < .001, with no interaction between age and condition. For each infant that performed at least one location match or mismatch, a proportion of trials was calculated on which infants matched the experimenter’s demonstrated location. This proportion was tested against chance level (50%). Infants matched the experimenter’s location in 67% (SD = 39%) of all cases which is significantly above chance, t(29) = 2.36, p < .05. The overall imitation score did not differ between both age groups, F(2, 54) = 1.5, p = .23 (see Table 1). 3.2. Mirror task Eighteen-month-olds were more likely to pass the mark test than 14-month-olds, 64% and 10% respectively, 2 (1, N = 57) = 18.9, p < .001. An ANOVA with age and passing the mark test as between-subjects factors revealed a main effect of passing the mark test on time looking in the mirror, M = 21.0 s, SD = 20.1 s for non-testpassers and M = 31.6 s, SD = 19.3 s for testpassers, F(1, 55) = 6.3, p < .05. Infants who passed the mark test spent more time looking in the mirror. However, there was no main effect of age; the time 14-month-olds spent looking in the mirror (M = 24.4 s, SD = 20.6 s) was comparable to the 18-month-olds (M = 31.6 s, SD = 21.6 s), F(1, 55) = 1.4, p = .25. Likewise, the number of times infants exhibited testing behavior did not differ between both age groups (14-month-olds: M = 1.4, SD = 1.5; 18-month-olds; M = 1.3, SD = 1.3), F(1, 55) = .1 p = .78. There was also no main effect of passing the mark test on the number of testing behaviors (non-testpassers: M = 1.3, SD = 1.5; testpassers: M = 1.4, SD = 1.4), F(1, 55) = 0.5, p = .65. 3.3. Relation between tasks The most relevant analysis of the data for the present study is the relation between the performance on the two tasks. The 14-month-old infants who showed testing behavior in front of the mirror (n = 16) had a significantly higher imitation score (M = 2.6, SD = 1.2) compared to the infants who showed no testing behavior (n = 14; M = 1.4, SD = 1.2), t(27) = 2.9, p < .01. Moreover, the number of times infants exhibited testing behavior was significantly correlated with the imitation score, r = .55, p < .001 (Pearson correlation), even after controlling for time looking in the mirror, r = .48, p < .01. This correlation was significant in the boxes absent condition, r = .63, p < .05, and approached significance in the boxes present condition, r = .50, p = .052. There was no association between 14-month-olds’ performance in the mark test and their imitation score, U(N = 30) = 33.5, p = .65, Mann–Whitney U-test. The 18-month-olds who passed the mark test (n = 18) had a higher imitation score (M = 2.1, SD = 1.2) than infants who did not (n = 9; M = 0.7, SD = 0.7), t(27) = 3.28, p < .01. This was the case for the boxes absent condition, U(N = 14) = 8.0; p < .05,
814
N. Zmyj et al. / Infant Behavior & Development 36 (2013) 809–816
Mann–Whitney U-test and for the boxes present condition, U(N = 13) = 5.0; p < .05, There was no relation between the imitation score and the occurrence of testing behavior, neither in all 18-month-olds, r = .07, p = .75, nor in the subgroup of 18-month-olds who did not pass the mark test, r = .26, p = .50. Across both age groups, infants who passed the mark test simultaneously imitated both components of the action, i.e., location and action style (at least once) (M = 24%) more often than infants who did not pass the mark test (M = 3%), p < .05, Fisher’s Exact Test. However, the number of infants imitating both components simultaneously in the 14- and 18-month-old groups showed no difference, M = 7%, and M = 15%, respectively, p = .41, Fisher’s Exact Test. 4. Discussion This is the first study to show that 14-month-olds’ testing behavior in front of a mirror is positively correlated with the imitation of the salient part of the observed action, a relation that was no longer present at the age of 18 months. Furthermore, the present results show that 18-month-olds who pass the mark test are more likely to imitate the salient part of the observed action than 18-month-olds who do not pass the mark test (for similar findings, see Asendorpf & Baudonniere, 1993; Asendorpf et al., 1996; Herold & Akhtar, 2008). The results further demonstrate that imitation of the complete action (i.e., action style and location) is correlated with passing the mark test and not with age. Hence, two different mirror selfimage reactions, testing behavior in 14-month-olds and passing the mark test (independent of age), are each correlated with imitation. The results of the present study are in line with the empiristic account of imitation. This account predicts that the detection of visual-motor contingencies in the mirror as measured with testing behavior and passing the mark test both coincide with imitation. The increased competence in relating one’s own behavior to the corresponding visual feedback possibly allows infants to reproduce the observed behavior of others more precisely. By contrast, the results of the present study are not in line with the nativist account of imitation. This account presupposes that infants are able to directly link observed actions of others to their own motor system by day 1 of life. However, infants would have to learn the relation between self-generated actions and the corresponding afferent feedback. According to the nativist account, no relation should be found between detecting visual-motor contingencies and imitation. The findings of the present study suggest that the seemingly different behaviors, testing behavior and passing the mark test, are based on the increasing ability to relate efferent motor and afferent visual information to each other (Ray & Heyes, 2011; Rochat, 2003). Nevertheless, it is a matter of debate whether passing the mark test represents a quantitative gain in detecting visual-motor contingencies (e.g., Heyes, 1994; Mitchell, 1993) or a qualitative gain in the ability to form secondary representations (Suddendorf & Whiten, 2001). In fact, there is some evidence that 18-month-olds form expectations about their own appearance in the mirror (Nielsen, Suddendorf, & Slaughter, 2006). We favor the former interpretation for explaining the present results because it provides a leaner account. However, the present results are also compatible with the latter idea that infants need to form a self-concept in order to pass the mark test (e.g., Asendorpf & Baudonniere, 1993; Asendorpf et al., 1996; Herold & Akhtar, 2008; Nielsen & Dissanayake, 2004). Since this ongoing controversy has not yet been resolved, an important avenue for future research is to investigate longitudinally the role of contingency detection as a potential precursor for the ability to pass the mark test. It should be emphasized that there are at least two alternative explanations for the correlation between the mirror selfimage reactions and the imitation of the salient part of the observed action. First, the correlation between imitation and the mirror task reactions might be based on differences in attentional capacities in infants. In other words, highly attentive infants might follow the behavior demonstrated in the imitation task more precisely, as well as their own mirror image in the mirror task. In turn, the closer observation of the model and their mirror image might have led to more imitative acts on the one hand, and to more testing behaviors or an increased likelihood of passing the mark test on the other hand. Indeed, the focus of attention has been shown to play an important role in infant imitation (Beisert et al., 2012). However, we found a high correlation between the imitation score and the testing behavior even after controlling for time spent looking in the mirror, which makes a purely attention-based interpretation less likely. Second, it might be argued that the correlation between 18-month-olds’ performance in the mirror task and the imitation task could depend on general age-related cognitive development and not on a specific maturation of visual-motor associations. However, the imitative behavior in the present task is not sensitive to the age-related cognitive development between 14 and 18 months of age: We found that the imitation of the complete action demonstrated (i.e., action style and location) was associated with passing the mark test but not with age. That is, infants who passed the test were more likely to imitate the complete action compared to those who did not. Infants who imitated both components of the action thus showed a better sensitivity to the demonstrator’s behavior. They not only imitated the salient component of the action (i.e., style in the box-absent condition; location in the box-present condition), but also imitated the less salient action. We found that the imitation score did not simply increase with age (see Carpenter et al., 2005, for similar findings); rather, we observed a specific correlation between imitation and testing behavior in 14-month-olds, as well as between imitation and passing the mark test in 18-month-olds. The question of why 18-month-olds’ testing behavior was not related to their imitation score remains open. One could speculate that the 18-month-olds were not as interested as 14-month-olds in the visual-motor contingencies. However, it is unclear why the sub-group of 18-month-olds who did not pass the mark test did not show a correlation between testing behavior and imitation. This lack of correlation might be due to the small sample size of this group (n = 9), and could also
N. Zmyj et al. / Infant Behavior & Development 36 (2013) 809–816
815
give rise to speculation that testing behavior is not a universal precursor for passing the mark test in all infants. Related to this, it remains open whether testing behavior and passing the mark test are indeed causally related. Our interpretation rests upon theoretical accounts that draw this developmental line (Ray & Heyes, 2011; Rochat, 2003). It is clear, however, that these issues warrant further exploration. In the present study, we described the infants’ behavior in the imitation task on a behavioral level: Infants preferentially imitate the most salient parts of the observed action (i.e., the location in the box-present condition, and the action style in the box-absent condition). There is indeed a more mentalistic interpretation that can be drawn, namely that infants’ imitation of the salient part of the observed action reflects their understanding of the model’s goal (Carpenter et al., 2005). Although we do believe this to be a possible interpretation, we chose a behavioral description since the focus of the present research was to investigate the link between the perception and production of actions. If we conceived inferring a model’s mental states as a central part of imitation, it would interfere with the aim of the present study to investigate the mechanism involved in imitation (Paulus, 2011). The investigation of how infants’ perception of the self influences their imitation of others lies at the heart of developmental psychology. Using a cross-sectional design, the current study shows that different aspects of mirror self-image reactions (i.e., testing behavior and passing the mark test) and imitation are closely related. We suggest that these abilities share cognitive capacities, namely the ability to detect and produce visual-motor contingencies. Acknowledgements This study was supported by the Max Planck Society. The authors thank Caterina Boettcher for the acquisition of the infants, and our student research assistants for help in data collection. The authors would also like to thank the infants and parents for participating in this project. References Amsterdam, B. (1972). Mirror self-image reactions before age two. Developmental Psychobiology: 5., 297–305. http://dx.doi.org/10.1002/dev.420050403 Anisfeld, M. (2005). No compelling evidence to dispute Piaget’s timetable of the development of representational imitation in infancy. In Perspectives on imitation: From neuroscience to social science: Vol. 2: Imitation, human development, and culture (pp. 107–131). Cambridge, MA, USA: MIT Press. Asendorpf, J. B., & Baudonniere, P.-M. (1993). Self-awareness and other-awareness: Mirror self-recognition and synchronic imitation among unfamiliar peers. Developmental Psychology: 29., 88–95. http://dx.doi.org/10.1037/0012-1649.29.1.88 Asendorpf, J. B., Warkentin, V., & Baudonniere, P.-M. (1996). Self-awareness and other-awareness II: Mirror self-recognition, social contingency awareness, and synchronic imitation. Developmental Psychology: 32., 313–321. http://dx.doi.org/10.1037/0012-1649.32.2.313 Bahrick, L. E., & Watson, J. S. (1985). Detection of intermodal proprioceptive-visual contingency as a potential basis of self-perception in infancy. Developmental Psychology: 21., 963–973. http://dx.doi.org/10.1037/0012-1649.21.6.963 Bard, K. A., Todd, B. K., Bernier, C., Love, J., & Leavens, D. A. (2006). Self-awareness in human and chimpanzee infants: What is measured and what is meant by the mark and mirror test? Infancy: 9., 191–219. http://dx.doi.org/10.1207/s15327078in0902 6 Barr, R., Dowden, A., & Hayne, H. (1996). Developmental changes in deferred imitation by 6- to 24-month-old infants. Infant Behavior & Development: 19., 159–170. http://dx.doi.org/10.1016/S0163-6383(96)90015-6 Beisert, M., Zmyj, N., Liepelt, R., Jung, F., Prinz, W., & Daum, M. M. (2012). Rethinking ‘rational imitation’ in 14-month-old infants: A perceptual distraction approach. PLoS ONE: 7., e32563. http://dx.doi.org/10.1371/journal.pone.0032563 Bischof, N. (1996). Das Kraftfeld der Mythen [The force field of myths]. München: Piper. Bischof-Köhler, D. (1989). Spiegelbild und Empathie. Die Anfänge der sozialen Kognition [Mirror image and empathy: The emergence of social cognition]. Bern: Hans Huber. Brugger, A., Lariviere, L. A., Mumme, D. L., & Bushnell, E. W. (2007). Doing the right thing: infants’ selection of actions to imitate from observed event sequences. Child Development: 78., 806–824. http://dx.doi.org/10.1111/j.1467-8624.2007.01034.x Carpenter, M., Akhtar, N., & Tomasello, M. (1998). Fourteen-through 18-month-old infants differentially imitate intentional and accidental actions. Infant Behavior & Development: 21., 315–330. http://dx.doi.org/10.1016/S0163-6383(98)90009-1 Carpenter, M., Call, J., & Tomasello, M. (2005). Twelve- and 18-month-olds copy actions in terms of goals. Developmental Science: 8., F13–F20. http://dx.doi.org/10.1111/j.1467-7687.2004.00385.x Daum, M. M., Prinz, W., & Aschersleben, G. (2011). Perception and production of object-related grasping in 6-month-olds. Journal of Experimental Child Psychology: 108., 810–818. http://dx.doi.org/10.1016/j.jecp.2010.10.003 Elsner, B. (2007). Infants’ imitation of goal-directed actions: The role of movements and action effects. Acta Psychologica: 124., 44–59. http://dx.doi.org/10.1016/j.actpsy.2006.09.006 Hauf, P., Elsner, B., & Aschersleben, G. (2004). The role of action effects in infants’ action control. Psychological Research/Psychologische Forschung: 68., 115–125. http://dx.doi.org/10.1007/s00426-003-0149-2 Herold, K. H., & Akhtar, N. (2008). Imitative learning from a third-party interaction: Relations with self-recognition and perspective taking. Journal of Experimental Child Psychology: 101., 114–123. http://dx.doi.org/10.1016/j.jecp.2008.05.004 Heyes, C. (1994). Reflections on self-recognition in primates. Animal Behaviour: 47., 909–919. http://dx.doi.org/10.1006/anbe.1994.1123 Heyes, C. (2001). Causes and consequences of imitation. Trends in Cognitive Sciences: 5., (6), 253–261. http://dx.doi.org/10.1016/S1364-6613(00)01661-2 Jones, S. S. (1996). Imitation or exploration? Young infants’ matching of adults’ oral gestures. Child Development: 67., 1952–1969. http://dx.doi.org/10.2307/1131603 Jones, S. S. (2007). Imitation in infancy: The development of mimicry. Psychological Science: 18., 593–599. http://dx.doi.org/10.1111/ j.1467-9280.2007.01945.x Leslie, A. M. (1987). Pretense and representation: The origins of “theory of mind”. Psychological Review: 94., 412–426. http://dx.doi.org/ 10.1037/0033-295X.94.4.412 Meltzoff, A. N., & Moore, M. (1977). Imitation of facial and manual gestures by human neonates. Science: 198., 75–78. Meltzoff, A. N. (1988). Infant imitation after a 1-week delay: Long-term memory for novel acts and multiple stimuli. Developmental Psychology: 24., 470–476. http://dx.doi.org/10.1037/0012-1649.24.4.470 Meltzoff, A. N. (1990). Foundations for developing a concept of self: The role of imitation in relating self to other and the value of social mirroring, social modeling, and self practice in infancy. In The self in transition: Infancy to childhood (pp. 139–164). Chicago, IL, USA: University of Chicago Press. Meltzoff, A. N., & Moore, M. (1997). Explaining facial imitation: A theoretical model. Early Development & Parenting: 6., 179–192. http://dx.doi.org/10.1002/(SICI)1099-0917(199709/12)6:3/43.0.CO;2-R
816
N. Zmyj et al. / Infant Behavior & Development 36 (2013) 809–816
Mitchell, R. W. (1993). Mental models of mirror-self-recognition: Two theories. New Ideas in Psychology: 11., 295–325. http://dx.doi.org/10.1016/0732-118X(93)90002-U Nielsen, M., & Dissanayake, C. (2004). Pretend play, mirror self-recognition and imitation: A longitudinal investigation through the second year. Infant Behavior & Development: 27., 342–365. http://dx.doi.org/10.1016/j.infbeh.2003.12.006 Nielsen, M., Suddendorf, T., & Slaughter, V. (2006). Mirror self-recognition beyond the face. Child Development: 77., 176–185. http://dx.doi.org/10.1111/j.1467-8624.2006.00863.x Papouˇsek, H., & Papouˇsek, M. (1979). Early ontogeny of human social interaction. In M. von Cranach, K. Foppa, W. Lepenies, & D. Ploog (Eds.), Human ethology (pp. 456–489). Cambridge: Cambridge University Press. Paulus, M. (2011). Imitation in infancy: Conceptual considerations. Theory & Psychology: 21., 849–856. http://dx.doi.org/10.1177/0959354310395990 Perner, J. (1991). Understanding the representational mind. Cambridge, MA, USA: The MIT Press. Prinz, W. (1997). Perception and action planning. European Journal of Cognitive Psychology: 9., 129–154. http://dx.doi.org/10.1080/713752551 Ray, E., & Heyes, C. (2011). Imitation in infancy: The wealth of the stimulus. Developmental Science: 14., 92–105. http://dx.doi.org/ 10.1111/j.1467-7687.2010.00961.x Rochat, P. (2003). Five levels of self-awareness as they unfold early in life. Consciousness and Cognition: An International Journal: 12., 717–731. http://dx.doi.org/10.1016/S1053-8100(03)00081-3 Rochat, P., & Morgan, R. (1998). Two functional orientations of self-exploration in infancy. British Journal of Developmental Psychology: 16., 139–154. Schmuckler, M. A., & Fairhall, J. L. (2001). Visual-proprioceptive intermodal perception using point light displays. Child Development: 72., 949–962. http://dx.doi.org/10.1111/1467-8624.00327 Seehagen, S., & Herbert, J. S. (2011). Infant imitation from televised peer and adult models. Infancy: 16., 113–136. http://dx.doi.org/10.1111/j.1532-7078.2010.00045.x Striano, T., Henning, A., & Stahl, D. (2005). Sensitivity to social contingencies between 1 and 3 months of age. Developmental Science: 8., 509–518. http://dx.doi.org/10.1111/j.1467-7687.2005.00442.x Suddendorf, T., & Whiten, A. (2001). Mental evolution and development: Evidence for secondary representation in children, great apes, and other animals. Psychological Bulletin: 127., 629–650. http://dx.doi.org/10.1037/0033-2909.127.5.629 Tomasello, M. (1999). The cultural origins of human cognition. Cambridge, MA, USA: Harvard University Press. Zmyj, N., Daum, M. M., Prinz, W., Nielsen, M., & Aschersleben, G. (2012). Fourteen-month-olds’ imitation of differently aged models. Infant and Child Development: 21., 250–266. http://dx.doi.org/10.1002/icd.750 Zmyj, N., Hauf, P., & Striano, T. (2009). Discrimination between real-time and delayed visual feedback of self-performed leg movements in the first year of life. Cognition, Brain, Behavior: An Interdisciplinary Journal: 13., 479–489. Zmyj, N., Jank, J., Schütz-Bosbach, S., & Daum, M. M. (2011). Detection of visual-tactile contingency in the first year after birth. Cognition: 120., 82–89. http://dx.doi.org/10.1016/j.cognition.2011.03.001
Contents lists available at ScienceDirect
Infant Behavior and Development
The relation between mirror self-image reactions and imitation in 14- and 18-month-old infants Norbert Zmyj a,∗ , Wolfgang Prinz b , Moritz M. Daum c a b c
Ruhr-Universität Bochum, Bochum, Germany Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany University of Zurich, Zurich, Switzerland
a r t i c l e
i n f o
Article history: Received 26 September 2012 Received in revised form 29 July 2013 Accepted 11 September 2013 Available online 10 October 2013 Keywords: Imitation Mirror image Testing behavior Self-recognition Contingency
a b s t r a c t Previous research suggests that sensitivity to aspects of the self and others develop in tandem. We tested 14- and 18-month-olds’ imitative abilities and mirror self-image reactions (i.e., testing behavior and passing the mark test). Results showed that 14-month-olds’ imitation was closely related to the occurrence of testing behavior in front of the mirror, where they checked whether they could control the movements of the mirror image. Eighteenmonth-olds, however, no longer showed this relation. Furthermore, in 18-month-olds, we found a high association between imitation and passing the mark test. These correlations suggest that infants’ mirror self-image reactions and imitation share the ability to detect and produce visual-motor contingencies. © 2013 Elsevier Inc. All rights reserved.
1. Introduction Imitation plays a fundamental role in the evolution of human culture (Tomasello, 1999). Very early in ontogeny, infants learn and copy what they observe in their social environment (Barr, Dowden, & Hayne, 1996; Meltzoff, 1988). The extent to which they learn from others depends on a variety of context factors, for example a model’s characteristics, such as age (e.g., Seehagen & Herbert, 2011; Zmyj, Daum, Prinz, Nielsen, & Aschersleben, 2012), a model’s intentionality (Carpenter, Akhtar, & Tomasello, 1998), and the relevance of actions for achieving an external result (Brugger, Lariviere, Mumme, & Bushnell, 2007). It is taken for granted in this line of research that infants are able to imitate observed actions. However, it is still a matter of debate as to how an observed action is transferred into a motor signal (correspondence problem, Heyes, 2001). Since the mechanism for this transformation remains unclear, we refer to imitation as the reproduction of an observed action (Paulus, 2011). The term action refers to object-directed bodily movements as well as to bodily movements that are not directed to an object. In the present study, we investigated infants’ imitation and mirror self-image reactions as two instances of observing and producing actions in order to explore whether they are related in development. 1.1. Visual-motor contingency and imitation Many researchers have suggested that the correspondence problem can be solved by the human cognitive system, as action perception and action production share common representations in that perceived actions are represented in the same format as planned and performed actions (Prinz, 1997). Accordingly, the perception of an action primes its execution.
∗ Corresponding author. Tel.: +49 234 3227783. E-mail address: [email protected] (N. Zmyj). 0163-6383/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.infbeh.2013.09.002
810
N. Zmyj et al. / Infant Behavior & Development 36 (2013) 809–816
This connection is already present in the first years of life (Daum, Prinz, & Aschersleben, 2011; Elsner, 2007; Hauf, Elsner, & Aschersleben, 2004). There are at least two accounts of how this link emerges so early in life. First, the nativist account of imitation suggests that a common link between action perception and action production is present from birth (Meltzoff & Moore, 1997). According to this idea, imitation builds upon the core human ability to represent actions from different modalities in the same supramodal format. This makes the correspondence problem obsolete. Newborns are thought to be able to imitate based on this mechanism (Meltzoff & Moore, 1977). However, there is still controversy as to whether a neonate’s execution of a facial gesture (e.g., tongue protrusion), after having observed this behavior, is based on a direct link between perception and production of these actions (as suggested by Meltzoff & Moore, 1997) or whether this behavior is based on simpler mechanisms (Anisfeld, 2005). For example, infants protrude the tongue when they observe something interesting, and observing a protruded tongue is more interesting than observing an open mouth (Jones, 1996). Second, the empiristic account of imitation suggests that imitation is the result of associative learning (Ray & Heyes, 2011). The basis for this associative learning is that the infant’s execution of an action often coincides with the observation of the same action. Sources for this visual-motor contingency are watching one’s own body moving directly or in reflecting surfaces, which additionally provide visual feedback of one’s own actions that are not directly observable (e.g., facial gestures), and the imitative feedback of parents (Papouˇsek & Papouˇsek, 1979). Infants are able to detect contingency (i.e., the occurrence of one event increases the likelihood that another event will occur) very early in life (Bahrick & Watson, 1985; Rochat & Morgan, 1998; Striano, Henning, & Stahl, 2005). According to the idea of associative learning, imitation builds upon the experience of seeing and doing the same action at the same time (Ray & Heyes, 2011). Despite the diverging viewpoints on the basis of imitation, both accounts suggest that learning is a viable foundation for how infants link self-generated actions to the corresponding visual and proprioceptive feedback. According to the nativist account of imitation, infants learn the relation between self-generated actions and the posture of different body parts via a so-called “body babbling”, according to which infants randomly generate actions and monitor the corresponding afferent feedback (Meltzoff & Moore, 1997). Similarly, the empiristic account of imitation suggests that self-observation is a potential source for linking seeing and doing the same action (Ray & Heyes, 2011). An infant’s ability to detect different forms of contingencies develops gradually over the first year of life for different modalities. At around 5 months of age, infants detect visual-motor contingencies (Bahrick & Watson, 1985; Schmuckler & Fairhall, 2001; Zmyj, Hauf, & Striano, 2009), and at around 10 months of age, they detect visual-tactile contingencies (Zmyj, Jank, Schütz-Bosbach, & Daum, 2011). Imitative abilities likewise develop gradually over the first years of life (Barr et al., 1996; Jones, 2007; Nielsen & Dissanayake, 2004). However, so far, the relation between contingency detection and imitation early in life has not been explored. 1.2. Mirror self-image reactions and imitation When infants observe themselves in the mirror, they are able to experience the direct relation between self-generated actions and the corresponding contingent visual feedback. Before the age of 18 months, infants primarily observe the mirror self-image as they move their body or parts of it (Amsterdam, 1972) and engage in testing behavior by intentionally speeding up or suddenly stopping their movement in order to test whether the mirror self-image displays the same behavior (BischofKöhler, 1989). This testing behavior is an example of how infants monitor the visual consequences of self-generated actions. At around the age of 18 months, infants start to locate a mark on their cheek by using the mirror self-image in the socalled mark test (Amsterdam, 1972). The cognitive prerequisites for passing the mark test are still a matter of debate (for an overview, see Bard, Todd, Bernier, Love, & Leavens, 2006). Some researchers have suggested a high-level interpretation for this ability in the sense of a qualitative improvement in representational development. In other words, infants become capable of forming secondary representations in addition to forming primary representations (Bischof, 1996; Perner, 1991; Suddendorf & Whiten, 2001). Primary representations of objects are experienced as coming from the outside world and carry the tag of “reality” (Leslie, 1987). By contrast, a secondary representation of objects is conceived of as an imagination of objects which are automatically experienced as a mere copy of a real object (Perner, 1991). To succeed in the mark test, infants have to link their mirror image (i.e., primary representation) to their imagined self (i.e., secondary representation). Hence, the two types of representation need to be coordinated in order to recognize the fact that the mirror image reflects their own appearance. This view, however, has been criticized by some researchers, who have suggested a low-level explanation for passing the mark test. For example, Heyes (1994) suggested that what has been described as mirror self-recognition is better described by the term “mirror-guided body inspection”. It merely requires a differentiation between sensory input from the infant’s own body and sensory input originating elsewhere. According to this explanation, individuals use the mirror to inspect the body without necessarily having a self-concept. Interestingly, even accounts that use a high-level interpretation of the mark test do consider this low-level explanation as a basis. It has been suggested that detecting contingencies between self-generated actions and the visual and proprioceptive feedback is the foundation for the later passing of the mark test (e.g., Rochat, 2003). Hence, testing behavior in front of the mirror and passing the mark test might lie along the same line of developmental progression. It should be noted, however, that this assumption still lacks empirical evidence. Despite the controversy regarding the cognitive underpinnings of passing the mark test, it is assumed that advances in using the mirror image are related to advances in imitation of others (Ray & Heyes, 2011; Suddendorf & Whiten, 2001). Some empirical evidence has been obtained for this view. Passing the mark test emerges in tandem with imitation of actions (e.g., Asendorpf & Baudonniere, 1993; Asendorpf, Warkentin, & Baudonniere, 1996; Herold & Akhtar, 2008; Nielsen & Dissanayake,
N. Zmyj et al. / Infant Behavior & Development 36 (2013) 809–816
811
2004). For example, 18-month-olds who engaged in longer periods of synchronic imitation (i.e., observing an experimenter’s action and copying this action at the same time) were also more likely to pass the mark test than infants who did not imitate in this context (Asendorpf et al., 1996). An uncharted territory, however, is the relation between the occurrence of testing behavior in front of the mirror and imitation of others’ actions. According to the empiristic account of imitation, these two behaviors are merely two sides of the same coin. Testing behavior is evidence of the ability to relate self-generated actions to the visual feedback of one’s own actions. Likewise, imitation of others’ actions is evidence of the ability to relate the visual input of others’ actions and self-generated actions. 1.3. The present study The empiristic account of imitation emphasizes that imitation is a learned mechanism to transform observed actions into one’s own actions. Learning about the contingency between an efferent motor signal for a self-generated action and the afferent visual feedback of this action is a possible source for acquiring the ability to generate an efferent motor signal for a self-generated action that is similar to the afferent feedback of an other-generated action. Accordingly, this account would predict a relation between the ability to detect the contingency between self-generated actions and visual feedback on the one hand and the ability to imitate others’ actions on the other hand. By contrast, the nativist account of imitation only suggests a learning mechanism for detecting the contingency between efferent motor signals for self-generated actions to the corresponding afferent feedback. However, this account assumes that there is an innate matching between observing actions of others and performing the same actions, without any contribution from advances in contingency detection between afferent and efferent information of the self. Accordingly, the nativist account of imitation would not predict a relation between the ability to detect the contingency between self-generated actions and the visual feedback on the one hand and the ability to imitate on the other hand. Imitation and self-perception in the mirror can be considered as two instances of seeing and doing the same action. Advances in detecting contingent behavior (i.e., testing behavior and passing the mark test) are then likely to coincide with advances in producing contingent behavior (i.e., imitation). We tested this hypothesis by presenting 14- and 18-month-olds with the mark test (analogous to Amsterdam, 1972) and an imitation task in which the infants were required to imitate a model’s salient component of the behavior (Carpenter, Call, & Tomasello, 2005). In this imitation task, infants observed how a model made a toy animal hop or slide toward one of two locations. The two locations were either salient, with one box present at each location, or not salient, without any boxes present. We tested these age groups because around half of 18-month-olds are expected to pass the mark test (Amsterdam, 1972). By contrast, 14-month-olds are not yet expected to pass the mark test, but half of this age group is expected to show testing behavior in front of the mirror (Amsterdam, 1972). The task, first introduced by Carpenter and colleagues, is ideal because children within the age range of 12–18 months performed at similarly moderate levels on this task, which makes ceiling effects less likely. 2. Method 2.1. Participants Participants were thirty 14-month-olds (mean age = 13 months, 29 days; range = 13 months, 15 days–14 months, 13 days; 14 girls) and twenty-seven 18-month-olds (mean age = 17 months, 26 days; range = 17 months, 13 days – 18 months, 15 days; 14 girls). An additional nineteen 14-month-olds and eighteen 18-month-olds were tested but not included in the final data analysis. As in the original study by Carpenter et al. (2005), the majority of the infants excluded from data analysis consistently responded in irrelevant ways in all trials of the imitation task; for example, they did not touch the toy animal or returned it to the experimenter (nine 14-month-olds and twelve 18-month-olds). Other infants were not included in the final sample due to fussiness (two 14-month-olds and five 18-month-olds), procedural errors (five 14-month-olds and one 18-month-old), or because they refused to look in the mirror (three 14-month-olds). This high attrition rate is analogous to the attrition rate reported by Carpenter et al. (2005). 2.2. Materials In the imitation task, we used a small pink rubber dragon (height 8 cm) and two rectangular boxes (10 cm × 10 cm × 10 cm) of different colors (yellow, green) that were open at the front and at the back. The boxes were fixed to a blue rectangular board (50 cm × 15 cm) 20 cm apart from each other. In the mirror task, we used a full-length mirror (120 cm × 40 cm), red lipstick, a baby doll, a tissue, three multi-color soft cubes, multi-color cups, and an earphone for the parent’s left ear. 2.3. Design and procedure Each participant and parent was first escorted to a reception room. For approximately 10 min, the infant was allowed to explore the room while the research assistant described the test procedure to the parent. All infants were tested individually
812
N. Zmyj et al. / Infant Behavior & Development 36 (2013) 809–816
with one parent present. The test room was unfurnished and contained only a table, two chairs and a mirror, which was hidden behind white curtains during the imitation task. 2.3.1. Imitation task The imitation task was similar to the one described by Carpenter et al. (2005). Infants sat on their parent’s lap at the table, facing the experimenter who sat at the opposite side of the table. Each infant was randomly assigned to one of two conditions (boxes present vs. boxes absent). In the boxes present condition, the boxes were located on the table during the whole task, whereas in the boxes absent condition, the boxes were not used at all for the task. Each condition consisted of four trials with the same demonstration of a dragon being moved from a center position to either the left or the right to a final location. The final location (left or right) and action style (hopping or sliding) of the action demonstrated was counterbalanced in a 2 × 2 design. The final locations had identical coordinates on the table in both the boxes present condition and the boxes absent condition. For the hopping action style, the experimenter made the dragon hop to one location in a straightforward manner. The series of approximately eight jumps was accompanied by the experimenter saying “bee” for each jump (i.e., “beebeebee. . .”). For the sliding action style, the experimenter slid the dragon along the surface of the table to the final location. The sliding was accompanied by the experimenter saying a continuous “beeeeeeeeee” for the duration of the slide action. Each condition consisted of four trials with the same demonstration. Whether the condition was boxes present or absent, with hopping or sliding action style to either a left or right final location, after demonstrating the action using the dragon, the experimenter picked up the dragon and placed it in front of the infant, equidistant to both final locations and told the infant, “Now it’s your turn!” The same sequence was repeated for a total of 4 times and the experimenter waited for 30 s before demonstrating the next action combination. 2.3.2. Mirror task The procedure of the mirror task was an adaptation of the tasks used by Asendorpf and Baudonniere (1993) and Amsterdam (1972). After the imitation task was finished, the table was removed from the room and the experimenter left the room. Parents were given instructions on how to administer the mirror task in the warm-up phase. During the experiment, the experimenter gave instructions to the parent step by step via an earphone from an adjacent room. First, the parent drew back the curtains that hid the mirror. When the infant had taken a look at him/herself in the mirror, the parent took the infant aside and showed the child a baby doll that was marked with a red lipstick dot on its cheek. The parent was asked to give the infant a tissue in order to clean the face of the doll. If the infant did not respond, the parent demonstrated how to remove the dot. Then, under the pretext of blowing the infant’s nose, the parent marked the infant’s cheek with a red dot that was attached to the tissue and the parent was asked to play with the infant out of view of the mirror for 30 s. After that, the parent placed the infant in front of the mirror and put multi-color soft cubes, multi-color cups, and the baby doll in front of the mirror in order to increase the interest of the area in front of the mirror. Infants had 3 min to explore themselves in the mirror. During this phase, the parents sat beside the mirror and were asked to only intervene if the infants left the area in front of the mirror, for example, hiding themselves behind the curtain. In this case, the parents were instructed to place their infants in front of the mirror again. 2.4. Coding 2.4.1. Imitation task The coding was analogous to Carpenter et al. (2005). We coded whether infants matched the demonstrated action in terms of location (left and right) and action style (hopping and sliding). Hopping was coded if the infants made the dragon break contact with the table at least once and sliding was coded when infants made the dragon move on the table without breaking contact with the table. Left and right location were coded when infants put the dragon either in one of the two boxes in the boxes present condition or at one of the corresponding locations on the table in the boxes absent condition. Any other behavior, for example, returning the toy to the experimenter or holding the toy in the hand, was not considered for coding. Only the infants’ first task-related performance (i.e., putting the object in a certain location or reproducing one of the two action styles) on every trial was used for the further analyses. We did not analyze the imitation of sound effects because this variable was not used for comparison with the mirror task. Carpenter et al. (2005) noted that infants copied the experimenter with respect to her perceived goal. That is, in a condition analogous to the boxes present condition, infants usually ignored the action style and copied the location. In contrast, in a condition analogous to the boxes absent condition, infants showed the opposite behavior: They ignored the location and copied the action style. Accordingly, each trial was scored with 1 when the infants correctly copied the location in the boxes present condition. Similarly, each trial was scored with 1 when the infants correctly copied the action style in the boxes absent condition. Each infant received a score between 0 and 4 in the imitation task. 2.4.2. Mirror task As in Asendorpf and Baudonniere (1993), infants were coded as having passed the task when they touched their cheek while looking at their reflection in the mirror or at the parent or when they verbally referred to the self (the child’s name or “I”). Furthermore, the number of testing behaviors in front of the mirror was coded. Testing behavior was defined as a
N. Zmyj et al. / Infant Behavior & Development 36 (2013) 809–816
813
Table 1 Mean number of imitative acts for each age group, condition, and dependant variable. Standard deviations are given in parentheses. Action style
14-month-olds 18-month-olds
Location
Boxes absent
Boxes present
Boxes absent
Boxes present
2.0 (1.4) 1.6 (1.2)
0.8 (1.4) 1.1 (1.1)
0.1 (0.3) 0.1 (0.3)
2.0 (1.2) 1.4 (1.3)
repetitive movement of body parts or a sudden freezing of a movement (as in Meltzoff, 1990) that was accompanied by the infants looking at their reflection in the mirror. 2.5. Reliability A trained observer who was unaware of the hypothesis and of the infants’ performance in the mirror task rated the task performance in the imitation task from videos. A second trained observer recoded 30% of infants (n = 17; nine 14-month-olds and eight 18-month-olds). Good levels of interrater reliability were obtained for action style and location (r = .94 and r = .91, respectively, intraclass correlations). The same coding procedure was applied for the mirror task with different independent coders and different infants. Again, good levels of reliability were obtained for passing the mirror task ( = .88, Cohen’s kappa), number of testing behaviors, and the time spent looking in the mirror (r = .72 and r = .97, respectively, intraclass correlations). The analyses were conducted parametrically (i.e., analysis of variance [ANOVA] and t-tests) when all test criteria were met. In all other cases (e.g., small sample size), we used non-parametrical tests (i.e., Mann–Whitney U-test or Chi-square test). All statistical tests were two-tailed with a p-value of .05. 3. Results 3.1. Imitation task An ANOVA with age and condition as between-subjects factors and the imitation score as dependent variable revealed that infants matched action styles significantly more often in the boxes absent condition than in the boxes present condition, F(2, 54) = 6.5, p < .05, with no interaction between age and condition (see Table 1). In contrast, infants matched the location more often in the box present condition than in the box absent condition, F(2, 54) = 53.0, p < .001, with no interaction between age and condition. For each infant that performed at least one location match or mismatch, a proportion of trials was calculated on which infants matched the experimenter’s demonstrated location. This proportion was tested against chance level (50%). Infants matched the experimenter’s location in 67% (SD = 39%) of all cases which is significantly above chance, t(29) = 2.36, p < .05. The overall imitation score did not differ between both age groups, F(2, 54) = 1.5, p = .23 (see Table 1). 3.2. Mirror task Eighteen-month-olds were more likely to pass the mark test than 14-month-olds, 64% and 10% respectively, 2 (1, N = 57) = 18.9, p < .001. An ANOVA with age and passing the mark test as between-subjects factors revealed a main effect of passing the mark test on time looking in the mirror, M = 21.0 s, SD = 20.1 s for non-testpassers and M = 31.6 s, SD = 19.3 s for testpassers, F(1, 55) = 6.3, p < .05. Infants who passed the mark test spent more time looking in the mirror. However, there was no main effect of age; the time 14-month-olds spent looking in the mirror (M = 24.4 s, SD = 20.6 s) was comparable to the 18-month-olds (M = 31.6 s, SD = 21.6 s), F(1, 55) = 1.4, p = .25. Likewise, the number of times infants exhibited testing behavior did not differ between both age groups (14-month-olds: M = 1.4, SD = 1.5; 18-month-olds; M = 1.3, SD = 1.3), F(1, 55) = .1 p = .78. There was also no main effect of passing the mark test on the number of testing behaviors (non-testpassers: M = 1.3, SD = 1.5; testpassers: M = 1.4, SD = 1.4), F(1, 55) = 0.5, p = .65. 3.3. Relation between tasks The most relevant analysis of the data for the present study is the relation between the performance on the two tasks. The 14-month-old infants who showed testing behavior in front of the mirror (n = 16) had a significantly higher imitation score (M = 2.6, SD = 1.2) compared to the infants who showed no testing behavior (n = 14; M = 1.4, SD = 1.2), t(27) = 2.9, p < .01. Moreover, the number of times infants exhibited testing behavior was significantly correlated with the imitation score, r = .55, p < .001 (Pearson correlation), even after controlling for time looking in the mirror, r = .48, p < .01. This correlation was significant in the boxes absent condition, r = .63, p < .05, and approached significance in the boxes present condition, r = .50, p = .052. There was no association between 14-month-olds’ performance in the mark test and their imitation score, U(N = 30) = 33.5, p = .65, Mann–Whitney U-test. The 18-month-olds who passed the mark test (n = 18) had a higher imitation score (M = 2.1, SD = 1.2) than infants who did not (n = 9; M = 0.7, SD = 0.7), t(27) = 3.28, p < .01. This was the case for the boxes absent condition, U(N = 14) = 8.0; p < .05,
814
N. Zmyj et al. / Infant Behavior & Development 36 (2013) 809–816
Mann–Whitney U-test and for the boxes present condition, U(N = 13) = 5.0; p < .05, There was no relation between the imitation score and the occurrence of testing behavior, neither in all 18-month-olds, r = .07, p = .75, nor in the subgroup of 18-month-olds who did not pass the mark test, r = .26, p = .50. Across both age groups, infants who passed the mark test simultaneously imitated both components of the action, i.e., location and action style (at least once) (M = 24%) more often than infants who did not pass the mark test (M = 3%), p < .05, Fisher’s Exact Test. However, the number of infants imitating both components simultaneously in the 14- and 18-month-old groups showed no difference, M = 7%, and M = 15%, respectively, p = .41, Fisher’s Exact Test. 4. Discussion This is the first study to show that 14-month-olds’ testing behavior in front of a mirror is positively correlated with the imitation of the salient part of the observed action, a relation that was no longer present at the age of 18 months. Furthermore, the present results show that 18-month-olds who pass the mark test are more likely to imitate the salient part of the observed action than 18-month-olds who do not pass the mark test (for similar findings, see Asendorpf & Baudonniere, 1993; Asendorpf et al., 1996; Herold & Akhtar, 2008). The results further demonstrate that imitation of the complete action (i.e., action style and location) is correlated with passing the mark test and not with age. Hence, two different mirror selfimage reactions, testing behavior in 14-month-olds and passing the mark test (independent of age), are each correlated with imitation. The results of the present study are in line with the empiristic account of imitation. This account predicts that the detection of visual-motor contingencies in the mirror as measured with testing behavior and passing the mark test both coincide with imitation. The increased competence in relating one’s own behavior to the corresponding visual feedback possibly allows infants to reproduce the observed behavior of others more precisely. By contrast, the results of the present study are not in line with the nativist account of imitation. This account presupposes that infants are able to directly link observed actions of others to their own motor system by day 1 of life. However, infants would have to learn the relation between self-generated actions and the corresponding afferent feedback. According to the nativist account, no relation should be found between detecting visual-motor contingencies and imitation. The findings of the present study suggest that the seemingly different behaviors, testing behavior and passing the mark test, are based on the increasing ability to relate efferent motor and afferent visual information to each other (Ray & Heyes, 2011; Rochat, 2003). Nevertheless, it is a matter of debate whether passing the mark test represents a quantitative gain in detecting visual-motor contingencies (e.g., Heyes, 1994; Mitchell, 1993) or a qualitative gain in the ability to form secondary representations (Suddendorf & Whiten, 2001). In fact, there is some evidence that 18-month-olds form expectations about their own appearance in the mirror (Nielsen, Suddendorf, & Slaughter, 2006). We favor the former interpretation for explaining the present results because it provides a leaner account. However, the present results are also compatible with the latter idea that infants need to form a self-concept in order to pass the mark test (e.g., Asendorpf & Baudonniere, 1993; Asendorpf et al., 1996; Herold & Akhtar, 2008; Nielsen & Dissanayake, 2004). Since this ongoing controversy has not yet been resolved, an important avenue for future research is to investigate longitudinally the role of contingency detection as a potential precursor for the ability to pass the mark test. It should be emphasized that there are at least two alternative explanations for the correlation between the mirror selfimage reactions and the imitation of the salient part of the observed action. First, the correlation between imitation and the mirror task reactions might be based on differences in attentional capacities in infants. In other words, highly attentive infants might follow the behavior demonstrated in the imitation task more precisely, as well as their own mirror image in the mirror task. In turn, the closer observation of the model and their mirror image might have led to more imitative acts on the one hand, and to more testing behaviors or an increased likelihood of passing the mark test on the other hand. Indeed, the focus of attention has been shown to play an important role in infant imitation (Beisert et al., 2012). However, we found a high correlation between the imitation score and the testing behavior even after controlling for time spent looking in the mirror, which makes a purely attention-based interpretation less likely. Second, it might be argued that the correlation between 18-month-olds’ performance in the mirror task and the imitation task could depend on general age-related cognitive development and not on a specific maturation of visual-motor associations. However, the imitative behavior in the present task is not sensitive to the age-related cognitive development between 14 and 18 months of age: We found that the imitation of the complete action demonstrated (i.e., action style and location) was associated with passing the mark test but not with age. That is, infants who passed the test were more likely to imitate the complete action compared to those who did not. Infants who imitated both components of the action thus showed a better sensitivity to the demonstrator’s behavior. They not only imitated the salient component of the action (i.e., style in the box-absent condition; location in the box-present condition), but also imitated the less salient action. We found that the imitation score did not simply increase with age (see Carpenter et al., 2005, for similar findings); rather, we observed a specific correlation between imitation and testing behavior in 14-month-olds, as well as between imitation and passing the mark test in 18-month-olds. The question of why 18-month-olds’ testing behavior was not related to their imitation score remains open. One could speculate that the 18-month-olds were not as interested as 14-month-olds in the visual-motor contingencies. However, it is unclear why the sub-group of 18-month-olds who did not pass the mark test did not show a correlation between testing behavior and imitation. This lack of correlation might be due to the small sample size of this group (n = 9), and could also
N. Zmyj et al. / Infant Behavior & Development 36 (2013) 809–816
815
give rise to speculation that testing behavior is not a universal precursor for passing the mark test in all infants. Related to this, it remains open whether testing behavior and passing the mark test are indeed causally related. Our interpretation rests upon theoretical accounts that draw this developmental line (Ray & Heyes, 2011; Rochat, 2003). It is clear, however, that these issues warrant further exploration. In the present study, we described the infants’ behavior in the imitation task on a behavioral level: Infants preferentially imitate the most salient parts of the observed action (i.e., the location in the box-present condition, and the action style in the box-absent condition). There is indeed a more mentalistic interpretation that can be drawn, namely that infants’ imitation of the salient part of the observed action reflects their understanding of the model’s goal (Carpenter et al., 2005). Although we do believe this to be a possible interpretation, we chose a behavioral description since the focus of the present research was to investigate the link between the perception and production of actions. If we conceived inferring a model’s mental states as a central part of imitation, it would interfere with the aim of the present study to investigate the mechanism involved in imitation (Paulus, 2011). The investigation of how infants’ perception of the self influences their imitation of others lies at the heart of developmental psychology. Using a cross-sectional design, the current study shows that different aspects of mirror self-image reactions (i.e., testing behavior and passing the mark test) and imitation are closely related. We suggest that these abilities share cognitive capacities, namely the ability to detect and produce visual-motor contingencies. Acknowledgements This study was supported by the Max Planck Society. The authors thank Caterina Boettcher for the acquisition of the infants, and our student research assistants for help in data collection. The authors would also like to thank the infants and parents for participating in this project. References Amsterdam, B. (1972). Mirror self-image reactions before age two. Developmental Psychobiology: 5., 297–305. http://dx.doi.org/10.1002/dev.420050403 Anisfeld, M. (2005). No compelling evidence to dispute Piaget’s timetable of the development of representational imitation in infancy. In Perspectives on imitation: From neuroscience to social science: Vol. 2: Imitation, human development, and culture (pp. 107–131). Cambridge, MA, USA: MIT Press. Asendorpf, J. B., & Baudonniere, P.-M. (1993). Self-awareness and other-awareness: Mirror self-recognition and synchronic imitation among unfamiliar peers. Developmental Psychology: 29., 88–95. http://dx.doi.org/10.1037/0012-1649.29.1.88 Asendorpf, J. B., Warkentin, V., & Baudonniere, P.-M. (1996). Self-awareness and other-awareness II: Mirror self-recognition, social contingency awareness, and synchronic imitation. Developmental Psychology: 32., 313–321. http://dx.doi.org/10.1037/0012-1649.32.2.313 Bahrick, L. E., & Watson, J. S. (1985). Detection of intermodal proprioceptive-visual contingency as a potential basis of self-perception in infancy. Developmental Psychology: 21., 963–973. http://dx.doi.org/10.1037/0012-1649.21.6.963 Bard, K. A., Todd, B. K., Bernier, C., Love, J., & Leavens, D. A. (2006). Self-awareness in human and chimpanzee infants: What is measured and what is meant by the mark and mirror test? Infancy: 9., 191–219. http://dx.doi.org/10.1207/s15327078in0902 6 Barr, R., Dowden, A., & Hayne, H. (1996). Developmental changes in deferred imitation by 6- to 24-month-old infants. Infant Behavior & Development: 19., 159–170. http://dx.doi.org/10.1016/S0163-6383(96)90015-6 Beisert, M., Zmyj, N., Liepelt, R., Jung, F., Prinz, W., & Daum, M. M. (2012). Rethinking ‘rational imitation’ in 14-month-old infants: A perceptual distraction approach. PLoS ONE: 7., e32563. http://dx.doi.org/10.1371/journal.pone.0032563 Bischof, N. (1996). Das Kraftfeld der Mythen [The force field of myths]. München: Piper. Bischof-Köhler, D. (1989). Spiegelbild und Empathie. Die Anfänge der sozialen Kognition [Mirror image and empathy: The emergence of social cognition]. Bern: Hans Huber. Brugger, A., Lariviere, L. A., Mumme, D. L., & Bushnell, E. W. (2007). Doing the right thing: infants’ selection of actions to imitate from observed event sequences. Child Development: 78., 806–824. http://dx.doi.org/10.1111/j.1467-8624.2007.01034.x Carpenter, M., Akhtar, N., & Tomasello, M. (1998). Fourteen-through 18-month-old infants differentially imitate intentional and accidental actions. Infant Behavior & Development: 21., 315–330. http://dx.doi.org/10.1016/S0163-6383(98)90009-1 Carpenter, M., Call, J., & Tomasello, M. (2005). Twelve- and 18-month-olds copy actions in terms of goals. Developmental Science: 8., F13–F20. http://dx.doi.org/10.1111/j.1467-7687.2004.00385.x Daum, M. M., Prinz, W., & Aschersleben, G. (2011). Perception and production of object-related grasping in 6-month-olds. Journal of Experimental Child Psychology: 108., 810–818. http://dx.doi.org/10.1016/j.jecp.2010.10.003 Elsner, B. (2007). Infants’ imitation of goal-directed actions: The role of movements and action effects. Acta Psychologica: 124., 44–59. http://dx.doi.org/10.1016/j.actpsy.2006.09.006 Hauf, P., Elsner, B., & Aschersleben, G. (2004). The role of action effects in infants’ action control. Psychological Research/Psychologische Forschung: 68., 115–125. http://dx.doi.org/10.1007/s00426-003-0149-2 Herold, K. H., & Akhtar, N. (2008). Imitative learning from a third-party interaction: Relations with self-recognition and perspective taking. Journal of Experimental Child Psychology: 101., 114–123. http://dx.doi.org/10.1016/j.jecp.2008.05.004 Heyes, C. (1994). Reflections on self-recognition in primates. Animal Behaviour: 47., 909–919. http://dx.doi.org/10.1006/anbe.1994.1123 Heyes, C. (2001). Causes and consequences of imitation. Trends in Cognitive Sciences: 5., (6), 253–261. http://dx.doi.org/10.1016/S1364-6613(00)01661-2 Jones, S. S. (1996). Imitation or exploration? Young infants’ matching of adults’ oral gestures. Child Development: 67., 1952–1969. http://dx.doi.org/10.2307/1131603 Jones, S. S. (2007). Imitation in infancy: The development of mimicry. Psychological Science: 18., 593–599. http://dx.doi.org/10.1111/ j.1467-9280.2007.01945.x Leslie, A. M. (1987). Pretense and representation: The origins of “theory of mind”. Psychological Review: 94., 412–426. http://dx.doi.org/ 10.1037/0033-295X.94.4.412 Meltzoff, A. N., & Moore, M. (1977). Imitation of facial and manual gestures by human neonates. Science: 198., 75–78. Meltzoff, A. N. (1988). Infant imitation after a 1-week delay: Long-term memory for novel acts and multiple stimuli. Developmental Psychology: 24., 470–476. http://dx.doi.org/10.1037/0012-1649.24.4.470 Meltzoff, A. N. (1990). Foundations for developing a concept of self: The role of imitation in relating self to other and the value of social mirroring, social modeling, and self practice in infancy. In The self in transition: Infancy to childhood (pp. 139–164). Chicago, IL, USA: University of Chicago Press. Meltzoff, A. N., & Moore, M. (1997). Explaining facial imitation: A theoretical model. Early Development & Parenting: 6., 179–192. http://dx.doi.org/10.1002/(SICI)1099-0917(199709/12)6:3/43.0.CO;2-R
816
N. Zmyj et al. / Infant Behavior & Development 36 (2013) 809–816
Mitchell, R. W. (1993). Mental models of mirror-self-recognition: Two theories. New Ideas in Psychology: 11., 295–325. http://dx.doi.org/10.1016/0732-118X(93)90002-U Nielsen, M., & Dissanayake, C. (2004). Pretend play, mirror self-recognition and imitation: A longitudinal investigation through the second year. Infant Behavior & Development: 27., 342–365. http://dx.doi.org/10.1016/j.infbeh.2003.12.006 Nielsen, M., Suddendorf, T., & Slaughter, V. (2006). Mirror self-recognition beyond the face. Child Development: 77., 176–185. http://dx.doi.org/10.1111/j.1467-8624.2006.00863.x Papouˇsek, H., & Papouˇsek, M. (1979). Early ontogeny of human social interaction. In M. von Cranach, K. Foppa, W. Lepenies, & D. Ploog (Eds.), Human ethology (pp. 456–489). Cambridge: Cambridge University Press. Paulus, M. (2011). Imitation in infancy: Conceptual considerations. Theory & Psychology: 21., 849–856. http://dx.doi.org/10.1177/0959354310395990 Perner, J. (1991). Understanding the representational mind. Cambridge, MA, USA: The MIT Press. Prinz, W. (1997). Perception and action planning. European Journal of Cognitive Psychology: 9., 129–154. http://dx.doi.org/10.1080/713752551 Ray, E., & Heyes, C. (2011). Imitation in infancy: The wealth of the stimulus. Developmental Science: 14., 92–105. http://dx.doi.org/ 10.1111/j.1467-7687.2010.00961.x Rochat, P. (2003). Five levels of self-awareness as they unfold early in life. Consciousness and Cognition: An International Journal: 12., 717–731. http://dx.doi.org/10.1016/S1053-8100(03)00081-3 Rochat, P., & Morgan, R. (1998). Two functional orientations of self-exploration in infancy. British Journal of Developmental Psychology: 16., 139–154. Schmuckler, M. A., & Fairhall, J. L. (2001). Visual-proprioceptive intermodal perception using point light displays. Child Development: 72., 949–962. http://dx.doi.org/10.1111/1467-8624.00327 Seehagen, S., & Herbert, J. S. (2011). Infant imitation from televised peer and adult models. Infancy: 16., 113–136. http://dx.doi.org/10.1111/j.1532-7078.2010.00045.x Striano, T., Henning, A., & Stahl, D. (2005). Sensitivity to social contingencies between 1 and 3 months of age. Developmental Science: 8., 509–518. http://dx.doi.org/10.1111/j.1467-7687.2005.00442.x Suddendorf, T., & Whiten, A. (2001). Mental evolution and development: Evidence for secondary representation in children, great apes, and other animals. Psychological Bulletin: 127., 629–650. http://dx.doi.org/10.1037/0033-2909.127.5.629 Tomasello, M. (1999). The cultural origins of human cognition. Cambridge, MA, USA: Harvard University Press. Zmyj, N., Daum, M. M., Prinz, W., Nielsen, M., & Aschersleben, G. (2012). Fourteen-month-olds’ imitation of differently aged models. Infant and Child Development: 21., 250–266. http://dx.doi.org/10.1002/icd.750 Zmyj, N., Hauf, P., & Striano, T. (2009). Discrimination between real-time and delayed visual feedback of self-performed leg movements in the first year of life. Cognition, Brain, Behavior: An Interdisciplinary Journal: 13., 479–489. Zmyj, N., Jank, J., Schütz-Bosbach, S., & Daum, M. M. (2011). Detection of visual-tactile contingency in the first year after birth. Cognition: 120., 82–89. http://dx.doi.org/10.1016/j.cognition.2011.03.001