Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Dec 1.
Published in final edited form as: Vision Res. 2010 Sep 7;50(24):2750–2757. doi: 10.1016/j.visres.2010.09.001

Teach to Reach: The Effects of Active Versus Passive Reaching Experiences on Action and Perception

Klaus Libertus 1, Amy Needham 2
PMCID: PMC2991490  NIHMSID: NIHMS234443  PMID: 20828580

Abstract

Reaching is an important and early emerging motor skill that allows infants to interact with the physical and social world. However, few studies have considered how reaching experiences shape infants’ own motor development and their perception of actions performed by others. In the current study, two groups of infants received daily parent guided play sessions over a two-week training period. Using “Sticky Mittens”, one group was enabled to independently pick up objects whereas the other group only passively observed their parent’s actions on objects. Following training, infants’ manual and visual exploration of objects, agents, and actions in a live and a televised context were assessed. Our results showed that only infants who experienced independent object apprehension advanced in their reaching behavior, and showed changes in their visual exploration of agents and objects in a live setting. Passive observation was not sufficient to change infants’ behavior. To our surprise, the effects of the training did not seem to generalize to a televised observation context. Together, our results suggest that early motor training can jump-start infants’ transition into reaching and inform their perception of others’ actions.

Keywords: Infant perception, motor development, perception-action, sticky mittens

An important and yet little studied question in development is how new abilities build upon existing abilities. Developmentalists are often criticized for collecting snapshots of static abilities: indeed, few methods have been devised that allow us to see developmental change in action and measure the consequences of these changes. However, development is a dynamic process that can be shaped by experiences and it is evident that infants readily learn from their own actions (e.g., Adolph, 1997; Campos, et al., 2000; DeCasper & Carstens, 1981; Rovee & Rovee, 1969).

As infants grow and acquire new motor skills, their behaviors and interactions with the world around them change. Attaining new motor skill provides infants with access to new kinds of information and opportunities for learning (Bushnell & Boudreau, 1993; Gibson, 1988). For example, manual exploration strategies determine the kind of information (e.g., shape, texture, or weight) that can be obtained from an object (Lederman & Klatzky, 2009). Advances in postural control increase infants’ interest in objects, ending a period of exclusive face-to-face interactions between mother and infant (Fogel, Messinger, Dickson, & Hsu, 1999). Once infants start to independently engage in manual exploration, they show more advanced object segregation abilities (Needham, 2000) and pay more attention to intermodal properties of objects (Eppler, 1995). Similarly, self-produced locomotion alters how infants respond to and interact with others (e.g., Bertenthal, Campos, & Kermoian, 1994; Clearfield, Osborne, & Mullen, 2008). Following the onset of independent walking, infants engage in more interactive exchanges with their mother (Clearfield, 2010) and show more expressions of emotion (Biringen, Emde, Campos, & Appelbaum, 1995). These findings suggest that acquiring a new motor skill can have influences across different domains of development. However, studies by Adolph and colleagues also show that learning from motor experiences can be quite specific and does not necessarily generalize to novel situations (e.g., Adolph, 1997; Adolph, 2000; Adolph & Berger, 2006). It remains unclear what exactly infants learn from motor experience and how far-reaching this learning really is.

Piaget suggested that self-produced action experiences also contribute the formation of action representations (Piaget, 1953). A number of studies have now shown that first-hand action experiences influence infants’ perception and understanding of actions (for review see Hauf, 2007). For example, infants’ ability to identify the goal of an observed action seems to depend on their own ability and experience (natural or artificial) with performing the same action (Sommerville & Woodward, 2005; Sommerville, Woodward, & Needham, 2005). Further, infants’ own actions on an object seem to increase their interest in actions performed by others on the same object (Hauf, Aschersleben, & Prinz, 2007). Recent findings show that the amount of self-produced object exploration is related to how quickly infants’ orient towards static faces (Libertus & Needham, submitted).

Neurophysiological studies support the involvement of the motor system in action observation. A mirror matching or mirror-neuron system (MNS) has been proposed as a potential neurological basis for the link between action and perception (Rizzolatti & Craighero, 2004). Studies with adults and infants have shown that action experiences and expertise can modulate the response of the MNS (Cross, Hamilton, & Grafton, 2006; van Elk, van Schie, Hunnius, Vesper, & Bekkering, 2008). Further, already in infancy the motor system seems to be involved in predicting the outcome of observed actions (Stapel, Hunnius, van Elk, & Bekkering, 2010). Together, behavioral and neurophyisological findings suggest that changes in infants’ own motor abilities may influence their perception of objects, actors, and actions.

However, determining the unique contributions of experiences on development is challenging. Assessing infants’ behavior once they have independently acquired a new motor skill leaves open the possibility that some general maturational processes may have caused both the acquisition of new motor skills and the changes in behavior. Age is a commonly used estimate for maturation, but age is confounded with experience. Even with age held constant, there are considerable differences between infants with regard to their skills and experiences. For example, while some infants start walking by eight months, others only start around 15 months of age (de Onis, 2006). When assessing behavior in 12-month-old infants, about half of the infants will have experienced walking for a month or longer whereas the other half will not have engaged in walking at all.

To study the effect of experiences while avoiding confounding influences of age, one can compare infants of the same age but with different levels of expertise (e.g., crawling and non-crawling seven-month-olds). Using this approach, studies have shown that infants’ behavior is greatly affected by their first-hand experiences (e.g., Campos, Bertenthal, & Kermoian, 1992). However, experimenters cannot randomly assign infants to the early-crawling and late-crawling groups. Thus, unknown factors could explain differences between the groups.

One way to address the issue of unknown confounding factors is to artificially alter experiences in same-age infants. A classic example of this approach is the kitten carousel study by Held & Hein (1963). Here, kittens were raised in the dark except for brief episodes in a patterned environment. Kittens were either allowed to actively move though the patterned environment or they only passively observed the environment from a sled that was pulled by a kitten from the active group. Both groups experienced roughly the same amount of visual stimulation, but only in the active group was the visual input contingent on the kitten’s own movement. A subsequent test using the “visual cliff” apparatus (Gibson & Walk, 1960) showed that only the actively-obtained visual experience shaped kittens’ visually-guided behaviors. This experiment demonstrated, very impressively, that self-produced actions and contingent feedback provide necessary information required for the development of (in this case) visually-guided behaviors.

The impact of active experiences on development has also been observed with human infants. For example, providing infants who cannot yet reach for objects themselves with reaching experience using ‘sticky mittens’ facilitates exploration behavior and action understanding (e.g., Needham, Barrett, & Peterman, 2002; Sommerville, et al., 2005). However, it is still unknown whether the changes following reaching training are due to the actual physical experience with reaching actions or due to other aspects of the training (such as context, exposure to objects, engagement with the parent or experimenter, or the encouragement provided by the parent). The current research addresses this issue by investigating the effects of active and passive reaching experiences (see Figure 1) on infants’ manual and visual exploration of objects, actors, and actions.

Figure 1.

Figure 1

Open in a new tab

Example of the Active- and Passive-Training procedures. a) Active Training (AT): Toys stick to the mittens upon contact and are moved by the infant. b) Passive Training (PT): Toys are moved by the parent and do not stick to the mittens.

The goal of the current research is to investigate the influences of active and passive ‘sticky mittens’ training experiences on infants’ manual and visual exploration of objects, actors, and actions. We report longitudinal changes in manual exploration behavior (sampling 5–6 times over a two-week period) and compare infants’ visual exploration of objects, actors, and actions in a live (action possible) and televised (observation only) context. Our analysis focused on reaching and grasping behavior, two important developmental milestones that enable infants to obtain objects for further inspection on their own. Despite earlier reports of reaching in newborns (Bower, Broughton, & Moore, 1970), most of the time newborns do not succeed in contacting or obtaining objects with their “prereaching” attempts (von Hofsten, 1982). Successful independent reaching does not emerge until about four to five months after birth (Pomerleau & Malcuit, 1980; von Hofsten & Ronnqvist, 1988). Therefore, the present study tested three-month-old infants who were not able to reach for objects on their own yet.

We predicted that active reaching experience would facilitate reaching and grasping behavior and heighten infants’ attention towards others’ actions (see Needham, et al., 2002; Sommerville, et al., 2005). We did not predict that passive reaching experiences would facilitate reaching and grasping behavior but expected that passive action observation would still increase infants’ attention to and understanding of others’ actions since infant’s repeatedly observed similar actions. It is currently unknown whether self-produced action experiences also influence the perception of others’ actions in a televised context. Children seem to show more difficulties understanding televised events (Troseth & DeLoache, 1998). However, at least by 6 months of age infants process televised and live events similarly (Hofer, Hauf, & Aschersleben, 2007). Therefore, we predicted that self-produced action experiences would alter perception of others’ actions in both a live and a televised context.

Method

Participants

Participants were 58 full-term infants from four groups: two trained groups and two naïve comparisons groups (see Table 1 for details). The “Active Training (AT) group consisted of 18 two- to three-month-old infants who were trained using “Sticky Mittens” for approximately two-weeks. The “Passive Training” (PT) group similarly consisted of 18 two- to three-month-old infants who observed their parent grasp and move objects for approximately two-weeks. A control group of 19 three-month-old infants that received no training (NT3) was additionally recruited. Finally, a control group of 23 five-month-old infants that received no training (NT5) was recruited to compare performance on the visual exploration task.

Table 1.

Group n #F Race Age 1st lab visit Training duration (min.) #of home visits Age 2nd lab visit Parent Edu. Birth weight
AT 18 9 15C, 1A, 2M 10.90 (1.75) 125 (23.70) 3.80 (0.38) 12.92 (1.77) 9.38 (2.99) 3621 (578)
PT 18 10 14C, 1B, 1A, 2M 10.90 (1.52) 144 (23.70) 3.80 (0.331) 12.93 (1.55) 9.94 (2.24) 3544 (470)
NT3 19 8 17C, 1B, 1M -- -- -- 12.61 (2.17) 10.05 (1.61) 3280 (377)
NT5 23 11 18C, 2B, 1A, 2M -- -- -- 19.70 (1.93) 9.60 (1.30) 3418 (515)
Open in a new tab

Note: The total number of participants in each group (n) and the number of females per group (#F) are indicated. All other values are group averages with standard deviations given in parentheses. Age is reported in weeks, birth weight in gram, training duration in minutes. Parents’ education level was assessed on a scale from 0 (no High School degree) to 6 (Post-doctoral Training) for each parent and summed (max. 12). Race abbreviations: C = Caucasian, B = Black or African American, A = Asian, M = More than one race.

Nine infants did not complete the televised-context assessment due to fussiness (AT=3, PT = 1, NT3 = 2, NT5 = 3) and three different infants from the NT3 group did not complete the live-context assessment due to fussiness (n = 1) or technical difficulties with the recording equipment (n = 2).

Stimuli

Live context

Manual and visual exploration in a live context were assessed using a small infant rattle that was easily graspable (0.8×6.4×11.5 cm, H×W×D). While infants were sitting on their parent’s lap, the rattle was presented on a reaching table with a half-circle (radius 23 cm) cut out on the infant’s side (lab visits) or on a reaching tray (home visits; see Figure 2). The reaching tray measured about 30 × 40 cm and had a similar half-circle (radius about 14 cm) cut out. The tray ensured similar testing situations during home and lab visits.

Figure 2.

Figure 2

Open in a new tab

Example of the four-step reaching assessment used in the live context. A small toy was sequentially placed I) beyond reach, II) far but within reach, III) close to hands at midline, and IV) placed into the infant’s hands. Each step lasted about 30 sec. This test was completed once on each lab and home visit. During home visits, a reaching tray as depicted in steps I–III was used.

Televised context

Visual exploration in a televised context was assessed using eight short (23 sec.) video clips presented on a 17” computer screen. In the video clips, an actor was seen with her right hand resting on a table. A bright and salient toy was on the left or right side of the table. After 5 seconds, the actor grasped the toy with her right hand from above and either lifted the toy straight up or slid the toy across the table (either action lasted 5 seconds). Following the action, the actor’s hand returned to its starting position and remained static for 5 seconds. During this sequence the actor looked down at the table in front of her.

Two different toys were used, an orange box with large blue dots and a yellow hexagon with two blue stripes and several small blue dots. Each toy was lifted twice and slid from left to right once and from right to left once. Movies were separated by a fixation stimulus. The actions used in the video clips were similar to actions infant’s would observe in the passive training procedure where parents grasped objects, moved objects over to one side, and lifted objects.

Apparatus

Sticky Mittens

Custom-made infant mittens with Velcro® (loop) sewn to the palm (“Sticky Mittens”, see Needham, et al., 2002) were placed on infants’ hands during training in both the AT and PT groups.

Training toys

A set of six Duplo® blocks were used as training toys in both the AT and PT group. Blocks measured 4.5 cm on each side with a round dome on top and were grouped into three sets of the same color (a singleton, a pair and a triplet). For the AT group, small squares of Velcro® (hook) were attached to the blocks, making them stick to the mittens. For the PT group, small squares of electrical tape were attached to the blocks, making them appear visually similar to the blocks in the AT group. In the PT group, the blocks would not stick to the mittens upon contact.

Eye Tracking

Eye gaze during visual exploration in a televised context was recorded using a remote eye tracker (Tobii 1750). Eye-tracking sessions were conducted in a dimly lit room with the infant sitting in a semi-reclined “bouncy chair” about 60 cm away from a 17” computer screen. To the infant’s right there was a wall covered with black fabric and on the infant’s left there were two large pieces of cardboard also covered with black fabric. Both the experimenter and parent were in the same room but hidden from the infants’ view.

Procedure

Training

In the AT and PT groups, parents were asked to train their infants for 10 minutes each day for a two-week period. Training started on the first lab visit with an experimenter-led demonstration session. Parents then took home all training materials and printed instructions. All remaining training sessions were parent-led except for short experimenter-led training sessions during home-visits to clarify the procedure.

In the AT group, one set of the training blocks (sticky) were first placed in front of the infant. Parents then demonstrated once (per set) that the blocks would stick to the infant’s mittens, placed the blocks back on the table, and drew attention to the blocks by pointing, touching, or commenting about the blocks. Infants had to reach out for the blocks themselves and were allowed to manipulate and shake the blocks for about 10 seconds following contact (see Figure 1a). The blocks were then placed back on the table and the sequence was repeated until a total of 10 minutes had passed (rotating thought the 3 sets of blocks).

The PT group was trained using an “Object Dance” procedure. One set of training blocks (non-sticky) were first placed in front of the infant and parents drew attention to the blocks. Parents then lifted one block, tapped it briefly on the table, moved over to the infant’s left hand, tapped the block on the table again, lifted the block to eye level and briefly touched the palm of the infant’s hand with the block (see Figure 1b). The block was then returned to its starting position and the same sequence was repeated on the infant’s right side. This procedure was repeated until a total of 10 minutes had passed. The PT group experienced similar levels of exposure to the training materials (mittens and blocks) as the AT group but did not engage in self-produced reaching. Wearing mittens in the PT group did prevent infants from grasping and manipulating the training toys (by covering the fingers). Thus, infants in this group were not able to experience self-produced reaching during training.

Assessments

Exploration behavior was assessed in a live context and a televised context. In the live context, a four-step reaching assessment (Figure 2) was used to measure manual and visual exploration in the NT3, AT, and PT groups. The NT5 group did not complete this assessment because their performance was expected to be at ceiling. The NT3 group completed this assessment once. The AT and PT groups completed this assessment 5–6 times, once each lab visit (before training and after two-weeks of training) and on up to 4 home visits (see Table 1). Home visits were conducted every 2–4 days during the two-week training period.

During the four-step reaching assessment, a toy was sequentially placed beyond reach, then far but within reach, then next to the hand, and finally inserted into the infant’s hand (see Figure 2 for schematic). If the infant dropped the toy immediately, up to 3 further attempts were made to place the toy into the infant’s hand. The four steps were always performed in the same order (Beyond, Far, Close, Holding) and each step lasted approximately 30 seconds before the experimenter moved on to the next step.

In the televised context, visual exploration was assessed once in all groups. The televised assessment was always conducted before the live-context assessment. In the AT and PT groups, the televised assessment was conducted following two-weeks of training on lab visit 2.

Measures

Live context

The four-step reaching assessment provided infants’ with different opportunities to interact with the toy. During Step I, infants could only passively look at the toy. Here, our analysis focused on infants’ visual attention to the toy and the experimenter. During both Steps II and III, the toy was placed within reach these steps were combined focus infants’ reaching and grasping actions. During Step IV, the toy was placed into the infant’s hands and no reaching was necessary. Here, our analysis focused on how frequently infants shifted their visual attention back and forth from the toy (looking episodes).

Televised context

The televised-context assessment provided similar measures as Steps I and IV of the live-context assessment: 1) attention to the toy or the actor (see Step I live), and 2) shifts in attention between the actor’s face and the toy (see Step IV live). This enabled us to compare visual exploration across different contexts: Live observation (Step I), Live hands-on (Step IV), and Televised observation. Eye tracking allows for more fine-grained analyses than looking-duration and switches (e.g., anticipation, latency). However, for comparison between live and televised context we decided to focus on measures that were available in both contexts for the present analysis.

Coding

Infants’ motor behavior was coded by trained observers using frame-by-frame coding software (Libertus, 2008). The following behaviors were assessed: looking at the toy or at the experimenter or elsewhere, reaching for the toy (extending hands towards toy while looking at it), touching the toy, grasping the toy (touching toy and bringing one corner of the toy off the table), bi-manual exploration, swatting at the toy, and mouthing the toy. Two different observers coded a random sample of 41 trials using frame-by-frame coding software and overall reliability was high (r = .90). To control for spurious oversampling and to compare our coding method to previous approaches, all data from the second lab visit (AT, PT, and NT3 groups, total 208 trials) were coded for looking at the toy and touching the toy using real-time coding software (see Needham, et al., 2002). Correlations between the different coding methods were high (touching: r = .90; looking: r = .75). For all analyses reported here, scores of the frame-by-frame coding procedure were used.

Results

Live-context assessment

Following the training period, infants in the AT group showed a decrease in attention towards the experimenter on Step I of the manual task, an increase in reaching and grasping behavior on Steps II and III, and an increased number of looking episodes to the toy on Step IV. Results are shown in Figure 3 and reported as proportions of the total trial duration. Gender differences have been reported for early motor behavior with males being more active than females (Eaton & Enns, 1986) but females showing faster and more accurate neonatal imitation skills (Nagy, Kompagne, Orvos, & Pal, 2007). Therefore gender was controlled for in all analyses by inclusion as factor. No effects of gender were found in this study.

Figure 3.

Figure 3

Open in a new tab

Results of the reaching assessment in the live context for a) Step I – Looking duration to experimenter; b) combined Steps II and III – Reaching and grasping behavior; and c) Step IV – Separate looking at toy episodes. Longitudinal patterns are reported from the first lab visit (Lab 1), over up to 4 home visits (Home 1–4), and the second lab visit (Lab 2). Significant differences (p < .05) between the AT and PT group are indicated with a *, differences between AT and untrained 3-month-old infants (NT3) with a †, and significant within-group differences for the AT group (lab visit 1 vs. lab visit 2) are indicated with a #. Error bars represent SEM.

Between-group differences were of key interest for our analysis. Therefore, on each of the lab visits, percentage scores were compared using a 2 (Gender) × 3 (Group) between-subjects analysis of variance (ANOVA). Because not all infants completed all home-visits, the AT and PT groups were compared to each other on each home visit via separate 2 (Gender) × 2 (Group) ANOVAs. On all measures, no significant group differences before training (all ps > .340) were observed. Changes from before to after training (lab 1 vs. lab 2) were expected a priori based on previous findings (see Needham, et al., 2002) and we performed unadjusted (and more powerful) t-test for these within-group comparisons.

Step I – Looking behavior

On Step I of the manual assessment, no group differences with regard to infants’ attention towards the toy were present following training (p = .361). Across the three groups, infants spent about the same amount of time looking at the toy (approximately 65% of the time; other options were experimenter, parent, or distracted). However, looking behavior differed when infants were looking at the experimenter. Significant differences between the AT (M = 6.07%, SD = 9.20) and the PT (M = 28.83%, SD = 29.03) group were present on the 3rd home visit (F(1,32) = 9.439, p = .004, η2 = .227). On the 2nd lab visit, an ANOVA indicated significant between-group differences (F(2,46) = 3.709, p = .032, η2 = .157) and planned between-group comparisons revealed that infants in the AT group (M = 3.22%, SD = 6.41) spent less time looking at the experimenter than infants in the PT group (M = 19.26%, SD = 27.97, p = .012, 95% CI [−28.39, −3.67]). Within-group analyses of behavior before and after training showed a borderline significant decrease in looking at the experimenter for the AT group (t(17) = 2.082, p = .053) but no effects in the PT group (t(17) = −1.054, p = .307).

Steps II and III – Reaching and Grasping behavior

On Steps II and III of the manual assessment, group-level differences were observed by the 3rd home visit where the AT group (M = 23.82%, SD = 29.66) showed significantly more reaching and grasping behavior than the PT group (M = 8.13%, SD = 13.12; F(1,32) = 4.380, p = .044, η2 = .128). Similarly, on the 4th home visit the AT group (M = 25.39%, SD = 28.00) showed more reaching and grasping behavior than the PT group (M = 7.56%, SD = 10.04; F(1,28) = 5.705, p = .024, η2 = .164). Finally, on the 2nd lab visit, an ANOVA revealed significant between-group differences (F(2,46) = 4.380, p = .018, η2 = .156) and planned comparisons indicated that infants in the AT group (M = 29.93%, SD = 27.42) engaged in more reaching and grasping than infants in the PT group (M = 11.79%, SD = 14.69; p = .009, [4.70, 31.57]). Within-group analyses of behavior before and after training revealed a significant increase in reaching and grasping for the AT group (t(17) = 3.875, p = .001) but not for the PT group (t(17) = 1.149, p = .266).

Step IV – Looking Episodes

On Step IV of the manual assessment, there were no group differences with regard to infants’ attention towards the toy or towards the experimenter following training (both ps > .598; but there was a significant effect of gender with males looking more at the experimenter than females). There were differences in how often (# of looking episodes) infants looked at the toy. On the 2nd home visit, the infants in the AT group (M = 3.44, SD = 2.28) produced more separate looking episodes to the toy than infants in the PT group (M = 1.61, SD = 1.14; F(1,32) = 8.630, p = .006, η2 = .206). On the 2nd lab visit, an ANOVA revealed a significant effect of group (F(2,46) = 3.594, p = .035, η2 = .123) and planned between-group comparisons indicated that the infants in the AT group (M = 3.44, SD = 2.28) showed more looking episodes than the infants in the PT group (M = 1.72, SD = 1.78; p = .011, [0.42, 3.03]). Within-group analyses of behavior before and after training confirmed a significant increase in the number of looking episodes for the AT group (t(17) = −2.257, p = .037) but not for the PT group (t(17) = −0.270, p = .790).

Televised-context assessment

Across all groups, there were no between-group differences in infants’ attention to the actor’s face or the toy during observation of televised actions (both ps > .164). However, there was a significant effect of group with regard to the number of gaze shifts (switches) infants produced between the face and the toy (F(3,61) = 2.847, p = .045, η2 = .115). Post-hoc comparisons between all groups (Tukey HSD) revealed that the older NT5 group (M = 28.45, SD = 19.30) produced significantly more switches than the NT3 group (M = 14.71, SD = 10.72; p = .029, 95% CI [1.06, 26.43]). No other group comparisons were significant (all ps > .268). Five-month-olds produced the highest number of gaze shifts between the toy and the face followed by the infants in the AT group (see Figure 4).

Figure 4.

Figure 4

Open in a new tab

Number of face-toy gaze shifts in the televised context. Untrained 3-month-old infants showed the least number of gaze shifts, untrained 5-month-old infants showed the highest number of gaze shifts. Both AT and PT groups fell in-between the younger and older untrained infants and showed similar amounts of gaze shifting. Error bars represent SEM. * p < .05.

Comparison: Live vs. Televised context

Infants’ attention to the face or the toy were assessed across three different contexts: in a live setting with the infant being a passive observer (Step I of the manual task), in a live setting with the infant actively manipulating the toy (Step IV of the manual task), and in a televised setting with the infant being a passive observer (visual task). The events observed in the live and televised context were not identical. Therefore, live and televised context cannot be compared directly. However, influences of context are still of interest and are summarized in Table 2. Regardless of the training experience, changes in context seemed to have similar influences on infants’ behavior with an increase in attention towards faces over toys in the televised context.

Table 2.

Measure Group Context
Live passive Live active TV passive
Face AT 3.23 (6.41) 25.69 (23.76) 48.11 (19.92)
PT 19.26 (27.97) 25.61 (31.03) 44.23 (18.76)
NT3 8.97 (11.55) 21.50 (21.55) 37.69 (17.27)
Toy AT 69.89 (25.75) 45.54 (36.10) 14.37 (10.05)
PT 59.63 (35.45) 53.54 (40.71) 16.43 (9.52)
NT3 70.85 (23.44) 38.17 (32.05) 17.41 (13.62)
Open in a new tab

Note: Average looking duration for all groups are reported as proportion of total looking time with standard deviations given in parentheses.

Discussion

The results reported here show that “Sticky Mittens” training encourages infants’ reaching and grasping behavior and changes their visual exploration of actors and objects in a live context. Two groups of three-month-old infants experienced approximately two hours of object directed training over a two-week period using different training procedures that were designed to be highly similar to each other. The two procedures only differed with regard to the infants’ own actions. In the Active-Training (AT) group, infants were able to contact and move objects themselves. In the Passive-Training (PT) group, infants observed the objects being moved and touched to their hands by their parents. Emphasizing the role of self-produced motor experiences in development, this seemingly small difference in procedure led to significant differences in behavior.

The idea that motor development influences perceptual and cognitive development is not new. Piaget described motor skills as a mechanism that drives development in other domains by generating new sensorimotor experiences (Piaget, 1953, 1954). More recent formulations of this idea describe our cognition and perception as “embodied” or grounded in the body and its actions (e.g., Gibson, 1988; Needham & Libertus, in press; Smith & Gasser, 2005). Nevertheless, it is surprising that infants’ behavior can be changed by only two hours of self-produced motor experiences distributed over two weeks and that closely matched but passive experiences are not sufficient. Several studies have shown that infants are highly sensitive to action-outcome relations and readily learn contingencies between their own behavior and outcomes in the world (e.g., DeCasper & Carstens, 1981; Rovee-Collier, 1999; Thelen, 1994). Experiencing success with their actions during active training may have encouraged and motivated infants to reproduce the outcome with different objects throughout the day and thereby fostered the development of reaching and grasping skills beyond the actual training duration.

Motor skills are important for development in general and affect what kinds of information can be extracted from the environment (e.g.,Bushnell & Boudreau, 1993; Gibson, 1988; Lederman & Klatzky, 2009). Similarly, infants’ engagement in simultaneous visual-manual object exploration predicts their 3D object completion skills (Soska, Adolph, & Johnson, 2010). Our results show that motor experiences affect both infants’ own motor behavior and their visual exploration of observed objects, actors, and actions. During observation in the live context, infants in the AT group showed only little interest in the experimenter while infants in the PT group showed considerable interest in the experimenter. This pattern matches the experiences provided by the AT and PT procedure: In the PT group, adults acted on objects for the infants. In the AT group, infants had to act on objects themselves. When given the opportunity to manipulate a toy in the live context, the AT and PT group showed similar interest in the toy and experimenter. However, infants in the AT group showed more looking back and forth between toy and the surrounding environment (experimenter, parent, room). Looking back and forth allows infants to compare the toy to other objects and to assess the interest of others in the toy. Thus, infants of the AT group showed more interest in actions and interactions between object and environment in a live context. This behavior may facilitate learning about the goals and actions of others.

However, we also observed evidence for specificity of learning from self-produced actions. Even though both motor behavior and visual exploration in a live setting were different between the AT and PT groups, no differences were present during observation in a televised context. Depending on the context experienced by the infant (live vs. televised), visual exploration of objects, actors, and actions seemed to be affected differently by self-produced reaching experiences. Infants in the AT group showed a clear decrease in attention to actors and an increase in looking episodes to objects in a live setting, but in a televised context these infants were not significantly different from their age-matched peers.

We are not the first to report different results between perception of a live and televised actions. In a study with 6–7-month-old infants, Shimada and Hiraki (2006) observed significantly different brain responses during action-observation and during observation of object-motion in a live context but not in a televised context, suggesting that infants process televised events differently than live events. Further, learning from televised actions seems to be harder for young children than learning from live observation (Troseth & DeLoache, 1998). Several factors differed between the live and televised context in our study and could explain the absence of an effect of motor training on perception during the televised context.

First, it is possible that infants had difficulty understanding actions in the televised context because of the impoverished and unnatural stimulus. Learning from their own actions might only generalize to others in similar contexts at first. In previous studies, the context and objects during training and action observation were highly similar (Sommerville, et al., 2005). This similarity should facilitate comparison processes such as structural alignment that could help infants identify correspondences between two scenarios (Gentner & Gunn, 2001; Markman & Gentner, 2000).

Second, infants’ action understanding abilities may not generalize across different postures of the infant. Motor learning is specific to the particular posture experienced during learning (e.g., Adolph, 1997, 2000). In the present study, infants’ posture differed between the manual and visual tasks. During observation in the live context, infants were seated on their parent’s lap in an upright posture. During observation in the televised context, infants were seated in a reclined bouncy chair. It is plausible that the infant action understanding system initially requires infants to be in a posture that would allow them to perform an action themselves in order to become engaged during observation. Supporting this hypothesis, previous studies already observed effects of posture on visual exploration in newborns (Fredrickson & Brown, 1975) and one to six month-old infants (Fogel, et al., 1999). We will investigate this idea in future research.

The findings reported here inform our understanding of the development of infants’ exploratory skills and the connection between these skills and infants’ ability to understand observed actions. Future studies should investigate the connections between action experience and action perception to further clarify the underlying brain circuits. In this domain and others, we need to learn more about how experience influences and shapes infant’s developing abilities.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Adolph KE. Learning in the development of infant locomotion. Monographs of the Society for Research in Child Development. 1997;62(3):I–VI. 1–158. [PubMed] [Google Scholar]
  2. Adolph KE. Specificity of learning: Why infants fall over a veritable cliff. Psychological Science. 2000;11(4):290–295. doi: 10.1111/1467-9280.00258. [DOI] [PubMed] [Google Scholar]
  3. Adolph KE, Berger SE. Motor Development. In: Kuhn D, Siegler RS, editors. Handbook of child psychology: Vol 2: Cognition, perception, and language. 6. New York: Wiley; 2006. pp. 161–213. [Google Scholar]
  4. Bertenthal BI, Campos JJ, Kermoian R. An Epigenetic Perspective on the Development of Self-produced Locomotion and its Consequences. Current Directions in Psychological Science. 1994;3(5):140–145. [Google Scholar]
  5. Biringen Z, Emde RN, Campos JJ, Appelbaum MI. Affective Reorganization in the Infant, the Mother, and the Dyad - the Role of Upright Locomotion and its Timing. Child Development. 1995;66(2):499–514. [PubMed] [Google Scholar]
  6. Bower TGR, Broughton JM, Moore MK. Demonstration of Intention in Reaching Behaviour of Neonate Humans. Nature. 1970;228(5272):679. doi: 10.1038/228679a0. [DOI] [PubMed] [Google Scholar]
  7. Bushnell EW, Boudreau JP. Motor Development and the Mind - the Potential Role of Motor Abilities as a Determinant of Aspects of Perceptual Development. Child Development. 1993;64(4):1005–1021. [PubMed] [Google Scholar]
  8. Campos JJ, Anderson DI, Barbu-Roth MA, Hubbard EM, Hertenstein MJ, Witherington D. Travel Broadens the Mind. Infancy. 2000;1(2):149–219. doi: 10.1207/S15327078IN0102_1. [DOI] [PubMed] [Google Scholar]
  9. Campos JJ, Bertenthal BI, Kermoian R. Early Experience and Emotional Development - the Emergence of Wariness of Heights. Psychological Science. 1992;3(1):61–64. [Google Scholar]
  10. Clearfield MW. Learning to Walk Changes Infants’ Social Interactions. Infant Behavior & Development. 2010 doi: 10.1016/j.infbeh.2010.04.008. Articel in Press, Corrected Proof. [DOI] [PubMed] [Google Scholar]
  11. Clearfield MW, Osborne CN, Mullen M. Learning by looking: Infants’ social looking behavior across the transition from crawling to walking. Journal of Experimental Child Psychology. 2008;100(4):297–307. doi: 10.1016/j.jecp.2008.03.005. [DOI] [PubMed] [Google Scholar]
  12. Cross ES, Hamilton AF, Grafton ST. Building a motor simulation de novo: observation of dance by dancers. Neuroimage. 2006;31(3):1257–1267. doi: 10.1016/j.neuroimage.2006.01.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. de Onis M. WHO Motor Development Study: Windows of achievement for six gross motor development milestones. Acta Paediatrica. 2006;95:86–95. doi: 10.1111/j.1651-2227.2006.tb02379.x. [DOI] [PubMed] [Google Scholar]
  14. DeCasper AJ, Carstens AA. Contingencies of stimulation: Effects on learning and emotion in neonates. Infant Behavior & Development. 1981;4:19–35. [Google Scholar]
  15. Eaton WO, Enns LR. Sex Differences in Human Motor Activity Level. Psychological Bulletin. 1986;100(1):19–28. [PubMed] [Google Scholar]
  16. Eppler MA. Development of manipulatory skills and the deployment of attention. Infant Behavior & Development. 1995;18(4):391–405. [Google Scholar]
  17. Fogel A, Messinger DS, Dickson KL, Hsu HC. Posture and gaze in early mother-infant communication: synchronization of developmental trajectories. Developmental Science. 1999;2(3):325–332. [Google Scholar]
  18. Fredrickson WT, Brown JV. Posture as a Determinant of Visual Behavior in Newborns. Child Development. 1975;46(2):579–582. [PubMed] [Google Scholar]
  19. Gentner D, Gunn V. Structural alignment facilitates the noticing of differences. Memory & Cognition. 2001;29(4):565–577. doi: 10.3758/bf03200458. [DOI] [PubMed] [Google Scholar]
  20. Gibson EJ. Exploratory behavior in the development of perceiving, acting and acquiring of knowledge. Annual Review of Psychology. 1988;(39):1–41. [Google Scholar]
  21. Gibson EJ, Walk RD. The “Visual Cliff”. Scientific American. 1960;202(4):64–72. [PubMed] [Google Scholar]
  22. Hauf P. Infants’ perception and production of intentional actions. In: VonHofsten C, Rosander K, editors. Progress in Brain Research. Vol. 164. Elsevier Science Bv; 2007. pp. 285–301. [DOI] [PubMed] [Google Scholar]
  23. Hauf P, Aschersleben G, Prinz W. Baby do-baby see!: How action production influences action perception in infants. Cognitive Development. 2007;22(1):16–32. [Google Scholar]
  24. Held R, Hein A. Movement-Produced Stimulation in Development of Visually Guided Behavior. Journal of Comparative and Physiological Psychology. 1963;56(5):872. doi: 10.1037/h0040546. [DOI] [PubMed] [Google Scholar]
  25. Hofer T, Hauf P, Aschersleben G. Infants’ perception of goal-directed actions on video. British Journal of Developmental Psychology. 2007;25:485–498. [Google Scholar]
  26. Lederman SJ, Klatzky RL. Haptic perception: A tutorial. Attention Perception & Psychophysics. 2009;71(7):1439–1459. doi: 10.3758/APP.71.7.1439. [DOI] [PubMed] [Google Scholar]
  27. Libertus K. StopFrameCoding (Version 0.9) [Coding Software] Durham, NC: 2008. [Google Scholar]
  28. Libertus K, Needham A. Reaching Experience Accelerates the Development of Social Orienting. (submitted) [Google Scholar]
  29. Markman AB, Gentner D. Structure mapping in the comparison process. American Journal of Psychology. 2000;113(4):501–538. [PubMed] [Google Scholar]
  30. Nagy E, Kompagne H, Orvos H, Pal A. Gender-related differences in neonatal imitation. Infant and Child Development. 2007;16(3):267–276. [Google Scholar]
  31. Needham A. Improvements in Object Exploration Skills May Facilitate the Development of Object Segregation in Early Infancy. Journal of Cognition and Development. 2000;1(2):131–156. [Google Scholar]
  32. Needham A, Barrett T, Peterman K. A pick-me-up for infants’ exploratory skills: Early simulated experiences reaching for objects using ‘sticky mittens’ enhances young infants’ object exploration skills. Infant Behavior & Development. 2002;25(3):279–295. [Google Scholar]
  33. Needham A, Libertus K. Wiley Interdisciplinary Reviews: Cognitive Science. Embodiment in Early Development. (in press) [DOI] [PubMed] [Google Scholar]
  34. Piaget J. The Origin of Intelligence in the Child. New York: Routledge & Kegan Paul Ltd; 1953. [Google Scholar]
  35. Piaget J. The Construction of Reality in the Child. London, England: Routledge and Kegan Paul Ltd; 1954. [Google Scholar]
  36. Pomerleau A, Malcuit G. Development of Cardiac and Behavioral Responses to a 3-Dimensional Toy Stimulation in One-month to 6-month-old Infants. Child Development. 1980;51(4):1187–1196. [PubMed] [Google Scholar]
  37. Rizzolatti G, Craighero L. The mirror-neuron system. Annual Review of Neuroscience. 2004;27:169–192. doi: 10.1146/annurev.neuro.27.070203.144230. [DOI] [PubMed] [Google Scholar]
  38. Rovee CK, Rovee DT. Conjugate Reinforcement of Infant Exploratory Behavior. Journal of Experimental Child Psychology. 1969;8(1):33. doi: 10.1016/0022-0965(69)90025-3. [DOI] [PubMed] [Google Scholar]
  39. Rovee-Collier CK. The development of infant memory. Current Directions in Psychological Science. 1999;8(3):80–85. [Google Scholar]
  40. Shimada S, Hiraki K. Infant’s brain responses to live and televised action. Neuroimage. 2006;32(2):930–939. doi: 10.1016/j.neuroimage.2006.03.044. [DOI] [PubMed] [Google Scholar]
  41. Smith L, Gasser M. The development of embodied cognition: six lessons from babies. Artificial Life. 2005;11(1–2):13–29. doi: 10.1162/1064546053278973. [DOI] [PubMed] [Google Scholar]
  42. Sommerville JA, Woodward AL. Pulling out the intentional structure of action: the relation between action processing and action production in infancy. Cognition. 2005;95(1):1–30. doi: 10.1016/j.cognition.2003.12.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Sommerville JA, Woodward AL, Needham A. Action experience alters 3-month-old infants’ perception of others’ actions. Cognition. 2005;96(1):B1–11. doi: 10.1016/j.cognition.2004.07.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Soska KC, Adolph KE, Johnson SP. Systems in Development: Motor Skill Acquisition Facilitates Three-Dimensional Object Completion. Developmental Psychology. 2010;46(1):129–138. doi: 10.1037/a0014618. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Stapel JC, Hunnius S, van Elk M, Bekkering H. Motor activation during observation of unusual versus ordinary actions in infancy. Social Neuroscience. 2010 doi: 10.1080/17470919.2010.490667. [DOI] [PubMed] [Google Scholar]
  46. Thelen E. Three-month-old Infants can Learn Task-specific Patterns of Interlimb Coordination. Psychological Science. 1994;5(5):280–284. [Google Scholar]
  47. Troseth GL, DeLoache JS. The medium can obscure the message: Young children’s understanding of video. Child Development. 1998;69(4):950–965. [PubMed] [Google Scholar]
  48. van Elk M, van Schie HT, Hunnius S, Vesper C, Bekkering H. You’ll never crawl alone: Neurophysiological evidence for experience-dependent motor resonance in infancy. Neuroimage. 2008;43(4):808–814. doi: 10.1016/j.neuroimage.2008.07.057. [DOI] [PubMed] [Google Scholar]
  49. von Hofsten C. Eye-Hand Coordination in the Newborn. Developmental Psychology. 1982;18(3):450–461. [Google Scholar]
  50. von Hofsten C, Ronnqvist L. Preparation for Grasping an Object - a Developmental-Study. Journal of Experimental Psychology-Human Perception and Performance. 1988;14(4):610–621. doi: 10.1037//0096-1523.14.4.610. [DOI] [PubMed] [Google Scholar]

RESOURCES